PRE2018 4 Group3: Difference between revisions

From Control Systems Technology Group
Jump to navigation Jump to search
 
(221 intermediate revisions by 5 users not shown)
Line 1: Line 1:
<div style="font-family: 'Arial'; font-size: 16px; line-height: 1.5; max-width: 900px; word-wrap: break-word; color: #333; font-weight: 400; box-shadow: 0px 25px 35px -5px rgba(0,0,0,0.75); margin-left: auto; margin-right: auto; padding: 70px; background-color: white; padding-top: 25px;">
<div style="font-family: 'Arial'; font-size: 16px; line-height: 1.5; max-width: 1300px; word-wrap: break-word; color: #333; font-weight: 400; box-shadow: 0px 25px 35px -5px rgba(0,0,0,0.75); margin-left: auto; margin-right: auto; padding: 70px; background-color: white; padding-top: 25px;">


== Group Members ==
== Group Members ==
Line 33: Line 33:


== Objective ==
== Objective ==
In this project our goal is to create a proof on concept of a drone that can collect pollen and subsequently pollinate flowers in a way that results in commercial quality produce. Furthermore we will give recommendations about the implementation of such a system of drones, focusing on the use of charging systems and noise reduction.


== User, society and enterprise ==
== Approach==
===Users===
Due to decreasing population of honey bees in recent years, there has been impacts in beekeeping industries as well as in the rate of pollination by bees. Because significant proportion of food consumption in the world depends on pollination by insects, those who provide materials for food processors definitely needs replacements for the future. Those who are involved in food affected by pollination, such as beekeepers and large scale plant owners whose plants depend on pollination by insects, will be primary users who will definitely consider this solution to be feasible.


At this time, some beekeepers make a living by renting their bees to farms and municipalities that require more pollination. The jobs of those people will be complemented with robotic bees, so their job is saved, even if bees become extinct. Beekeepers will have to focus on repairing the robotic bee, rather than care for actual animals. If beekeepers can hande this shift in their business, they will not be affected by the robotic bee in a negative way.
In order to reach our objectives, the following approach will be taken.


===Society===
Firstly, a literature study will be done, focusing on the state of the art techniques that are currently being researched or developed in artificial pollination.
There are more honey bees in this world than any other type of bee and pollinating insects. This means that honey bees are the most important pollinators of our food crops. Approximately one third of our food relies on the pollination by bees. Without honey bees, we would have a global food crisis that would kill a lot of people. This food shortage in case of an extinction will be prevented if an artificial pollinator replaces bees in time. The protection of our food chain is essential and vital to humanity's survival.
Subsequently, research will be done into the important aspects of artificial pollination that are needed to design our drone. We will reach out to a stakeholder to discuss the requirements of an artificial pollinator, and in what way our product would be useful for the stakeholder.


===Enterprise===
When a significant part of the research is concluded, a model or proof of concept of the drone, with its collection and pollination devices will be made.
Plants will be in trouble if pollinators die out. A lot of them would go extinct. This would lead to mass disruption of insect and wildlife life cycles. It would be hard to predict exactly what would happen, but there would be many negative impacts on user and society alike. There will be huge demand for other (Artificial) Pollination solution. Robotic bees could be the solution and be very beneficial for enterprises to invest in
 
== Requirements ==
The things users will require the drones to meet are;
 
*The reusability of the drones
*They need to be mass producible
*Environment friendly materials need to be used preferably bio degradable
*The drones need to be energy efficient so they last long on one charge
*Easy to control or automated so that 1 person can control multiple drones at once
*During pollination of the flowers the flowers should not be damaged by the drone
*The drones need to be replaceble by one another like real bees are in a swarm
*There must be a way to locate the drones if they break in order to reuse them
 
== Approach==
The following appraoch will be used to meet the requirements:
 
First a literature study will be done on the techniques and requirements described earlier. Next will be a literature study on the current state of the art of artificial pollination. when the research is done a model and/or prototype will be build.


== Milestones ==
== Milestones ==
Line 76: Line 57:
! scope="row" | 2
! scope="row" | 2
|  
|  
* Choosing a subject, define who the stakeholders are and finish the planning
* Choosing a subject, define who the stakeholders are and making and finnishing a planning for the project.
|-
|-
! scope="row" | 3
! scope="row" | 3
|
|
*concretely define problem and starting in-depth research into required recourses
*concretely define problem and starting in-depth research into required recourses and decide on which parts the focus of the project will be
|-
|-
! scope="row" | 4
! scope="row" | 4
|  
|  
* Main part of research is completed
* most of the basic background
* Design
* Design concept start
|-
|-
! scope="row" | 5
! scope="row" | 5
|  
|  
* -
* have an interview with a stakeholder and see if requirements match
|-
|-
! scope="row" | 6
! scope="row" | 6
Line 98: Line 79:
! scope="row" | 7
! scope="row" | 7
|  
|  
* Finalize reasearch and design
* Finalize research and design
* Finish prototype
* Finish proof of concept
* Finish wiki/report
* Finish wiki/report
|-
|-
Line 112: Line 93:


== Deliverables ==
== Deliverables ==
* Research about pollination
* Proof of concept of drone design
* Advice about e-hives and deployment
* This wiki
== Planning ==
== Planning ==
Our up-to-date planning can be found with the following link: [https://docs.google.com/spreadsheets/d/154mZ1OlmBa66Y3iH9zsQIhkTriqf30_fSkKlRaEFLFc/edit#gid=0].
Our up-to-date planning can be found with the following link: [https://docs.google.com/spreadsheets/d/154mZ1OlmBa66Y3iH9zsQIhkTriqf30_fSkKlRaEFLFc/edit#gid=0].


== USE Stakeholders ==
===Users===
Due to decreasing population of honey bees in recent years, there has been impacts in beekeeping industries as well as in the rate of pollination by bees. Because significant proportion of food consumption in the world depends on pollination by insects, those who provide materials for food processors definitely needs replacements for the future. Therefore, farmers and other food producers will profit from artificial pollination, as it allows them to continue their businesses. Farmers do not care whether their crops are pollinated by bees or artificially, as long as it does not affect their ways of farming too much. They will see artificial pollinators as a solution when the bee extinction starts to harm the global food production.


==Assumptions==
At this time, some beekeepers make a living by renting their bees to farms and municipalities that require more pollination. The jobs of those people will be complemented with robotic bees, so their job is saved, even if bees become extinct. Beekeepers will have to focus on repairing the robotic bee, rather than care for actual animals. If beekeepers can hande this shift in their business, they will not be affected by the robotic bee in a negative way. They do need to get used to not working with bees, as that was probably not just their business, but also their hobby. In that case, bee keepers can keep both ordinary bees and robotic bees, so their happiness is not affected.  
For our project on robotic bees we decided to narrow down the research and design by focussing on the pollination of apple trees.
This was decided since there exist around 300.000 species of flowering plant in the world, making the design a robotic bee suited for all species of flowers too complicated to achieve.


The apple tree was chosen since the apple is the most eaten fruit in the Netherlands, as well as an important export product. Furthermore, when it comes to pollination, the apple tree is self-incompatible, which means that it must be cross-pollinated to bloom. This would accomodate the testing of a prototype of robotic bee, since the self-incompatibility helps to ensure that the bee is the only pollinator.
===Society===
There are more honey bees in this world than any other type of bee and pollinating insects. This means that honey bees are the most important pollinators of our food crops. Approximately one third of our food relies on the pollination by bees. Without honey bees, we would have a global food crisis that would kill a lot of people. This food shortage in case of an extinction will be prevented if an artificial pollinator replaces bees in time. The protection of our food chain is essential and vital to humanity's survival. Our global society as a whole profits from the replacement of bees.  


== State of the Art ==
===Enterprise===
===Pollination===
Plants will be in trouble if pollinators die out. A lot of them would go extinct. This would lead to mass disruption of insect and wildlife life cycles. It would be hard to predict exactly what would happen, but there would be many negative impacts on user and society alike. There will be huge demand for other (artificial) pollination solutions. Robotic bees could be the solution and be very beneficial for enterprises to invest in, as they can rent their robotic bees to farmers, who can pollinate their crops. For the rest of the plants and trees, the government can either hire robotic bees from bee keepers or buy and maintain them on their own, perhaps as part of 'Staatsbosbeheer' or 'Rijkswaterstaat'.
====Pollen collection====
Honey bees collect pollen as a source of food, used to satisfied the need for protein, minerals and fats. Aside from pollen, honey bees harvest nectar from flowers as part of their diet.
While a bee harvests nectar from a plant, the hairs on its body are used to collect pollen from inside the flower. Since the pollen stick to every part of its body, the bee grooms itself with the pollen brush on the back of its legs and collects the pollen in pockets on its hind legs. This is done while the bee is in flight.


Study shows that the density and length of the bees hairs is an important factor in the adhesion of pollen. <ref> Honey bee hairs and pollenkitt are essential for pollen capture and removal, Guillermo J Amador et al 2017 Bioinspir. Biomim. 12 026015, https://doi.org/10.1088/1748-3190/aa5c6e </ref> It is found that the pollen brushes have a hair spacing of 30 micro meter, which results in a hair density of 1100 mm^2. Furthermore it is found that these hairs have a length of 250 micro meter and stand at an angle of 35 degrees with respect to the leg-surface.
As the world will need quite some of these drones, the producers of artificial pollinators will have a good business model with the production, marketing and selling of the drone pollinators.
Subsequently the study shows that there is a strong relation between the pollen size and its adhesion to the hairs. The relation between the diameter of the pollen and the spacing of the bee's hairs defines the difficulty of removal. When the diameter of the pollen is significantly smaller than the spacing of the hairs, the pollen settles deeply into the hairs. However, as the diameter/hair-spacing ration increases, the pollen are suspended between the hairs, which better facilitates the removal of the pollen.


If we want to replicate a bee's hairs as a form of pollen collection, it is required to think of the pollination method. If the goal is to hop from flower to flower and thereby pollinating flowers with the accumulated pollen from one bee, it would be best suited to use a hair setup that easily releases the collected pollen onto a flower. This would mean a patch made up of relatively short hairs, with a high density.
=== Interview ===


However, if the priority would be to collect the pollen, a set of longer hairs would be required.
We went to the Philips Fruittuin to explain our research and ask some questions about it and the bee probem. We spoke with the owner, Carlos Faes.


==== Artificial Pollination in Kiwifruit and Olive Trees<ref>Tacconi Gianni and Michelotti Vania (June 6th 2018). Artificial Pollination in Kiwifruit and Olive Trees, Pollination in Plants, Phatlane William Mokwala, IntechOpen, DOI: 10.5772/intechopen.74831. Available from: https://www.intechopen.com/books/pollination-in-plants/artificial-pollination-in-kiwifruit-and-olive-trees</ref> ====
* What family of apples do you produce?
** All kinds of apples, different families of apples is good for the pollination
* Do you think the bees will be able to pollinate our food in the future?
** Yes, I think the bee problem will solve itself eventually. Nature has a way of fixing itself. If the bee population declines by half, people will have a food shortage and die out as well. This will give the bees room to grow again.
** People do not need to die if farmers become more artisan.
** People should have mutual respect for bees when dealing with them. Right now, their nectar is replaced with sober sugar water, which is not good enough for the bees.
*** It is like with insects. In the chain of food productions, insects were a problem for the farmers, so people poisoned them. Because of this, a lot of farmers produce way more food, which also means the price dropped a lot, which makes it very hard for farmers to make some money. Once the farmers kick the bucket, there is going to be a food shortage in the world and people will die. This will result in a growth of insects again.
*** If every farmer produces half of what they do now, everyone will be saved. But people are selfish and some farmers will still produce more (for the higher price), so not a single farmer will half their production.
* Do you have a method for pollination or do you let nature handle it?
** I hire bees when the trees are flowering. I need around 20 bee colonies, which is 2 colonies per hectare and approximately 40,000 bees per colony.
** Keeping bees myself would cost a lot of energy and time, so I do not do that. They are like pets, you need to take care of them.
* What do you think about the concept of a robotic bee / artificial pollinator
** It would be possible to pollinate artificially, it already happens a lot in china with labourers, who pollinate using brushes.
** It would be a shame if this would really be necessary. The problem should be solved by protecting the bees, but it is useful to research this anyway.
* Do you notice the reduction of the amount of bees?
** Not really yet. I think I'll be dead before the beeproblem gets too serious. I do not notice much difference because this land was too big to pollinate without bees before the decline of bees, and hired bees are still easy to come by.
* How much have you thought about pollination while planting your trees?
** I have thought about this a lot, because it is very important for the quality of your apples. There needs to be at least 30% of 'strange' pollen
* Have you ever thought about artificial pollination / do you think it is possible to create an artificial bee?
** Read a bit about it, but I have my bees, so it is not really necessary.
** Pollen inside of flowers is only ripe for a few specific hours, which is a different moment for every single flower. Bees can immediately notice whether pollen is useful, while a drone cannot do this easily. That is why bees stay on certain flowers longer than on others. Bees are very smart, if you can achieve to copy their ways and senses, you have build a great robot and you have found a golden business plan. This will be extremely hard to do tho.
* What would you expect from a robotic bee?
** I would go crazy if it makes a lot of sound, like drones do at this moment.
** Not too big
** It should not need adjustments of my orchard (like roofing in)
* What way should be spread the pollen? From flower to flower or first collect, then spray liquified pollen
** I do not know whether it is possible to suck out pollen out of apples, that is not my area of expertise.
** Liquifying is probably possible.
** If you can create specific air flows, you could pollinate through the air, but this is hard without adding a roof to the orchard.


