Chelsea Barabas (CMS, ’15) and Jude Mwenda (MAS, ’15) spent January’s Independent Activities Period in Kenya, where they collaborated with wildlife conservancies to conduct a feasibility assessment for the use of drone technologies to support anti-poaching activities. They conducted a series of workshops with wildlife conservation workers and local communities who reside on the border of wildlife parks in order to understand the challenges they face in deploying drone technologies in a sustainable way. The goal of the project was to identify support mechanisms that need to be put into place in order for drones to be a sustainable intervention for conservation efforts in Kenya. The four sections of this piece were authored as reports from the field.
1. Drones in Ol Pejeta
I arrived in Kenya with two colleagues, Jude Mwenda and John Kidenda, to work on a project exploring the potential of applying unmanned aerial vehicle technology — drones — to promote conservation efforts in Kenya. Earlier this week we travelled to Ol Pejeta, a private wildlife conservancy located near Mt. Kenya in the northern part of the country. In addition to being home to a rich ecosystem of flora and fauna, Ol Pejeta serves as the guardian territory for over one hundred rhinos, including three of the five remaining northern white rhinos in the world. Given the immense price a northern white rhino horn fetches in the black market, this particular rhino species has been nearly driven to extinction. Ol Pejeta spends a large amount of time and effort every year guarding these animals, as well as other endangered species, from poaching activities.
Last year the conservancy conducted a drone pilot study to explore ways they could use this emerging technology to supplement their ongoing efforts to catch and deter poachers from killing animals on their land. Our team was interested in this project, because to date it is one of the most ambitious drone projects undertaken on the African continent, where access to this type of technology is still in the early stages of development. Jude, John, and I travelled to the conservancy to learn more about how the project was going, as well as investigate ways we could support Ol Pejeta’s work in the future.
Over the last couple of days we have had some very interesting and informative meetings with some of the key leaders in the conservancy’s drone project. They explained some of the challenges they faced in using drones specifically for anti-poaching efforts. For example, Ol Pejeta covers a very large swath of land, making it difficult to pinpoint exactly where a poaching situation may be taking place once a gunshot has been heard. Even if an approximate location can be identified through triangulating data from different park ranger watch towers, poachers are likely to be on the run within minutes of firing their weapons at an animal. This means that a drone needs to have enough battery life not only to arrive at the site of violence but also to track poachers once they start running from the crime scene. With a full battery, the drone that Ol Pejeta used for their initial pilot could fly for a maximum of ninety minutes. However, the park guards estimate that they would need at least four hours of flight time in order for the drone to be useful during a chase down of poachers. Moreover, even with an expensive, commercial-grade camera it was very difficult for the park staff to glean critical information about exactly how many people might be involved in a poaching hit and whether or not they were armed. Without such information, it was difficult for rangers to know how to best approach a band of fleeing poachers.
After reviewing some of these challenges, we explored with the Ol Pejeta staff other potential use-cases for the drone technology in their work. We discussed the potential for using drones to conduct more frequent animal censuses, as well as tracking the health and expansion of specific vegetation. Our team is now in the process of using this fieldwork as a launch point for more brainstorms about how our specific backgrounds in citizen science and grassroots technology development might support this work. We are also ramping up to conduct our first pilots of our own drone technology. We have DJI Phantom II drone, as well as three different types of cameras we want to experiment with to understand what type of imaging is possible with-low grade cameras, as well as how resilient the drone components are to the elements in east Africa.
2. Plant Analysis Drone Pilot
During our trip one of the key things we have learned about are the challenges of using drones to track and catch poachers. The folks at Ol Pejeta explained to us that they understood the primary benefit of launching an anti-poaching drone project to be the deterrent factor of having an “eye in the sky.” Drones were unlikely to be able to proactively spot and prevent poachers from entering the park, but they could mitigate poaching activity by making surveillance and protection measures both more apparent and ominous to outside observers. However, we wondered how long the deterrent effect would last if the rates at which poachers were actually apprehended did not actually increase with the drone system. Given challenges in the duration of drone flight time and the toll that the rugged environment took on the drone hardware, it remains unclear whether or not drones would actually increase park rangers’ effectiveness at tracking poachers.
In light of these findings, our team decided to spend the last couple of weeks exploring a different conservation use case for drones: vegetation monitoring and spotting illegal charcoal burning. In Kenya, charcoal burning is a growing ecological concern, because it often involves the indiscriminate clear-cutting of trees and shrubs on public land. Experts are concerned that the rapid rates of deforestation could exacerbate the effects of draught, increasing soil erosion and reducing the vibrancy of the surrounding flora. Currently, there is no efficient and methodical way to monitor such activity.
Over the last week or so, our team hacked together two different types of cameras to see whether or not we could observe charcoal burning from the sky. The first camera we used was an Infragram camera from Public Lab. This camera provides an open source platforms to analyze plant health with near infrared imagery. The second camera was a cheap Raspberry Pi camera module. In contrast to the cameras piloted at Ol Pejeta, which cost around $7,000, our cameras were quite cheap. We spent less than $100 for the parts we assembled. In addition to these two cameras we also had a GoPro camera (around $250) and mount that we used to compare the quality of the images we gathered. Our plan was to assemble these cameras, then to collect data out in a place where charcoal is frequently burned out in the open. To do this, we travelled to a town about 40 km outside of Nairobi, called Mattu. A good friend of Jude’s named Jason is from this area, so he accompanied us as a local guide.
Given large amounts of traffic on the roads leaving Nairobi, it took us much longer than we expected to reach our destination. As we pulled into town we had only about an hour of sunlight left in the day! As such, we decided that our safest bet would be to travel to Jason’s family land and build our own fire in a pit in the ground, simulating an actual charcoal burning experience. Jude and I rapidly gathered firewood for the task as John and Jason prepared the drone and cameras for flight.
As we rapidly went about our work, a small crowd of curious onlookers came to observe what we were doing. With only about thirty minutes of daylight left, we launched our drone in the air, flying it over the plot where the fire was burning. We took photos on all three of our cameras, capturing not only the fire but also a magnificent sunset against the hills of Mattu.
By the time the sun had set, we had captured several hundred images. Success! (We weren’t 100% confident that the Pi camera we built would actually work the first time, but it did!) The next task for us was to repurpose some code that Jude found to analyze the data. The code involves an algorithm that analyzes aerial satellite images to identify ground fires based on the level of vegetation on the ground and the smoke billowing from the fire. Over the weekend Jude adapted this code to serve our needs.
3. Drone Workshop at the iHub
Last Friday our team held a drone workshop with our partners at the iHub, a center for collaboration and tech incubation in Nairobi. The goal of the workshop was to convene entrepreneurs and hobbyists who are currently working with drones in Kenya to discuss current and future applications of drone technology in the local context. We were keen to learn from participants about the challenges and potential opportunities they identified for using drones for entrepreneurial activities, as well as addressing social and conservation issues.
We were not sure how many people would attend the workshop. In prior weeks, we conducted some research and invited about ten individuals who were working with drones or aerial technology more generally. In addition, we posted the event on the iHub website. We were aiming to have about a dozen participants attend. However, on the day of the event we were pleasantly surprised by the number of people who turned up, around forty participants in total. The participants ranged from young adults from Kibera’s Tunapanda Institute (a center that provides tech and computer training to young adults with minimal resources) who had never seen a drone before, to seasoned engineers and entrepreneurs who were building their own drones from scratch and developing business models based on their drone expertise.
We opened our workshop with a discussion of how drones are currently being used in Kenya. Right now, the most common applications are media driven: using drones to capture aerial images of events like weddings and journalistic reporting of large public events. Some people were using drones to develop promotional and marketing material, such as filming commercial real estate. We then moved to talking about aspirational uses of drones in the future. People had some great ideas, such as using drones to monitor traffic flows and delivering medicine and remote diagnostics to medical patients in rural areas. Participants also conceptualized drones as a potential new infrastructure, which could make some services such as wifi hotspots more accessible to the general population.
After discussing some of the inspirational potential use-cases of drones, we moved to discuss some of the significant challenges we as a community of drone enthusiasts faced in fostering innovation and experimentation with this technology. Kenya, like most countries, has not yet developed a robust policy framework for regulating drone activity. However, as recent as last week the Kenyan Civil Aviation Authority issued a notice requiring anyone who wants to deploy drones for personal or public use to receive permission before doing so. The proclamation was received with mixed emotions. On one hand, this provides some clarity and protection for people who feared being arrested or harassed by the police for flying their drones. However, it remains unclear how difficult it will actually be to obtain permission from the KCAA. Additionally, the group expressed a desire for clear guidelines regarding operator liability and the establishment of permissible civilian fly zones.
In addition to policy issues, the group also discussed challenges with the hardware and software platforms on which drones operate. It is very difficult for local enthusiasts to repair and replace parts on their drones. Not only is it difficult to source manufactured goods from abroad, but it’s also challenging to find the resources and materials necessary to manufacture locally. One participant described his frustration after attempting to access one of the few 3D printers located in Nairobi in order to recreate a component on his drone. Additionally, finding skilled individuals who have the ability to develop and implement local designs was also a challenge. The drone market is not big enough to provide a steady income for software engineers and designers who are in high demand in other sectors.