In this study, they tested what the best way to collect,store and spread pollen for kiwifruits. Pollen samples were collected with two different systems, but was irrelevant to the conclusion. They timing of when and how to store was more important. Th best way to store to guarantee the highest qualtiy of pollen obtained when the pollen were picked up from the collecting machines about every hour. This is to avoid any stres on the pollen.
This interview gave us a new insight in the bee problem and how to solve it. However, we did not agree with his solution for the problem. We would like to use technology to fix the problem, rather than let the decline of the human population get the bee population back in balance. His requirements for the robotic bee were very useful to create a drone design.
For short term storage the pollen needed to be stored at 4°C for no more than 7 days. For long tern storage the pollen needed to be stored at −18°C for no more than 3 years low humidity or pre-dried to 10–12% with silica gel at 4°C.


For spreading the pollen they used liquid and dry pollination with varying machines in different flowering stages of the kiwifruit flower. There both as effect if done at the specific flowering. for liquid pollination it was Early Petals Fall and for dry pollination it was Petals Fall.
== Requirements ==
The things users will require the drones to meet are:


They used the same technique on olive trees to better understand the moment for pollination in relation to the flowering stage during flowering as they were as they were effective as well.
*The drones need to be energy efficient so they last long enough on one charge
*The flowers should not be damaged by the artificial pollination
*The drones must be fully charged in a short timespan
*The drones must be able to reach a charging station in time before a battery runs out
*The drones must be able to hover over flowers while collecting and pollinating
*The pollen must be efficiently collected from and spread on a flower
*The drones need to be charged efficiently; charging times should be similar or lower than flying times
*Charging should be safe for the environment and the users
*Charging should work no matter the weather conditions


==== Pollination efficiency of artificial and bee pollination practices in kiwifruit <ref>Agustín Sáez, Pedro Negri, Matias Viel, Marcelo A. Aizen, Pollination efficiency of artificial and bee pollination practices in kiwifruit Scientia Horticulturae, Volume 246, 2019, Pages 1017-1021, ISSN 0304-4238, http://doi.org/10.1016/j.scienta.2018.11.072.(http://www.sciencedirect.com/science/article/pii/S0304423818308525) </ref> ====
== Pollen collection and Pollination ==


In this study they state that the efficiency of artificial pollination has never been compared with that provided by bees and will do so themselves. When comparing bee pollination with artificial. Bee pollination did not only increase the number of kiwifruit produced, but also the number of seeds per fruit, fruit weight and even higher homogeneityin.
=== Assumptions ===
For the research into pollination and collection we decided to narrow down the research and design by focussing on the pollination of apple trees.
This was decided since there exist around 300.000 species of flowering plant in the world <ref> http://www.theplantlist.org/browse/A/ </ref>, making the design of a robotic bee suited for all species of flowers too complicated to achieve in the limited time that is available.


Something to also note:
The apple tree was chosen since the apple is the most eaten fruit in the Netherlands, as well as an important export product. Furthermore, when it comes to pollination, the apple tree is self-incompatible, which means that it must be cross-pollinated to bloom. This would accomodate the testing of a prototype of robotic bee, since the self-incompatibility helps to ensure that the bee is the only pollinator.


Almost all the fruits produced in the bee-pollinated plants were of export quality while that of artificially pollinated were not.This is because Artificially pollination happened once, when ∼10% of all flowers remained as buds.as for the open flower that were sprayed with pollen, some of them were already senescent. The senescent flowers causes higher chances of producing malformed fruits or no fruit at all.
Furthermore, since appleblossom grows all around the branches and thus also face towards the ground, the drone cannot land on all of the flowers in order to pollinate them. Therefore it was decided to focus on a drone that only hovers over the flowers while pollinating or collecting pollen.


==== Effects of natural and artificial pollination on fruit and offspring quality <ref>Alexander Chautá-Mellizo, Stuart A. Campbell, Maria Argenis Bonilla, Jennifer S. Thaler, Katja Poveda, Effects of natural and artificial pollination on fruit and offspring quality, Basic and Applied Ecology, Volume 13, Issue 6,
=== Pollen collection ===
2012, Page 524-532, ISSN 1439-1791, https://doi.org/10.1016/j.baae.2012.08.013. (http://www.sciencedirect.com/science/article/pii/S143917911200093X)</ref> ====


In this study they research the effects natural and artificial pollination on cape gooseberry. The test the effects of fruit and offspring characteristics on honey and bumble bee pollination compared to manual outcrossing and autonomous self-pollination. Compared to manual and self-pollination, bee pollination increased fruit size, seed set and germination rates. On the other hand , plant growth rate and herbivore resistance were significantly and marginally greater in manually outcrossed plants compared to self-pollinated offspring, suggesting that inbreeding reduces offspring quality. Herbivore resistance and plant growth did not differ between one honeybee visit and self-pollination suggesting that multiple pollinator visits are needed to prevent inbreeding events. bees visitation can significantly alter ecologically and economically relevant traits in this agroecosystem.
For the collection of the pollen, two main methods can be distinguised.


==== Materially Engineered Artificial Pollinators <ref>Svetlana A. Chechetka, Yue Yu, Masayoshi Tange, Eijiro Miyako, Materially Engineered Artificial Pollinators, Chem, Volume 2, Issue 2, 2017, Pages 224-239, ISSN 2451-9294, https://doi.org/10.1016/j.chempr.2017.01.008. (http://www.sciencedirect.com/science/article/pii/S2451929417300323)</ref> ====
Firstly, the pollen could be collected by means of touching the flower, causing the pollen to stick to the surface they make contact with. Bees use this method to collect pollen for themselves. The pollen are collected in the bee's fur and then transfered to pollen baskets on their legs for transport.
Study shows that the density and length of the bees hairs is an important factor in the adhesion of pollen <ref> Honey bee hairs and pollenkitt are essential for pollen capture and removal, Guillermo J Amador et al 2017 Bioinspir. Biomim. 12 026015, https://doi.org/10.1088/1748-3190/aa5c6e </ref>; the relation between the diameter of the pollen and the spacing of the bee's hairs defines the difficulty of removal. When the diameter of the pollen is significantly smaller than the spacing of the hairs, the pollen settles deeply into the hairs. However, as the pollen diameter to hair-spacing ratio increases, the pollen are suspended between the hairs, which better facilitates the removal of the pollen.


In this study, multifunctionality  from  synthesized  ionic  liquidgels (ILGs) for biotechnology is presented. ILGs exhibit unique properties and coating vertically aligned animal hair with ILGs can be used for effective pollen collection. When place onto a radiowave-controllable UAV it could successfully pollinate L.  japonicumflowers.
This conclusion could be used to design a collection patch, by imitating the fur of a bee for optimal adhesion. According to the aformentioned research, the ratio of hair-spacing to pollen diameter should be 1 for optimal pollen adhesion. However, since apple pollen have an elliptic form, of on average 45 by 25 micrometer <ref> Pollen Morphology Of The Genus Malus (Rosaceae), V. Nazeri Joneghani, 2008 Ir. Journ. of Science&Technology, vol 32, n A2 </ref>, the pollen will never exactly fit into an evenly spaced grid of hairs. Thus a hair spacing of the average, 35 micrometer, would be the best suited option.


==== Development of strawberry pollination system using ultrasonic radiation pressure <ref>Hiroshi Shimizu, Taito Sato, Development of strawberry pollination system using ultrasonic radiation pressure, IFAC-PapersOnLine, Volume 51, Issue 17, 2018, Pages 57-60, ISSN 2405-8963, https://doi.org/10.1016/j.ifacol.2018.08.060 (http://www.sciencedirect.com/science/article/pii/S2405896318311765)</ref> ====
However, a disatvantage of pollination with a patch is the precision required from the drone. Since the blossoms of an apple tree have a diameter of about 3 to 4 cm, the robot will have to aim precisely to be able to touch the patch to the stamens of the flower. Adding this to the fact that the robot will need to visit a large amount of flowers, the usage of a patch would be a relatively slow collection method.


In this study they developed an artificial pollination system using ultrasonic radiation pressure as a substitute technique for bee pollination for strawberry cultivation in a plant factory. It has a higher marketable rate than that of no pollination treatment or brush pollination.
Furthermore, since we aim to pollinate the flowers on scheduled times to optimize fruit yield, it is not desireable for the flowers to be pollinated unintentionally. When collecting pollen, the robotic bees will have to visit multiple flowers for collection before returning to the hive to charge and dispense the pollen, to be able to make the system efficient. As a consequence, a patch will already contain pollen that have been collected when the next flower is visited, risking accidental pollination. Thus, the patch will need to be designed in such a way that as many pollen as possible will remain stuck to the patch. However, it should still be possible to easily remove the collected pollen at the hive, when the bee returns to empty its storage. This could be achieved by using electrostatic energy.
[[File:bee+charge.png|400px|thumb|charge on flower and bee]]


=== (Autonomous) Drones ===
Drones flying through the air will generate a small static charge on their surface just like a helicopter does. With plants having a slight negtive charge on them ,<ref>Warnke U (1977) Information transmission by means of electrical biofields. Proceedings of The Symposium on Electromagnetic Bio-Information of Marburg, pp. 55-79.</ref> as well as on their pollen grains, this means that the pollen will stick on the surface of the drone more easily. The same principle is used by bees already. Bees generate a small static charge on their fur while flying, which makes the pollen stick much better to their fur and so ensuring that the pollen don't fall off when flying. 
==== Autonomous drone is making test flights in Kansas, Illinois <ref> P. J. Griekspoor, "Autonomous drone is making test flights in Kansas, Illinois," Southwest Farm Press, 2018. Available: https://search.proquest.com/docview/2088344724?accountid=27128</ref> ====


In this project, a drone was created which can fly without an operator or pilot on the scene. It has been created for the purpose of surveillance. This project shows how an autonomous drone which keeps track of a map spawns more than 30 GB of data to fly in an area of around 400 hectares.
We can use this static charge the drone generates to make the pollen stick better to the patch. When the pollen are collected from the flower and need to be stored in the hive, we can flip the polarity of the static charge so that all the pollen will fall off easily, since the pollen and the patch repel eachother.
To make sure no pollen will fall of the patch, a charge needs to be put on the surface of the patch, preferably a bigger one then the one generated by friction between the rotors and small particles.


==== Watching the watchmen: Drone privacy and the need for oversight <ref>Jenkins, Ben. "Watching the watchmen: Drone privacy and the need for oversight." Ky. LJ 102 (2013): 161.</ref> ====
A second possibility could be collection of pollen by means of a vacuum cleaner type of attachment.
The use of suction as a collection method would eliminate the problem of accidental pollination, as opposed to pollination by touching, since the collected pollen could be collected in a closed-off reservoir. Furthermore, the drone would need to be less precise when approaching the flower, since suction in the general area of the stamen would be sufficient to collect the pollen.  


However, the method of suction itself is less precise than the touch-method. When using a vacuum cleaner there is a significant risk of sucking in unwanted material such as small insects or blockages of the system . In addition, the suction power should be well-balanced, enough to enable the robot to collect the pollen, but not damage the stamens or gynoecium of the plant.
A last and important disadvantage of the suction method is the power needed to use the attachment. Since an essential part of the design of the robotic bee is its power usage and action-radius, additional electical components would influence the efficiency of the robot.


This paper explores the privacy concerns that is associated with drones and other UAVs. It shows how a 'privacy by design (PbD)' approach helps to ensure that the aqcuired data is protected and the privacy is protected from an early stage of development.
=== Storage and treatment ===


==== Privacy, data protection and ethics for civil drone practice: A survey of industry, regulators and civil society organisations <ref> Rachel L. Finn, David Wright, "Privacy, data protection and ethics for civil drone practice: A survey of industry, regulators and civil society organisations",
==== Suspension consistency ====
Computer Law & Security Review, vol. 32, num. 4, p.p. 577 - 586</ref> ====
When pollinating with a solution of pollen, there are several factors to take into account.
Firstly, pollen are very delicate when it comes to storage. Shaking the suspension or storing it a too high temperatures causes the pollen to burst and thus become unsuited for pollination. Furthermore, the ability for the pollen to germinate, which is necessary for succesfull pollination, while being stored or suspended, depends on the treatment of the pollen. When the pollen are hydrated before being suspended, the germination rates of the pollen increase significantly. Subsequently, the composition of the suspension liquid plays an important role in the germination rate of the pollen.
According to research focussed on the conservation of pollen viability in several suspensions <ref>  M. E. Hopping & E. M. Jerram (1980) I. Development of suspension media, New Zealand Journal of Agricultural Research, 23:4, 509-515, DOI: 10.1080/00288233.1980.10417875 </ref>, a solution of calcium nitrate, boric acid and CMC (sodium carboxymethyl cellulose), each at 0.01%. However, a problem that remains is protecting the pollen grains while they dry on the flower, since the unprotected drying of the suspension causes the pollen grains to lose their capability to pollinate.


This article presents the findings from a survey of the drone industry, regulators and civil society organisations. It uses these results to show that the drone industry is diverse in applications and payloads. The industry sometimes has a lack of knowledge about privacy, ethics and data protection. Operators are often not aware of their obligations within the European law about these subjects. Bringing together watchdogs and regulatory organisations could help to educate drone operators and members of the public.
=== Pollination techniques ===


==== Experimentally Validated Extended Kalman Filter for UAV State Estimation Using Low-Cost Sensors <ref>Driessen, S. P. H., et al. "Experimentally Validated Extended Kalman Filter for UAV State Estimation using Low-Cost Sensors." IFAC-PapersOnLine, vol. 51, no. 15, 2018, pp. 43-48. SCOPUS, www.scopus.com, doi:10.1016/j.ifacol.2018.09.088.</ref> ====
There are 3 ways for our use case to apply pollen to the flowers. Dry pollination, liquid pollination and just by rubbing it on. We would discourage pollination through rubbing immediately as it is a very inefficient way of applying. From all the options, it would cost the most effort to apply and the worst results.<ref>Meng-Ying Tsai, Su-Hwa Chen, Wen-Yuan Kao,Floral morphs and seed production from hand-pollination in a population of Oxalis corymbosa in Taiwan, Flora, Volume 226, 2017, Pages 89-95, ISSN 0367-2530, https://doi.org/10.1016/j.flora.2016.11.011. (http://www.sciencedirect.com/science/article/pii/S0367253016301852)</ref> <ref>Hiroshi Shimizu, Taito Sato, Development of strawberry pollination system using ultrasonic radiation pressure, IFAC-PapersOnLine, Volume 51, Issue 17, 2018, Pages 57-60, ISSN 2405-8963, https://doi.org/10.1016/j.ifacol.2018.08.060 (http://www.sciencedirect.com/science/article/pii/S2405896318311765)</ref>


Visually based velocity and position estimations can make sure an UAV does not depend on GPS systems. This paper explores a sensor-fusion algorithm, which uses a few different sensors to achieve this. In the experiments, varying parameters were removed in case of different environmental situations. The results show that the velocity and attitude can be estimated, dispite various (indoor) environments.
According to the article Artificial Pollination in Kiwifruit and Olive Trees<ref>Tacconi Gianni and Michelotti Vania (June 6th 2018). Artificial Pollination in Kiwifruit and Olive Trees, Pollination in Plants, Phatlane William Mokwala, IntechOpen, DOI: 10.5772/intechopen.74831. Available from: https://www.intechopen.com/books/pollination-in-plants/artificial-pollination-in-kiwifruit-and-olive-trees</ref>
, dry pollination and liquid pollination are both very viable solutions. Both options give very similar results if applied correctly. However, the tests were applied to kiwi fruits, but when they used the same tests on olives trees the researchers got similar results. This makes it safe to assume that this princple also applies to the apple trees. Further research is of course needed for concrete evidence on this.


The way they sprayed these kiwis is quite different from how we will spray apples, as they did not accurately target the flower when spraying instead of spraying the whole tree. In our case, we will spray precisely onto the flowers as spraying the whole tree wastes a lot of pollen. It is easier to accurate spray onto flowers with liquids than with a dust like substance as pollen. Hence, we will use liquids for pollinating apple tree flowers.


==== Wireless charging possibilities ====
Since liquid pollination is in our opinion better achievable with drones than dry pollination, it was decided to have the drones pollinate the flowers with a solution of pollen. When doing this, there will also be an implemented pollination schedule. this would consist of spraying every flower in intervals of 4 hours at least 3 times. This will assure higher quality apples through full pollination as discussed in [[PRE2018_4_Group3_Literature#Seeds]].


*solar energy
==== Liquid Pollination ====
to see if this option is viable first we need to know how much power a micro or nano drone uses. and how much power we can generate from the sun.
The power of sunlight per square meter is 1000 W at the surface of the earth.
to calculate what the power is that a drone uses 2 sizes of drones were looked into one of 4x4cm and one of 10x10 cm


The 10 by 10 cm drone has 1100 mAh battery with 3.7 V output and can fly for 13 min
In a recent research project on the comparison between different pollination methods <ref> Kubersky, U., Boecking, O. & Wittmann, D. 2005, "Are pollen spraying and pollen dispensers alternatives to conventional pollination by bees for apple trees?", Erwerbs-Obstbau, vol. 47, no. 5, pp. 117-123. </ref> , a suspension of pollen in water was used as one of the studied pollination methods. For this experiment the liquid pollination was done in two days. On the first day a preparation of 1.2 grams of pollen in 5L of water was used. On the second day the pollen density was doubled to 2.4 grams of pollen in 5L of water. According to the researchers, this resulted in a spread of 3 pollen grains per cm^2 after spraying. However, this pollination method resulted in only 10 apples per apple tree, which was traced back to the low density of pollen per cm^2.
To be safe we use 10 minutes for the calculation (600 seconds).  
1100mAh * 3600 gives  3960000 mA seconds = 3960 A seconds
If you divide this by the flight time of 600 seconds you get the current the motors draw from the battery which is 6.6 A to calculate the power the drone uses while flying you multiply the current with the voltage. This gives a power of 24.42 W.


For the nano drone:  
In research into artificial pollination in kiwifruit trees <ref>Tacconi Gianni and Michelotti Vania (June 6th 2018). Artificial Pollination in Kiwifruit and Olive Trees, Pollination in Plants, Phatlane William Mokwala, IntechOpen, DOI: 10.5772/intechopen.74831. Available from: https://www.intechopen.com/books/pollination-in-plants/artificial-pollination-in-kiwifruit-and-olive-trees</ref>, a pollen density of 12 grams of pollen per liter of water was used, along with a ratio of 50L of suspension per hectare of trees. Since this ratio of pollen per liter of water gave decent harvesting numbers, we can assume that this density is a good reference for liquid pollination.
Its 4 by 4 cm so much smaller however it can’t fly as long as the bigger counterpart, only 5 minutes
However, since this research was conducted on kiwifruit trees and not apple trees, we cannot know for sure this ratio will work for our research purposes. Furthermore, the suspension was sprayed all over the tree and not targeted directly at the individual flowers.
The battery specifications are 3.7 V output and 100mAh
This gives a power of 4.44 W which is a lot less than the bigger drone.


The area of the drones is 0.01 square meter and 0.0016 square meter.
However, in research into the influence of pollination of kiwi fruit size is stated that 1mg of pollen per flower is needed to ensure commercial quality produce, but the researchers only sprayed the flowers once. From this we can conclude that 1mg solution per flower would be enough to pollinate.<ref>A. Alspach, P & Pyke, Nick & G. T. Morgan, C & E. Ruth, J. (1992). Influence of application rates of bee-collected pollen on the fruit size of kiwifruit. New Zealand Journal of Crop and Horticultural Science - N Z J CROP HORTICULT SCI. 19. 19-24. 10.1080/01140671.1991.10418101.</ref>. We estimate that the drone will need to spray 1 ml of water per flower. From this follows that a concentration of 1 g/L would be sufficient.
if you multiply the power that sunlight has per square meter with the area of the drones you get 10W for the bigger drone and 1.6W for the smaller one.
because this is much smaller than the power required and in the most optimal conditions, purfect sunny weather and no clouds or anything this sollution will not work if the motors of the drones do not get more efficient.


*landing platforms
== The Drone ==
*charge while flying
[[File: drone with collection path.jpg|400px|thumb|Collection patch attached to the drone]]
this problem is currently being solved by GET and they have figured out a way to power drones using modern day techniques.  
In order to be able to design the robotic bee and its attachments, we will use an existing drone body as basis. There are several factors to take into account when selecting the correct drone.
Their solution is to have an elektric net in which the drone can hover to charge. The range in which the drone needs to be is acording to them 10 meters which means that for our project the artificial bees could continue to pollunate flowers while being charged by the network they came up with.
They use a large drone for there testing with a rapid chargable battery which can be charged fully in 6 minutes. After charging they can fly 25 minutes with it.


For us this could help a lot by exploiding this solution and creating several (porable)charging locations inside the area where the artificial bees need to pollunate.
A large drone usually has a larger battery capacity and could have a larger tank for pollen storage. A bigger drone is also less influenced by external factors, like wind. However, large drones damage the flowers rather than helping them with pollination. We made a consideration between these arguments to create good balance between the advantages and disadvantages of large and small drones. The drone we will use is the Tello drone and will be around 100mm by 100mm.  
The website of GET where they have some videos explaning and showing how their tech works (http://getcorp.com/)


=== Image recognition ===
==== Functionalities of the drone ====
====Cats or CAT scans: transfer learning from natural or medical image source datasets? <ref>Cheplygina, V. (2019). Cats or CAT scans: Transfer learning from natural or medical image source data sets?.
Since our objective is to develop a proof of concept of a drone attachment that will collect and transport pollen, we need to discuss how the attachment affects the performance of the drone that will carry the attachment. Because the drone will hover over apple tree flowers to collect and deliver pollen, the drone will not be similar in terms of its size, carrying capacity, and travel distances. Rather, the drones will transport pollen within the range of the charging station acting as hives for the bees and cover multiple apple tree flowers. The drones will not be able to land on the flower due to the size that is needed to carry large amount of pollen for multiple flowers.
===== Collection and pollination estimation =====


</ref>====
Apple trees, when cultivated commercially, produce around 600 fruits in a season <ref> https://wikifarmer.com/apple-tree-harvest-yields/ </ref>. Combining this with the estimation that  90% of the flowers grows into an apple, a tree will produce around 670 flowers that will need to be pollinated.


In this article the usefullness of transfer learning is explaned for medical image analysis. Because in medical image analysis there is not much data avaiilable for training a neural network. To do this there was a large amount of data used that had nothing to do with medical images but that could be classified in different cathegories.
Honey bees visit on average 9,2 flowers per minute according to research <ref> M.J. Couvillon, C.M. Walter, E.M. Blows, T.J. Czaczkes, K.L. Alton, F.L.W. Ratnieks, "Busy Bees: Variation in Insect Flower-Visiting Rates across Multiple Plant Species" Psyche, vol. 2015, Article ID 134630, 7 pages, 2015. https://doi.org/10.1155/2015/134630. </ref>. However, since normal bees visit certain flowers for a longer amount of time in order to drink nectar, it is our estimate that the robotic bee will need 5 seconds per flower to be able position itself, collect pollen and move on to the next flower. In addition, the robotic bee will need 10 - 20 seconds to travel between the hive and the first or last flower.