After brainstorming some of these challenges we broke into small groups to discuss specific short-term and long-term solutions to each of these problem areas. This was probably the most fruitful part of our workshop, as it gave us a chance to dig into tangible next-steps we could take in order to develop the experimental drone community in Kenya. There was great enthusiasm around the idea of establishing a permanent group of lobbyists and hobbyists to exchange knowledge and advocate for more robust policies from the government. We have already established a working group that will be working on an “internal code of ethics” and certification process for the drone community in Kenya. Our hope is to form a citizen community that self-regulates their drone activity, as well as provides training to novices who are new to the risks and challenges of flying drones. We are also planning to build a website which will serve as an online hub for drone operators across east Africa.
Overall, the workshop was a great success.
Although the number of people interested in drones in Kenya is quite large, they had never held a formal convening to get to know one another and share their experiences. Our event provided an opportunity for these disparate individuals to coalesce into a community. The workshop also enabled our team to identify tangible opportunities to extend our work moving forward. In the coming months, I look forward to contributing to this emerging community and developing more interventions to overcome the challenges we identified during these initial discussions.
4. Final Reflections
Over the last few weeks, we’ve had some diverse and highly educational experiences working on our UAV project in Kenya. The goal of our project was to survey the landscape for future innovation and applications using drone technology in the region. This work involved interviewing individuals and organizations currently experimenting with drones in Kenya, as well as building and piloting some of our own technology to gain some practical insights into the challenges and opportunities of using drones in this context.
From these experiences, we learned a few key things:
- There already exists a large number of UAV enthusiasts, designers, and hackers in Kenya. But before January this group had not coordinated themselves into a cohesive community. This is rapidly changing. The workshop we held at the iHub a couple of weeks ago has sparked the formation of a group calling themselves the Association of UAV Operators in Kenya. There are currently about twenty members actively engaging in conversations and debates about how to further the interests of the drone community moving forward. For example, this past weekend, a small group convened at a local cafe to start early conversations around the development of an advocacy and lobbying group to interface with the government regarding future regulations and guidelines for drone use. The formation of this group is quite timely, as the Kenyan Civil Aviation Authority just recently announced that any and all individuals using drones (for personal or commercial use) must receive permission from the government before doing so. It remains unclear how difficult it will be to obtain such permission, but this group is dedicated to building a support network to facilitate the process.
- There is a lot of potential for drones to be a highly-beneficial leapfrog technology in the Kenyan context. Our discussions with thought-leaders on the ground revealed many thoughtful and creative potential use cases for drones in the future. Many of these ideas conceptualized drones as a potential support infrastructure, on which critical services and products, such as the delivery of medicine or disaster response, could be implemented more efficiently. Developing “as the crow flies” distribution channels could be immensely beneficial, particularly in areas of the country where the road infrastructure is underdeveloped (rural areas) or where traffic is quite dense (Nairobi).
- Drones provide an interesting educational platform for the development of critical skills necessary for the future growth of the Kenyan tech scene. Not only do drones provide opportunities for individuals to develop skills in handling both software and hardware, but they also provide a unique opportunity for young entrepreneurs to familiarize themselves with the uncertainty that comes along with working with a cutting-edge technology. The legal and social structures currently in place have not had the opportunity to catch up to speed with the rapid development of UAV technology. Critical to any new technology’s success is having a cadre of people who are able to lead the development of new legal and social frameworks for dealing with the ambiguity and contingencies brought about by a new technology. Drones provide an opportunity to develop these skills, which can strengthen the future trajectory of the Kenyan tech scene overall.
- Although there is much enthusiasm around drones in Kenya, there are very few UAV-based projects that have managed to establish sustainable long-term operations. The price of drones is still quite high, while the accessibility of critical parts for maintenance is still quite low. Entrepreneurs also struggle to hire employees with the skills necessary to scale the design and manufacture of locally-sourced drones. However, there are some promising developments underway in Kenya, such as the opening of Gearbox, a shared maker space designed to “bridge the gaps preventing new inventions and ideas from becoming commercializable products.” Gearbox is scheduled to open this year, with the primary aim being to unleash industrial innovation in Kenya, building a base for widespread economic growth. Organizations like this will be critical to the future development of an ecosystem where cuttingedge technologies like drones are able to thrive.
Overall, our time in Kenya has yielded some very interesting insights for our team. In the coming months, we will continue to collaborate with our partners on the ground in order to support the future development of the drone-innovation ecosystem in the country. This includes bridging promising projects, such as the work in Ol Pejeta, with the partners and resources necessary to increase their sustainability. We will also continue to participate in the emerging UAVadvocacy community as they take critical next steps to pooling their resources and efforts to educate policy makers and the general public about UAV technology.
Originally published in slightly different form by the MIT Public Service Center.