====Multispectral images of flowers reveal the adaptive significance of using long-wavelength-sensitive receptors for edge detection in bees<ref>Vasas, V., Hanley, D., Kevan, P. and Chittka, L. (2019). Multispectral images of flowers reveal the adaptive significance of using long-wavelength-sensitive receptors for edge detection in bees.</ref>====
The drone that is used as fundament for our robotic bee has a flight time of about 600 seconds or 10 minutes. If 100% of the power of the drone would be available for flight, it would mean that the robotic bee would be able to visit 114 flowers before needing to return to the hive to recharge. However, assuming that the robot will not be completely efficient, we assume that on average 100 flowers will be visited per trip.


In nature bees and other insects need to detect flowers because its their main source of nutrients. They do this by detecting the edges of flowers by using a single type of receptors. the ones for long wavelengths. These receptors gave the highest signal to noise ratio, therefore it would be a good suggestion of what to look for in a camera or method to get an image.
With the assumption that a standard orchard has a tree density of around 250 trees per hectare <ref> https://wikifarmer.com/planting-apple-trees/ </ref>, an amount of 1675 trips of 10 minutes would be needed to pollinate one hectare of apple trees. This would come down to 280 hours of work, excluding the time needed to recharge the drones.


====Deap Learning<ref>LeCun, Y., Bengio, Y. and Hinton, G. (2015). Deep learning. Nature, 521, p.436.</ref>====
This also means that 167.5 thousand flowers need to be pollinated per hectare and since the drones will polinate each flower three times to optimize the chance of a successful pollination, around 5000 trips of 10 minutes would be needed to pollinate one hectare of apple trees.  
Deep learning drastically improved visual object recognition in the current state of the art, therefore is it a really good method to use in our project to determine if something is a flower or not without having an operator determining this. multiple examples of types of neural networks are given and how well some perform on images.


====Deep Learning of Representations for Unsupervised and Transfer Learning <ref>Bengio. Y  (2012)Deep Learning of Representations for Unsupervised and
The drones have no more than 4 weeks to pollinate all flowers. The sun needs to shine in order to successfully pollinate the trees. Apple trees are fertile around the end of April. In this period, there were around 240 hours of sun the last few years<ref>http://www.zonurencalculator.nl/sun_hours_calculation</ref>. Rounded up, 50 trips per hour are required. A drone can make 3 trips in an hour, so approx. 17 drones are needed per hectare. We need spare drones as backup, which makes the amount of needed drones: 25.
Transfer Learning JMLR</ref> ====


In the article they discuss the use of unsupervised training of a neural network and how it can bennefit. Especially when training on data that is not from the same cathegorie as the test data
===== Noise reduction =====
Our stakeholder told us he would only use an artificial pollinator if it would not make too much noise. In his words, the constant buzz of a drone would drive him crazy. There are a few possibilities to make our drone quieter.


====Food image recognition using deep convolutional network with pre-training and fine-tuning <ref>K. Yanai and Y. Kawano, "Food image recognition using deep convolutional network with pre-training and fine-tuning," 2015 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), Turin, 2015, pp. 1-6.
====== Noise cancelling ======
doi: 10.1109/ICMEW.2015.7169816
To cancel parts of the noise, noise-cancelling headphones use active noise control. This technique uses a microphone to measure the noise and uses a speaker to output the same noise in opposite phase, thus effectively cancelling it together. This technique might be usefull in our drones to eliminate the noise for bystanders. It has already been applied on drones successfully<ref>Castro, Victor et al. “Active Noise Cancellation System for UAVs.” (2017).</ref>. This method adds quite some electronic components to the drone, which makes it a bit heavier and more expensive. As the added components also use power, a drone equipped with noise cancelling will not be able to fly as far as a drone without this hardware. We will therefore only try to add noise cancelling if the other options turn out not to be effective enough.  
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7169816&isnumber=7169738
</ref>====


Here a more difficult task is performed, food is being recognised and because types of food sometimes looks a lot like a different type of food this is a more difficult problem than for example determining if there is a cat in a picture. In this article they use a dataset from ImageNet this website also has a lot of flower pictures we could use.
====== Reduce the RPM of the drone ======
The noise of a drone is generated by its high rotational speed. Reducing the RPM of the propellors will lead to a more silent drone. A smaller rotational speed needs to be compensated by increasing the radius of the blades in order to keep it in the air. A compromise needs to be found between size and noise. This can be done by testing multiple designs in practise.  


=== Causes and effects of bee extinction ===
====== Alter the blade design ======
[[File:silent_propellor.jpg|400px|thumb|Silent blade]]
'Northwest UAV' has used modeling software to optimize the design of the blade in terms of noise. Multiple cycles of prototyping and testing resulted in a propellor that looks like Batman's 'batarangs', thus in the shape of a bat-wing. This shape of blades will be tested on the artificial bees.


====Climate change: impact on honey bee populations and diseases <ref>Y. Le Conte and M. Navajas, "Climate change: impact on honey bee populations and diseases". (Sept 2008). Revue scientifique et technique (International Office of Epizootics). 27(2). p 499-510.</ref>====
====== Use sound dampening materials ======
Noise reduction in flying machines is a hot topic in the research area. A new material was designed recently, which reduced noise in combination with the strategic placement of this material<ref> A. Liszewski, New Noise-Blocking Material Could Make Jets and Drones Super Quiet, 12-03-2019, https://gizmodo.com/new-noise-blocking-material-could-make-jets-and-drones-1833229326</ref>. It still has to be applied on machines like drones, but this method has a lot of potential.


In this study the effects of climate change on honey bee behaviour, habitat and disease interaction are discussed. It is found that although the species shows great environmental adaptation capabilities, the added stress of climate change will worsen the factors already endangering the honey bee in certain regions of the world.
===== Concentration and tank =====
An apple flower is around 5-10 cm in diameters, 0.5-1 cm in diameters consist of the area we need spray. This would take about around 0.5 ml if every drop of liquid the drone spray lands on this area. This would of course would not be reasonable. I would make an assumption that 50% of what we spray will land on this area. Meaning we need 2x as much water per flower. Out of the pollen collection, we would be visiting an average of 100 flowers. This would mean we have a total of 100 ml of liquid the robot needs to store.


==== Bee declines driven by combined stress from parasites, pesticides, and lack of flowers <ref> D. Goulson, E. Nicholls, C. Botías, E.L. Rotheray. "Bee declines driven by combined stress from parasites, pesticides, and lack of flowers". (March 2015) Science 347(6229). DOI: 10.1126/science.1255957 </ref> ====
==== Component design ====
===== Rotation arm =====
[[Image:rotation arm.gif|center|upright-0.5|Full range of motion of the rotation arm]] Due to Tello drone not being able to fly perpendicular to the ground and wide differences in the direction of flowers, additional range of motion to reach the flowers was needed. As a result, we have decided to put a rotation arm powered by a single servo motor at the front of the drone. This arm moves 180 degrees facing the sky and the ground for the flowers facing towards either the sky or the ground.


The article discusses the causes of honey and wild bee declines. Drivers of declines and colony losses are mentioned to be habitat loss, the increasing use of pesticides, monotonous diets. The article further discusses the potential effect of climate change. Lastly, the authors offer potential solutions for sustainable pollination in the future.
===== Nozzle concept =====
[[File:initial nozzle.jpg|400px|thumb|Nozzle concept]] One of the two modules will comprise of a small container for storing pollen mixture and a nozzle that spreads pollen by drops. The nozzle will have holes facing towards into the flowers so that the pollen so that it sprays directly into the flowers. The similar concept of nozzle can be found in medicial fields with cleaning arteries where the nozzle has holes facing different directions spraying substances that removes deposits. Since the nozzle is small enough to operate and be placed inside and cleanse arteries, this can be applied to spraying pollen closer inside flowers. The spraying nozzle is small and light for easy transport and low power consumptions, yet will have enough power to spray pollen.


==== Global pollinator declines: trends, impacts and drivers <ref> S.G. Potts, J.C. Biesmeijer, C. Kremen, P. Neumann, O. Schweiger, W.E. Kunin. "Global pollinator declines: trends, impacts and drivers" (June 2010) Trends in Ecology & Evolution 25(6), p 345-353. DOI: 10.1016/j.tree.2010.01.007 </ref> ====
===== Tank concept =====
[[File: tank concept.jpg|400px|thumb|Angled mixture tank at the bottom of the drone]]In order for mixture of pollen to be collected through the bendable pipes, the tank needs to be in a shape such that the pollen can be collected into a central point. Because the tank is placed at the bottom of the drone, the mixture will be concentrated towards the bottom center of the tank. Giving this point an enough angle for mixture to slide towards the center will allow for more efficient collection of mixture via bendable pipes. Having a cube-shaped tank will prevent the pipe from efficiently collecting the mixture as it does not have a central location where the mixture is concentrated.


The authors of this article discusss the nature and extend of wild and domesticated pollinator declines and its causes. Furthermore, the impact of pollinator decline on wild flower and cultivated crop pollination is discussed, thereby analyzing the economical impact of pollinator losses.
====== One-way valve ======
[[File:springcheckball.jpg|400px|thumb|one way valve]] The way we refill this tank is through a one-way valve, allows fluid (liquid or gas) to flow through it in only one direction, to achieve this we will be using a spring energized ball check valve. The spring helps keeping the valve shut.  Without the spring we would be using the reverse flow of the water tank to keep the valve shut, this would not be ideal as the reverse flow pressure is quite weak.  By placing this on the tank we would have a way to refill it.


==== Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands <ref> Biesmeijer, J.C.,  Roberts, S.P.M.b,  Reemer, M.c, Ohlemüller, R.d, e.a. "Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands". (July 2006) Science 313(5785), p 351-354. DOI: 10.1126/science.1127863. </ref> ====
==== Component placements ====
===== Nozzle placement =====
[[File: nozzle placement.jpg|400px|thumb|Nozzle placed in an abstract drone design]] The nozzle needs to spray the liquid substance inside the flower, but the flowers may face different directions. The first idea was to have the nozzle attached without any mobility. The drone can easily maneuver horizontally to adjust the aim towards the flowers that are parallel to the ground. The drones cannot fly perpendicular to the horizon for flowers facing towards either the sky or the ground. To be able to target these flowers, a rotation mechanism needed to be implemented with the nozzle. The most straightforward methods to achieve multi-directional spraying is to have a motor that can precisely rotate every direction. However, this functionality will be redundant of what the normal quadcopter drones can perform. To prevent having redundant functionalities, we have decided to implement that the nozzle would only turn 180 degrees in vertical direction of the drone. For these functionalities, several motors are available including servo motors and precision stepper motor. The servo motor was chosen that best fits the purpose as it can precisely change the desired angles and having the precision stepper motor is not necessary as the rotation of the nozzle only needs to be within 180 degrees.


In this article the authors research the decline of local bee diversities in both the Netherlands and Britain. It was found that the largest declines could be seen in nonmigrant and flower specialist species. Furthermore, it was found that plant species relying on the declining polinators were declining in relation to other plant species as well.
===== Tank placement =====
[[File: Tank placement.jpg|400px|thumb|Tank placed underneath the abstract drone model]] For the placement of storage tank of liquid mixture, we needed to consider stability of the drone and the durability of the tank. Placing the tank in the bottom center of the drone provides stability as it distributes the weight evenly throughout the body of the drones. Having the tank underneath the drone will help self stabilization, whereas having the tank above the drone will require more energy to stabilize as the liquid mixture can shift during the transport, making the drone compensate for the instability. Having the tank underneath also prevents damages to the drones during leakage in the tanks and allows for easier tank refilling.


==== Importance of pollinators in changing landscapes for world crops <ref> A.M. Klein, B.E. Vaissière, J.H. Cane, I. Steffan-Dewenter, S.A. Cunningham, C. Kremen, T. Tscharntke. "Importance of pollinators in changing landscapes for world crops" (February 2007) Proceedings of the Royal Society B: Biological Sciences, 274(1608), p 303-313 </ref> ====
===== Internal components =====
[[File: internals.jpg|400px|thumb|Internal components of the drone attachment]]With the rotating arm powered by servo motors for nozzle attachment, there are some considerations regarding the internal pipes that pulls the mixture from the tanks. The pipes will be placed inside the drone rather than outside because the connecting cables and pipes related to spraying mixture will interfere with the movement of the drone while hovering over flowers when they are placed outside. When the nozzle begins to spray the pollen mixture, it will pull the pollen mixture from the tank, Since the pipes will be placed inside the frame of the drone and needs to go through rotation arm, the pipes need to be bendable so that it does not interfere with rotation arm's movement.


The authors of the article discuss the importance of pollinators in global agriculture by researching the volume of pollinator dependant food crops and their pollinator dependence level. Furthermore, it was found that agricultural policies and intensification jeopardize pollination stability.
== E-hives ==


====How many flowering plants are pollinated by animals? <ref> Ollerton, J.,  Winfree, R.,  Tarrant, S. "How many flowering plants are pollinated by animals?" (March 2011) Oikos 120(3), p 321-326 </ref> ====
When the robotic bees run out of energy, they return to their e-hive to charge. They also release the collected pollen here.


In this article the authors research the amount of flowering plants that depend on pollination by animals. It is concluded that in temperate climates 78% and in tropical climates 94% of flowering plant species depend on animal pollination.
The 10x10 drone uses approx. 25W as calculated in [[PRE2018_4_Group3_Literature#Wireless_charging_possibilities|Wireless charging possiblities]] and can fly around 600 seconds with a maximum speed of 7.75 m/s. As it will not fly in a straight line, but must pollinate the tree flowers in the meantime, we assume that the average speed equals half the maximum speed, 3.88 m/s. This means it can fly around 2.3 kilometers.


=== Microdrones ===
An e-hive needs to be able to power many drones simultaneously. If there is an e-hive every kilometer in the area, drones will not have to fly more than 2 times that distance before reaching the next hive, so 2 kilometers. This gives our drone some margin distance, so it can divert from its course a bit more for pollination and handle bad weather circumstances, like headwinds.
In this section, the articles discuss the microdrones and their application and implementation challenges within set area for operation. The purpose of research in this area is to find out how microdrones operate within a specified range for specialized functions, which can be replaced with artificial pollination.
==== Collaborative Microdrones: Applications and Research Challenges <ref> Quaritsch, Markus & Stojanovski, Emil & Bettstetter, Christian & Friedrich, Gerhard & Hellwagner, Hermann & Rinner, Bernhard & Hofbaur, Michael & Shah, Mubarak. (2008). Collaborative microdrones: Applications and research challenges. 38. 10.1145/1487652.1487690. </ref> ====
This article discusses the collaborative application of microdrons specialized for monitoring environmental changes, surveilence, and disaster management. These operations are based on the aerial images provided from the cameras attached to the drones. These drones are also connected to a wireless network, allowing for cooperation to gather data from multiple units of drones for future analysis.  


==== Monitoring CCS Areas using Micro Unmanned Aerial Vehicles <ref> P.P. Neumann, S. Asadi, V. Hernandez Bennetts, A.J. Lilienthal, M. Bartholmai,
The 10x10 drone has a LiPo battery with a capacity of 1100 mAh. We assume it takes 10 minutes to charge the battery of one drone.
Monitoring of CCS Areas using Micro Unmanned Aerial Vehicles (MUAVs), Energy Procedia, Volume 37, 2013, Pages 4182-4190, ISSN 1876-6102, https://doi.org/10.1016/j.egypro.2013.06.320 (http://www.sciencedirect.com/science/article/pii/S1876610213005638) </ref> ====
In this article, the author presents a method to develop Micro Unmanned Aerial Vehicles (MUAVs) with the already existing Carbon Capture and Storage technologies to detect leakage of CO{{sub|2}} gas within the storage area, which requires devices that are mobile and quickly deployable. The author conducts real life experiments within predefined areas for gathering measurements and concludes that the MUAVs are feasible for monitoring.


==== Unmanned aerial systems for photogrammetry and remote sensing: A review <ref> I. Colomina, P. Molina, Unmanned aerial systems for photogrammetry and remote sensing: A review, ISPRS Journal of Photogrammetry and Remote Sensing, Volume 92, 2014, Pages 79-97, ISSN 0924-2716, https://doi.org/10.1016/j.isprsjprs.2014.02.013. (http://www.sciencedirect.com/science/article/pii/S0924271614000501) </ref>====
=== Power ===
In this article, the author provides an overview of the development of Unmanned Aerial Vehicles (UAVs) and discusses the state-of-the-art usage of UAVs, specifically photogrammetry and remote sensing. The drones are implemented to provide precise and accurate aerial images which can be used for both photogrammetry and remote sensing within defined areas for operation. The UAVs are also required to communicate to the ground station and with each other to avoid air collisions.


==== UAV-Based Augmented Monitoring–Real-Time Georeferencing and Integration of Video Imagery with Virtual Globes <ref> Eugster, Hannes & Nebiker, Stephan. (2008). UAV-Based Augmented Monitoring–Real-Time Georeferencing and Integration of Video Imagery with Virtual Globes. </ref>====
One of the possibilities to keep the e-hive energy efficient is to power them using solar panels. Solar panels cost quite some space. The standard solar panel has an input rate of around 1000 Watt per square meter, but you will only gain roughly 15-20% efficiency at best. A solar panel of one square meter with an efficiency of 20% will therefore be approximately 200 Watt. A day has approximately 5 sun hours, so the solar panel will produce around 1 kWh per day. This energy production depends on the weather of course. Because of this, solar power alone is not very reliable. The e-hive will still need power from other sources to ensure that the drones can always be charged. 1 kWh per day corresponds to 41.67 W, which is not enough to charge multiple drones at once.
This article discusses a completion of virtual globe monitoring with the use of UAVs in areas such as forest fires, traffic, and surveillance. The project uses existing MD4-200 UAV platform with geosensors such as cameras, GPS, and compass and is augmented to virtual globe monitoring. The author concluded that the use of UAVs in the virtual globe is very promising with the data it gathers but also very challenging as real-time processing of the aerial images to models requires a new approach to overcome variables such as weather conditions and lighting.


==== Analysis of Unmanned Aerial System-Based CIR Images in Forestry—A New Perspective to Monitor Pest Infestation Levels <ref> Lehmann, Jan & Nieberding, Felix & Prinz, Torsten & Knoth, Christian. (2015). Analysis of Unmanned Aerial System-Based CIR Images in Forestry—A New Perspective to Monitor Pest Infestation Levels. Forests. 6. 594-612. 10.3390/f6030594. </ref> ====
As the e-hive are not that far apart, we will try to keep the solar panels small, which is why the solar panels will not be a lot larger than 1 square meter. The rest of the energy comes from the grid.
This article analyzes the use of UAVs to detect pest infestation at a forest. With the color infrared camera attached to the drone, the author was able to determine which trees were infested with pest. The author concluded with positive notes that these implementations of drones will save time and costs and be easy to implement for private forest owners.


===Apple===
==== Charging and filling tank ====
The collected pollen will be collected at the e-hives in liquid form. Drones that are charging can also fill up their pollen storage.


====Quality factors====
To save space at the charging station, we looked at the possibility of charging while the drones are hovering in the neigbourhood of the hive, but this costs a lot of power, as the drone needs to keep flying. This article shows that it is possible using their GET system, but not yet in a range of more than a few meters. <ref>https://www.teslarati.com/wireless-charging-drone-in-air-ces-global-energy/</ref> and on their site <ref>http://getcorp.com/</ref>. These drones are also way larger than the pollination drones. This gives them a larger charging area. In order to make it useful for our project, we would have to increase this range and make sure it can charge smaller drones. The consequence of this method is that the pollen needs to be refilled after each other instead of simultaneously.
Consumers assess an apples appearance by its colour, size, shape, absence of defects and then by its eating quality. There are 3 groups of factors affecting the quality: Genetic, Environmental and Agronomic factors.


Quality of a apple is strictly linked to its dna, there are many strains of apples that has been produced in cultivation by selective breeding(Delicious,Gala,Fuji,Golden Delicious,Jonathan,Rome)
The other possibility was to let the drones land on a platform and connect a charging cable. This increases the charge capacity a lot and makes sure the drones can fill up their pollen storage simultaneously, but this would increase the size of the hives significantly, as all the drones that charge at the same time need an own landing spot.


Apples should be grown environments that are most suited for the strain apple, this will produce higher quality apples. The most important environmental factors for high quality apples are temperature and light.
== Discussion ==
If we look at the current state of art of the drones and all the technical components needed to build the drone, the production of robotic bees would be feasible. Small drones already exist and, looking at the battery capacity and the research and development of even better batteries, the flight time will not be a problem.


Apples are highly temperature-responsive in the first 40 days after pollination. Fruit expansion was 10 times greater at a mean temperature of 20 °C than at a mean of 6 °C.<ref>Warrington, I.J., Fulton, T.A., Halligan, E.A., de Silva, H.N., 1999. Apple fruit growth and maturity are affected by early season temperatures</ref>
Currently it is already possible to quickly charge the drones or even charge the drones while they are in the air. In the coming decade these techniques will be improved, so there is no bottleneck in the charging functionalities. Furthermore, cameras have become much better over the last couple of years and therefore, using those cameras to help navigate around the flowers and to the right position will be fairly easy. Thus there is low risk of harming the flowers by flying against them.  
With all the technology available at this point in time and looking at the trend of the development it is safe to say that the feasibility of the robotic bee will not depend on technology.


Light and fruit quality are strongly associated. if high quality commercial apples get less than 50% of the total light energy they usually get, they won't develop. However, fruit exposed to high levels of light will cause some defects, such as bitter pit, internal breakdown and rotting. Areas characterized by high light intensity and temperature can also present sunburn problems. In the temperate zone, it is possible to lose from 5 to 10% of apples due to sunburn damage.<ref>Schrader, L.E., Felicetti, D.A., Sun, J., Xu, J.-Z., Zhang, J.-G., Kahn, C.B., 2008. Effects of high temperature and high solar irradiance on sunburn fruit quality, and skin pigments of apple</ref> The most common of the sunburns is browning, it results in a yellow/bronze colouration of the apple. This doesn't affect the flesh of the apple itself but when fruit surface temperature reaches about 52 degrees.<ref>Schrader, L.E., Felicetti, D.A., Sun, J., Xu, J.-Z., Zhang, J.-G., Kahn, C.B., 2008. Effects of high temperature and high solar irradiance on sunburn fruit quality, and skin pigments of apple</ref> The flesh of the apple starts to die out.  
Another thing that might cause a problem will be the pollinating techniques, and the quality of the fruit. Flowers that are not correctly pollinated will grow into not perfectly round apples or even no apple will grow at all. This is something that should be avoided, because only the apples will bring up money for the user.
The pollination technique (might be slightly changed if we have a final result) that we use can in theory pollinate every flower sufficiently, but this needs to be verified by tests that can be done after a prototype is build. The technology available at this point will definitely suffice for building and testing the pollination parts of the drone. However, even without testing, taking into account that the flowers will be sprayed upon multiple times, the chance that it will be pollinated well is very high.


Wind velocity plays a role in sunburn. demonstrated that a wind speed increase from 0.3 m s−1 to 4ms−1 reduced the fruit surface temperature by 5 °C  <ref>Thorpe, M.R., 1974. Radiant heating of apples. J. Appl. Ecol.</ref>
The only problem that could still exist is the price tag of the drone. Since a standard drone of the size chosen for this project is around €99,- and a lot of drones are required for pollination, the drones themselves will be quite expensive. Subsequently,the nozzle and the tank need to be build and placed on the drone. Lastly, charging stations and the pollen collecting patch need to be designed and made. So all in all the drone would be quite expensive. However, taking into account that all the materials are going to be mass produced, the price for 1 drone should be somewhere in between €100 and €150 .  


Reported that seeds in different apples (‘Golden Delicious’, ‘Annurca’ and ‘Red Delicious’) are able to affect also fruit weight, firmness, acidity, calcium concentration and ripening. <ref>Buccheri, M., Di Vaio, C., 2004. Relationship among seed number, quality, and calcium content in apple fruits. J. Plant Nutr. 27 (10), 1735–1746.</ref>
When renting beehives to pollinate an orchard it will cost around €60 per beehive per move. In which a move is defined as the transportation of the beehive to a new location. The Philips Fruittuin we interviewed needs 20 beehives to pollinate its whole orchard, which is around 10 hectares. So the total cost per year using this method comes to €1200 to pollinate the trees.  


=== Robotic bee replacements ===
For a field of this size we would need 250 drones to pollinate everything within the time frame in which the flowers are fertile. Current electromotors have a life span of around 2 years. If research and development is capable of improving the life span of the motors, the durability of the drones will increase. This means that repair costs will decrease over time since the drones will last longer. Therefore the drone could be a viable option if farmers and companies want to make sure food productions stay on the same level and if the bee population drops in the upcoming years.


==== What should robotic bees replace? ====
== State of the Art ==
 
===== Travel distance =====
From approximately 12,000 bees studied, the distances of honey bees traveled from their apiary for pollination ranged from 45 meters to almost 6 kilometers.<ref> Hagler, J. R., Mueller, S., Teuber, L. R., Machtley, S. A., & Van Deynze, A. (2011). Foraging range of honey bees, Apis mellifera, in alfalfa seed production fields. Journal of insect science (Online), 11, 144. doi:10.1673/031.011.14401 </ref> The rate of bees returning to thier apiary of origin decreased exponentially as the travel distance increased from 800 meters. Because robot bees are not only replacing the functions of honey bees but also to improve upon the capabilities of honey bees, the robotic bees will be program to return apiary of origin, which is the charging station.
 
===== Carrying capacity =====
An average honey bee can carry at least 15 miligrams of pollen and make at least 1 million trips per year to collect pollens in order to maintain its colony. <ref> Winston, Mark L. (1987). The biology of the honey bee. Cambridge, Mass: Harvard University Press, 177 </ref> Robotic bees with improved carrying capacity should not only carry the minimum amount to maintain but also shorten the amount of time needed to maintain.
 
====Design of an Autonomous Precision Pollination Robot <ref> N. Ohni, K. Lassak, R. Watson, J. Strader, Y. Du e.a. "Design of an Autonomous Precision Pollination Robot". (Dec 2018) 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, Madrid. DOI: 10.1109/IROS.2018.8594444 </ref> ====
 
The article describes the development of the 'BrambleBee', an autonomous robot designd to pollinate bramble plants in a greenhouse environment. It uses mapping of the environment, flower identification and high precision arm control to pollinate bramble flowers.
 
==== Autonomous pollination of individual kiwifruit flowers: Toward a robotic kiwifruit pollinator <ref> H. Williams, M. Nejati, S. Hussein, N. Penhall, J.Y. Lim, e.a. "Autonomous pollination of individual kiwifruit flowers: Toward a robotic kiwifruit pollinator" (2019) Journal of Field Robotics. DOI: 10.1002/rob.21861 </ref> ====


The paper presents an evaluation of a kiwifruit pollination robot, which is able to pollinate 79,5% of tested kiwifruit flowers. Furthermore, results show that even though several artificially pollinated flowers grew into fruit, the fruit standard was below commercial requirements.
[[PRE2018_4_Group3_Literature]]


== References ==
== References ==
<references />
<references/>

Latest revision as of 15:08, 24 June 2019

Group Members

Name Student Id
Han Wei Chia 1002684
Niek Brekelmans 1017203
Floris Verheijen 0948592
Esmee Esselaar 0987206
Minjin Song 1194206

Problem statement

In the past few years beekeepers around the world have seen sudden dissapearances of wild and domesticated bees and a steady decline in the amount of honey bee colonies. According to research, causes of the observed decline can be found in the increase in pesticide use around the world and steadily increasing urbanization. Even climate change may be a factor that influences bee population decline.

Since around a third of the global food consumption depends on pollination by insects, of which the bee is a significant contributor, the decline or even extinction of these pollinators would have a large impact on our lives.

Besides proposed solutions to stop further decline of the bee population, there is a need to compensate the already occurred loss of pollinators. In this project, the intent is to research and design a replacement for bee pollination, in the form of a robotic and drone-like bee.

Objective

In this project our goal is to create a proof on concept of a drone that can collect pollen and subsequently pollinate flowers in a way that results in commercial quality produce. Furthermore we will give recommendations about the implementation of such a system of drones, focusing on the use of charging systems and noise reduction.

Approach

In order to reach our objectives, the following approach will be taken.

Firstly, a literature study will be done, focusing on the state of the art techniques that are currently being researched or developed in artificial pollination. Subsequently, research will be done into the important aspects of artificial pollination that are needed to design our drone. We will reach out to a stakeholder to discuss the requirements of an artificial pollinator, and in what way our product would be useful for the stakeholder.

When a significant part of the research is concluded, a model or proof of concept of the drone, with its collection and pollination devices will be made.

Milestones

Week Milestones
1
  • -
2
  • Choosing a subject, define who the stakeholders are and making and finnishing a planning for the project.
3
  • concretely define problem and starting in-depth research into required recourses and decide on which parts the focus of the project will be
4
  • most of the basic background
  • Design concept start
5
  • have an interview with a stakeholder and see if requirements match
6
  • -
7
  • Finalize research and design
  • Finish proof of concept
  • Finish wiki/report
8
  • Present
9
  • -

Deliverables

  • Research about pollination
  • Proof of concept of drone design
  • Advice about e-hives and deployment
  • This wiki

Planning

Our up-to-date planning can be found with the following link: [1].

USE Stakeholders

Users

Due to decreasing population of honey bees in recent years, there has been impacts in beekeeping industries as well as in the rate of pollination by bees. Because significant proportion of food consumption in the world depends on pollination by insects, those who provide materials for food processors definitely needs replacements for the future. Therefore, farmers and other food producers will profit from artificial pollination, as it allows them to continue their businesses. Farmers do not care whether their crops are pollinated by bees or artificially, as long as it does not affect their ways of farming too much. They will see artificial pollinators as a solution when the bee extinction starts to harm the global food production.

At this time, some beekeepers make a living by renting their bees to farms and municipalities that require more pollination. The jobs of those people will be complemented with robotic bees, so their job is saved, even if bees become extinct. Beekeepers will have to focus on repairing the robotic bee, rather than care for actual animals. If beekeepers can hande this shift in their business, they will not be affected by the robotic bee in a negative way. They do need to get used to not working with bees, as that was probably not just their business, but also their hobby. In that case, bee keepers can keep both ordinary bees and robotic bees, so their happiness is not affected.

Society

There are more honey bees in this world than any other type of bee and pollinating insects. This means that honey bees are the most important pollinators of our food crops. Approximately one third of our food relies on the pollination by bees. Without honey bees, we would have a global food crisis that would kill a lot of people. This food shortage in case of an extinction will be prevented if an artificial pollinator replaces bees in time. The protection of our food chain is essential and vital to humanity's survival. Our global society as a whole profits from the replacement of bees.

Enterprise

Plants will be in trouble if pollinators die out. A lot of them would go extinct. This would lead to mass disruption of insect and wildlife life cycles. It would be hard to predict exactly what would happen, but there would be many negative impacts on user and society alike. There will be huge demand for other (artificial) pollination solutions. Robotic bees could be the solution and be very beneficial for enterprises to invest in, as they can rent their robotic bees to farmers, who can pollinate their crops. For the rest of the plants and trees, the government can either hire robotic bees from bee keepers or buy and maintain them on their own, perhaps as part of 'Staatsbosbeheer' or 'Rijkswaterstaat'.

As the world will need quite some of these drones, the producers of artificial pollinators will have a good business model with the production, marketing and selling of the drone pollinators.

Interview

We went to the Philips Fruittuin to explain our research and ask some questions about it and the bee probem. We spoke with the owner, Carlos Faes.

  • What family of apples do you produce?
    • All kinds of apples, different families of apples is good for the pollination
  • Do you think the bees will be able to pollinate our food in the future?
    • Yes, I think the bee problem will solve itself eventually. Nature has a way of fixing itself. If the bee population declines by half, people will have a food shortage and die out as well. This will give the bees room to grow again.
    • People do not need to die if farmers become more artisan.
    • People should have mutual respect for bees when dealing with them. Right now, their nectar is replaced with sober sugar water, which is not good enough for the bees.
      • It is like with insects. In the chain of food productions, insects were a problem for the farmers, so people poisoned them. Because of this, a lot of farmers produce way more food, which also means the price dropped a lot, which makes it very hard for farmers to make some money. Once the farmers kick the bucket, there is going to be a food shortage in the world and people will die. This will result in a growth of insects again.
      • If every farmer produces half of what they do now, everyone will be saved. But people are selfish and some farmers will still produce more (for the higher price), so not a single farmer will half their production.
  • Do you have a method for pollination or do you let nature handle it?
    • I hire bees when the trees are flowering. I need around 20 bee colonies, which is 2 colonies per hectare and approximately 40,000 bees per colony.
    • Keeping bees myself would cost a lot of energy and time, so I do not do that. They are like pets, you need to take care of them.
  • What do you think about the concept of a robotic bee / artificial pollinator
    • It would be possible to pollinate artificially, it already happens a lot in china with labourers, who pollinate using brushes.
    • It would be a shame if this would really be necessary. The problem should be solved by protecting the bees, but it is useful to research this anyway.
  • Do you notice the reduction of the amount of bees?
    • Not really yet. I think I'll be dead before the beeproblem gets too serious. I do not notice much difference because this land was too big to pollinate without bees before the decline of bees, and hired bees are still easy to come by.
  • How much have you thought about pollination while planting your trees?
    • I have thought about this a lot, because it is very important for the quality of your apples. There needs to be at least 30% of 'strange' pollen
  • Have you ever thought about artificial pollination / do you think it is possible to create an artificial bee?
    • Read a bit about it, but I have my bees, so it is not really necessary.
    • Pollen inside of flowers is only ripe for a few specific hours, which is a different moment for every single flower. Bees can immediately notice whether pollen is useful, while a drone cannot do this easily. That is why bees stay on certain flowers longer than on others. Bees are very smart, if you can achieve to copy their ways and senses, you have build a great robot and you have found a golden business plan. This will be extremely hard to do tho.
  • What would you expect from a robotic bee?
    • I would go crazy if it makes a lot of sound, like drones do at this moment.
    • Not too big
    • It should not need adjustments of my orchard (like roofing in)
  • What way should be spread the pollen? From flower to flower or first collect, then spray liquified pollen
    • I do not know whether it is possible to suck out pollen out of apples, that is not my area of expertise.
    • Liquifying is probably possible.
    • If you can create specific air flows, you could pollinate through the air, but this is hard without adding a roof to the orchard.

This interview gave us a new insight in the bee problem and how to solve it. However, we did not agree with his solution for the problem. We would like to use technology to fix the problem, rather than let the decline of the human population get the bee population back in balance. His requirements for the robotic bee were very useful to create a drone design.

Requirements

The things users will require the drones to meet are:

  • The drones need to be energy efficient so they last long enough on one charge
  • The flowers should not be damaged by the artificial pollination
  • The drones must be fully charged in a short timespan
  • The drones must be able to reach a charging station in time before a battery runs out
  • The drones must be able to hover over flowers while collecting and pollinating
  • The pollen must be efficiently collected from and spread on a flower
  • The drones need to be charged efficiently; charging times should be similar or lower than flying times
  • Charging should be safe for the environment and the users
  • Charging should work no matter the weather conditions

Pollen collection and Pollination

Assumptions

For the research into pollination and collection we decided to narrow down the research and design by focussing on the pollination of apple trees. This was decided since there exist around 300.000 species of flowering plant in the world [1], making the design of a robotic bee suited for all species of flowers too complicated to achieve in the limited time that is available.

The apple tree was chosen since the apple is the most eaten fruit in the Netherlands, as well as an important export product. Furthermore, when it comes to pollination, the apple tree is self-incompatible, which means that it must be cross-pollinated to bloom. This would accomodate the testing of a prototype of robotic bee, since the self-incompatibility helps to ensure that the bee is the only pollinator.

Furthermore, since appleblossom grows all around the branches and thus also face towards the ground, the drone cannot land on all of the flowers in order to pollinate them. Therefore it was decided to focus on a drone that only hovers over the flowers while pollinating or collecting pollen.

Pollen collection

For the collection of the pollen, two main methods can be distinguised.

Firstly, the pollen could be collected by means of touching the flower, causing the pollen to stick to the surface they make contact with. Bees use this method to collect pollen for themselves. The pollen are collected in the bee's fur and then transfered to pollen baskets on their legs for transport. Study shows that the density and length of the bees hairs is an important factor in the adhesion of pollen [2]; the relation between the diameter of the pollen and the spacing of the bee's hairs defines the difficulty of removal. When the diameter of the pollen is significantly smaller than the spacing of the hairs, the pollen settles deeply into the hairs. However, as the pollen diameter to hair-spacing ratio increases, the pollen are suspended between the hairs, which better facilitates the removal of the pollen.

This conclusion could be used to design a collection patch, by imitating the fur of a bee for optimal adhesion. According to the aformentioned research, the ratio of hair-spacing to pollen diameter should be 1 for optimal pollen adhesion. However, since apple pollen have an elliptic form, of on average 45 by 25 micrometer [3], the pollen will never exactly fit into an evenly spaced grid of hairs. Thus a hair spacing of the average, 35 micrometer, would be the best suited option.

However, a disatvantage of pollination with a patch is the precision required from the drone. Since the blossoms of an apple tree have a diameter of about 3 to 4 cm, the robot will have to aim precisely to be able to touch the patch to the stamens of the flower. Adding this to the fact that the robot will need to visit a large amount of flowers, the usage of a patch would be a relatively slow collection method.

Furthermore, since we aim to pollinate the flowers on scheduled times to optimize fruit yield, it is not desireable for the flowers to be pollinated unintentionally. When collecting pollen, the robotic bees will have to visit multiple flowers for collection before returning to the hive to charge and dispense the pollen, to be able to make the system efficient. As a consequence, a patch will already contain pollen that have been collected when the next flower is visited, risking accidental pollination. Thus, the patch will need to be designed in such a way that as many pollen as possible will remain stuck to the patch. However, it should still be possible to easily remove the collected pollen at the hive, when the bee returns to empty its storage. This could be achieved by using electrostatic energy.

charge on flower and bee

Drones flying through the air will generate a small static charge on their surface just like a helicopter does. With plants having a slight negtive charge on them ,[4] as well as on their pollen grains, this means that the pollen will stick on the surface of the drone more easily. The same principle is used by bees already. Bees generate a small static charge on their fur while flying, which makes the pollen stick much better to their fur and so ensuring that the pollen don't fall off when flying.

We can use this static charge the drone generates to make the pollen stick better to the patch. When the pollen are collected from the flower and need to be stored in the hive, we can flip the polarity of the static charge so that all the pollen will fall off easily, since the pollen and the patch repel eachother. To make sure no pollen will fall of the patch, a charge needs to be put on the surface of the patch, preferably a bigger one then the one generated by friction between the rotors and small particles.

A second possibility could be collection of pollen by means of a vacuum cleaner type of attachment. The use of suction as a collection method would eliminate the problem of accidental pollination, as opposed to pollination by touching, since the collected pollen could be collected in a closed-off reservoir. Furthermore, the drone would need to be less precise when approaching the flower, since suction in the general area of the stamen would be sufficient to collect the pollen.

However, the method of suction itself is less precise than the touch-method. When using a vacuum cleaner there is a significant risk of sucking in unwanted material such as small insects or blockages of the system . In addition, the suction power should be well-balanced, enough to enable the robot to collect the pollen, but not damage the stamens or gynoecium of the plant. A last and important disadvantage of the suction method is the power needed to use the attachment. Since an essential part of the design of the robotic bee is its power usage and action-radius, additional electical components would influence the efficiency of the robot.

Storage and treatment

Suspension consistency

When pollinating with a solution of pollen, there are several factors to take into account. Firstly, pollen are very delicate when it comes to storage. Shaking the suspension or storing it a too high temperatures causes the pollen to burst and thus become unsuited for pollination. Furthermore, the ability for the pollen to germinate, which is necessary for succesfull pollination, while being stored or suspended, depends on the treatment of the pollen. When the pollen are hydrated before being suspended, the germination rates of the pollen increase significantly. Subsequently, the composition of the suspension liquid plays an important role in the germination rate of the pollen. According to research focussed on the conservation of pollen viability in several suspensions [5], a solution of calcium nitrate, boric acid and CMC (sodium carboxymethyl cellulose), each at 0.01%. However, a problem that remains is protecting the pollen grains while they dry on the flower, since the unprotected drying of the suspension causes the pollen grains to lose their capability to pollinate.

Pollination techniques

There are 3 ways for our use case to apply pollen to the flowers. Dry pollination, liquid pollination and just by rubbing it on. We would discourage pollination through rubbing immediately as it is a very inefficient way of applying. From all the options, it would cost the most effort to apply and the worst results.[6] [7]

According to the article Artificial Pollination in Kiwifruit and Olive Trees[8] , dry pollination and liquid pollination are both very viable solutions. Both options give very similar results if applied correctly. However, the tests were applied to kiwi fruits, but when they used the same tests on olives trees the researchers got similar results. This makes it safe to assume that this princple also applies to the apple trees. Further research is of course needed for concrete evidence on this.

The way they sprayed these kiwis is quite different from how we will spray apples, as they did not accurately target the flower when spraying instead of spraying the whole tree. In our case, we will spray precisely onto the flowers as spraying the whole tree wastes a lot of pollen. It is easier to accurate spray onto flowers with liquids than with a dust like substance as pollen. Hence, we will use liquids for pollinating apple tree flowers.

Since liquid pollination is in our opinion better achievable with drones than dry pollination, it was decided to have the drones pollinate the flowers with a solution of pollen. When doing this, there will also be an implemented pollination schedule. this would consist of spraying every flower in intervals of 4 hours at least 3 times. This will assure higher quality apples through full pollination as discussed in PRE2018_4_Group3_Literature#Seeds.

Liquid Pollination

In a recent research project on the comparison between different pollination methods [9] , a suspension of pollen in water was used as one of the studied pollination methods. For this experiment the liquid pollination was done in two days. On the first day a preparation of 1.2 grams of pollen in 5L of water was used. On the second day the pollen density was doubled to 2.4 grams of pollen in 5L of water. According to the researchers, this resulted in a spread of 3 pollen grains per cm^2 after spraying. However, this pollination method resulted in only 10 apples per apple tree, which was traced back to the low density of pollen per cm^2.

In research into artificial pollination in kiwifruit trees [10], a pollen density of 12 grams of pollen per liter of water was used, along with a ratio of 50L of suspension per hectare of trees. Since this ratio of pollen per liter of water gave decent harvesting numbers, we can assume that this density is a good reference for liquid pollination. However, since this research was conducted on kiwifruit trees and not apple trees, we cannot know for sure this ratio will work for our research purposes. Furthermore, the suspension was sprayed all over the tree and not targeted directly at the individual flowers.

However, in research into the influence of pollination of kiwi fruit size is stated that 1mg of pollen per flower is needed to ensure commercial quality produce, but the researchers only sprayed the flowers once. From this we can conclude that 1mg solution per flower would be enough to pollinate.[11]. We estimate that the drone will need to spray 1 ml of water per flower. From this follows that a concentration of 1 g/L would be sufficient.

The Drone

Collection patch attached to the drone

In order to be able to design the robotic bee and its attachments, we will use an existing drone body as basis. There are several factors to take into account when selecting the correct drone.

A large drone usually has a larger battery capacity and could have a larger tank for pollen storage. A bigger drone is also less influenced by external factors, like wind. However, large drones damage the flowers rather than helping them with pollination. We made a consideration between these arguments to create good balance between the advantages and disadvantages of large and small drones. The drone we will use is the Tello drone and will be around 100mm by 100mm.

Functionalities of the drone

Since our objective is to develop a proof of concept of a drone attachment that will collect and transport pollen, we need to discuss how the attachment affects the performance of the drone that will carry the attachment. Because the drone will hover over apple tree flowers to collect and deliver pollen, the drone will not be similar in terms of its size, carrying capacity, and travel distances. Rather, the drones will transport pollen within the range of the charging station acting as hives for the bees and cover multiple apple tree flowers. The drones will not be able to land on the flower due to the size that is needed to carry large amount of pollen for multiple flowers.

Collection and pollination estimation

Apple trees, when cultivated commercially, produce around 600 fruits in a season [12]. Combining this with the estimation that 90% of the flowers grows into an apple, a tree will produce around 670 flowers that will need to be pollinated.

Honey bees visit on average 9,2 flowers per minute according to research [13]. However, since normal bees visit certain flowers for a longer amount of time in order to drink nectar, it is our estimate that the robotic bee will need 5 seconds per flower to be able position itself, collect pollen and move on to the next flower. In addition, the robotic bee will need 10 - 20 seconds to travel between the hive and the first or last flower.

The drone that is used as fundament for our robotic bee has a flight time of about 600 seconds or 10 minutes. If 100% of the power of the drone would be available for flight, it would mean that the robotic bee would be able to visit 114 flowers before needing to return to the hive to recharge. However, assuming that the robot will not be completely efficient, we assume that on average 100 flowers will be visited per trip.

With the assumption that a standard orchard has a tree density of around 250 trees per hectare [14], an amount of 1675 trips of 10 minutes would be needed to pollinate one hectare of apple trees. This would come down to 280 hours of work, excluding the time needed to recharge the drones.

This also means that 167.5 thousand flowers need to be pollinated per hectare and since the drones will polinate each flower three times to optimize the chance of a successful pollination, around 5000 trips of 10 minutes would be needed to pollinate one hectare of apple trees.

The drones have no more than 4 weeks to pollinate all flowers. The sun needs to shine in order to successfully pollinate the trees. Apple trees are fertile around the end of April. In this period, there were around 240 hours of sun the last few years[15]. Rounded up, 50 trips per hour are required. A drone can make 3 trips in an hour, so approx. 17 drones are needed per hectare. We need spare drones as backup, which makes the amount of needed drones: 25.

Noise reduction

Our stakeholder told us he would only use an artificial pollinator if it would not make too much noise. In his words, the constant buzz of a drone would drive him crazy. There are a few possibilities to make our drone quieter.

Noise cancelling

To cancel parts of the noise, noise-cancelling headphones use active noise control. This technique uses a microphone to measure the noise and uses a speaker to output the same noise in opposite phase, thus effectively cancelling it together. This technique might be usefull in our drones to eliminate the noise for bystanders. It has already been applied on drones successfully[16]. This method adds quite some electronic components to the drone, which makes it a bit heavier and more expensive. As the added components also use power, a drone equipped with noise cancelling will not be able to fly as far as a drone without this hardware. We will therefore only try to add noise cancelling if the other options turn out not to be effective enough.

Reduce the RPM of the drone

The noise of a drone is generated by its high rotational speed. Reducing the RPM of the propellors will lead to a more silent drone. A smaller rotational speed needs to be compensated by increasing the radius of the blades in order to keep it in the air. A compromise needs to be found between size and noise. This can be done by testing multiple designs in practise.

Alter the blade design
Silent blade

'Northwest UAV' has used modeling software to optimize the design of the blade in terms of noise. Multiple cycles of prototyping and testing resulted in a propellor that looks like Batman's 'batarangs', thus in the shape of a bat-wing. This shape of blades will be tested on the artificial bees.

Use sound dampening materials

Noise reduction in flying machines is a hot topic in the research area. A new material was designed recently, which reduced noise in combination with the strategic placement of this material[17]. It still has to be applied on machines like drones, but this method has a lot of potential.

Concentration and tank

An apple flower is around 5-10 cm in diameters, 0.5-1 cm in diameters consist of the area we need spray. This would take about around 0.5 ml if every drop of liquid the drone spray lands on this area. This would of course would not be reasonable. I would make an assumption that 50% of what we spray will land on this area. Meaning we need 2x as much water per flower. Out of the pollen collection, we would be visiting an average of 100 flowers. This would mean we have a total of 100 ml of liquid the robot needs to store.

Component design

Rotation arm
Full range of motion of the rotation arm
Due to Tello drone not being able to fly perpendicular to the ground and wide differences in the direction of flowers, additional range of motion to reach the flowers was needed. As a result, we have decided to put a rotation arm powered by a single servo motor at the front of the drone. This arm moves 180 degrees facing the sky and the ground for the flowers facing towards either the sky or the ground.
Nozzle concept
Nozzle concept
One of the two modules will comprise of a small container for storing pollen mixture and a nozzle that spreads pollen by drops. The nozzle will have holes facing towards into the flowers so that the pollen so that it sprays directly into the flowers. The similar concept of nozzle can be found in medicial fields with cleaning arteries where the nozzle has holes facing different directions spraying substances that removes deposits. Since the nozzle is small enough to operate and be placed inside and cleanse arteries, this can be applied to spraying pollen closer inside flowers. The spraying nozzle is small and light for easy transport and low power consumptions, yet will have enough power to spray pollen.
Tank concept
Angled mixture tank at the bottom of the drone
In order for mixture of pollen to be collected through the bendable pipes, the tank needs to be in a shape such that the pollen can be collected into a central point. Because the tank is placed at the bottom of the drone, the mixture will be concentrated towards the bottom center of the tank. Giving this point an enough angle for mixture to slide towards the center will allow for more efficient collection of mixture via bendable pipes. Having a cube-shaped tank will prevent the pipe from efficiently collecting the mixture as it does not have a central location where the mixture is concentrated.
One-way valve
one way valve
The way we refill this tank is through a one-way valve, allows fluid (liquid or gas) to flow through it in only one direction, to achieve this we will be using a spring energized ball check valve. The spring helps keeping the valve shut. Without the spring we would be using the reverse flow of the water tank to keep the valve shut, this would not be ideal as the reverse flow pressure is quite weak. By placing this on the tank we would have a way to refill it.

Component placements

Nozzle placement
Nozzle placed in an abstract drone design
The nozzle needs to spray the liquid substance inside the flower, but the flowers may face different directions. The first idea was to have the nozzle attached without any mobility. The drone can easily maneuver horizontally to adjust the aim towards the flowers that are parallel to the ground. The drones cannot fly perpendicular to the horizon for flowers facing towards either the sky or the ground. To be able to target these flowers, a rotation mechanism needed to be implemented with the nozzle. The most straightforward methods to achieve multi-directional spraying is to have a motor that can precisely rotate every direction. However, this functionality will be redundant of what the normal quadcopter drones can perform. To prevent having redundant functionalities, we have decided to implement that the nozzle would only turn 180 degrees in vertical direction of the drone. For these functionalities, several motors are available including servo motors and precision stepper motor. The servo motor was chosen that best fits the purpose as it can precisely change the desired angles and having the precision stepper motor is not necessary as the rotation of the nozzle only needs to be within 180 degrees.
Tank placement
Tank placed underneath the abstract drone model
For the placement of storage tank of liquid mixture, we needed to consider stability of the drone and the durability of the tank. Placing the tank in the bottom center of the drone provides stability as it distributes the weight evenly throughout the body of the drones. Having the tank underneath the drone will help self stabilization, whereas having the tank above the drone will require more energy to stabilize as the liquid mixture can shift during the transport, making the drone compensate for the instability. Having the tank underneath also prevents damages to the drones during leakage in the tanks and allows for easier tank refilling.
Internal components
Internal components of the drone attachment
With the rotating arm powered by servo motors for nozzle attachment, there are some considerations regarding the internal pipes that pulls the mixture from the tanks. The pipes will be placed inside the drone rather than outside because the connecting cables and pipes related to spraying mixture will interfere with the movement of the drone while hovering over flowers when they are placed outside. When the nozzle begins to spray the pollen mixture, it will pull the pollen mixture from the tank, Since the pipes will be placed inside the frame of the drone and needs to go through rotation arm, the pipes need to be bendable so that it does not interfere with rotation arm's movement.

E-hives

When the robotic bees run out of energy, they return to their e-hive to charge. They also release the collected pollen here.

The 10x10 drone uses approx. 25W as calculated in Wireless charging possiblities and can fly around 600 seconds with a maximum speed of 7.75 m/s. As it will not fly in a straight line, but must pollinate the tree flowers in the meantime, we assume that the average speed equals half the maximum speed, 3.88 m/s. This means it can fly around 2.3 kilometers.

An e-hive needs to be able to power many drones simultaneously. If there is an e-hive every kilometer in the area, drones will not have to fly more than 2 times that distance before reaching the next hive, so 2 kilometers. This gives our drone some margin distance, so it can divert from its course a bit more for pollination and handle bad weather circumstances, like headwinds.

The 10x10 drone has a LiPo battery with a capacity of 1100 mAh. We assume it takes 10 minutes to charge the battery of one drone.

Power

One of the possibilities to keep the e-hive energy efficient is to power them using solar panels. Solar panels cost quite some space. The standard solar panel has an input rate of around 1000 Watt per square meter, but you will only gain roughly 15-20% efficiency at best. A solar panel of one square meter with an efficiency of 20% will therefore be approximately 200 Watt. A day has approximately 5 sun hours, so the solar panel will produce around 1 kWh per day. This energy production depends on the weather of course. Because of this, solar power alone is not very reliable. The e-hive will still need power from other sources to ensure that the drones can always be charged. 1 kWh per day corresponds to 41.67 W, which is not enough to charge multiple drones at once.

As the e-hive are not that far apart, we will try to keep the solar panels small, which is why the solar panels will not be a lot larger than 1 square meter. The rest of the energy comes from the grid.

Charging and filling tank

The collected pollen will be collected at the e-hives in liquid form. Drones that are charging can also fill up their pollen storage.

To save space at the charging station, we looked at the possibility of charging while the drones are hovering in the neigbourhood of the hive, but this costs a lot of power, as the drone needs to keep flying. This article shows that it is possible using their GET system, but not yet in a range of more than a few meters. [18] and on their site [19]. These drones are also way larger than the pollination drones. This gives them a larger charging area. In order to make it useful for our project, we would have to increase this range and make sure it can charge smaller drones. The consequence of this method is that the pollen needs to be refilled after each other instead of simultaneously.

The other possibility was to let the drones land on a platform and connect a charging cable. This increases the charge capacity a lot and makes sure the drones can fill up their pollen storage simultaneously, but this would increase the size of the hives significantly, as all the drones that charge at the same time need an own landing spot.

Discussion

If we look at the current state of art of the drones and all the technical components needed to build the drone, the production of robotic bees would be feasible. Small drones already exist and, looking at the battery capacity and the research and development of even better batteries, the flight time will not be a problem.

Currently it is already possible to quickly charge the drones or even charge the drones while they are in the air. In the coming decade these techniques will be improved, so there is no bottleneck in the charging functionalities. Furthermore, cameras have become much better over the last couple of years and therefore, using those cameras to help navigate around the flowers and to the right position will be fairly easy. Thus there is low risk of harming the flowers by flying against them. With all the technology available at this point in time and looking at the trend of the development it is safe to say that the feasibility of the robotic bee will not depend on technology.

Another thing that might cause a problem will be the pollinating techniques, and the quality of the fruit. Flowers that are not correctly pollinated will grow into not perfectly round apples or even no apple will grow at all. This is something that should be avoided, because only the apples will bring up money for the user. The pollination technique (might be slightly changed if we have a final result) that we use can in theory pollinate every flower sufficiently, but this needs to be verified by tests that can be done after a prototype is build. The technology available at this point will definitely suffice for building and testing the pollination parts of the drone. However, even without testing, taking into account that the flowers will be sprayed upon multiple times, the chance that it will be pollinated well is very high.

The only problem that could still exist is the price tag of the drone. Since a standard drone of the size chosen for this project is around €99,- and a lot of drones are required for pollination, the drones themselves will be quite expensive. Subsequently,the nozzle and the tank need to be build and placed on the drone. Lastly, charging stations and the pollen collecting patch need to be designed and made. So all in all the drone would be quite expensive. However, taking into account that all the materials are going to be mass produced, the price for 1 drone should be somewhere in between €100 and €150 .

When renting beehives to pollinate an orchard it will cost around €60 per beehive per move. In which a move is defined as the transportation of the beehive to a new location. The Philips Fruittuin we interviewed needs 20 beehives to pollinate its whole orchard, which is around 10 hectares. So the total cost per year using this method comes to €1200 to pollinate the trees.

For a field of this size we would need 250 drones to pollinate everything within the time frame in which the flowers are fertile. Current electromotors have a life span of around 2 years. If research and development is capable of improving the life span of the motors, the durability of the drones will increase. This means that repair costs will decrease over time since the drones will last longer. Therefore the drone could be a viable option if farmers and companies want to make sure food productions stay on the same level and if the bee population drops in the upcoming years.

State of the Art

PRE2018_4_Group3_Literature

References

  1. ↑ http://www.theplantlist.org/browse/A/
  2. ↑ Honey bee hairs and pollenkitt are essential for pollen capture and removal, Guillermo J Amador et al 2017 Bioinspir. Biomim. 12 026015, https://doi.org/10.1088/1748-3190/aa5c6e
  3. ↑ Pollen Morphology Of The Genus Malus (Rosaceae), V. Nazeri Joneghani, 2008 Ir. Journ. of Science&Technology, vol 32, n A2
  4. ↑ Warnke U (1977) Information transmission by means of electrical biofields. Proceedings of The Symposium on Electromagnetic Bio-Information of Marburg, pp. 55-79.
  5. ↑ M. E. Hopping & E. M. Jerram (1980) I. Development of suspension media, New Zealand Journal of Agricultural Research, 23:4, 509-515, DOI: 10.1080/00288233.1980.10417875
  6. ↑ Meng-Ying Tsai, Su-Hwa Chen, Wen-Yuan Kao,Floral morphs and seed production from hand-pollination in a population of Oxalis corymbosa in Taiwan, Flora, Volume 226, 2017, Pages 89-95, ISSN 0367-2530, https://doi.org/10.1016/j.flora.2016.11.011. (http://www.sciencedirect.com/science/article/pii/S0367253016301852)
  7. ↑ Hiroshi Shimizu, Taito Sato, Development of strawberry pollination system using ultrasonic radiation pressure, IFAC-PapersOnLine, Volume 51, Issue 17, 2018, Pages 57-60, ISSN 2405-8963, https://doi.org/10.1016/j.ifacol.2018.08.060 (http://www.sciencedirect.com/science/article/pii/S2405896318311765)
  8. ↑ Tacconi Gianni and Michelotti Vania (June 6th 2018). Artificial Pollination in Kiwifruit and Olive Trees, Pollination in Plants, Phatlane William Mokwala, IntechOpen, DOI: 10.5772/intechopen.74831. Available from: https://www.intechopen.com/books/pollination-in-plants/artificial-pollination-in-kiwifruit-and-olive-trees
  9. ↑ Kubersky, U., Boecking, O. & Wittmann, D. 2005, "Are pollen spraying and pollen dispensers alternatives to conventional pollination by bees for apple trees?", Erwerbs-Obstbau, vol. 47, no. 5, pp. 117-123.
  10. ↑ Tacconi Gianni and Michelotti Vania (June 6th 2018). Artificial Pollination in Kiwifruit and Olive Trees, Pollination in Plants, Phatlane William Mokwala, IntechOpen, DOI: 10.5772/intechopen.74831. Available from: https://www.intechopen.com/books/pollination-in-plants/artificial-pollination-in-kiwifruit-and-olive-trees
  11. ↑ A. Alspach, P & Pyke, Nick & G. T. Morgan, C & E. Ruth, J. (1992). Influence of application rates of bee-collected pollen on the fruit size of kiwifruit. New Zealand Journal of Crop and Horticultural Science - N Z J CROP HORTICULT SCI. 19. 19-24. 10.1080/01140671.1991.10418101.
  12. ↑ https://wikifarmer.com/apple-tree-harvest-yields/
  13. ↑ M.J. Couvillon, C.M. Walter, E.M. Blows, T.J. Czaczkes, K.L. Alton, F.L.W. Ratnieks, "Busy Bees: Variation in Insect Flower-Visiting Rates across Multiple Plant Species" Psyche, vol. 2015, Article ID 134630, 7 pages, 2015. https://doi.org/10.1155/2015/134630.
  14. ↑ https://wikifarmer.com/planting-apple-trees/
  15. ↑ http://www.zonurencalculator.nl/sun_hours_calculation
  16. ↑ Castro, Victor et al. “Active Noise Cancellation System for UAVs.” (2017).
  17. ↑ A. Liszewski, New Noise-Blocking Material Could Make Jets and Drones Super Quiet, 12-03-2019, https://gizmodo.com/new-noise-blocking-material-could-make-jets-and-drones-1833229326
  18. ↑ https://www.teslarati.com/wireless-charging-drone-in-air-ces-global-energy/
  19. ↑ http://getcorp.com/
Retrieved from "https://dsdwiki.wtb.tue.nl/index.php?title=PRE2018_4_Group3&oldid=80424"