Cargo Drones Deliver in the Amazon Rainforest

Cross-posted from WeRobotics

The Amazon is home to thousands of local indigenous communities spread across very remote areas. As a result, these sparsely populated communities rarely have reliable access to essential medicines and public health services. Local doctors in the region of Contamana report an average of 45 snakebites per month and no rapid access to antivenom, for example. We recently traveled to the rainforest to learn more about these challenges, and to explore whether cargo drones (UAVs) could realistically be used to overcome some of these problems in a sustainable manner. We’re excited to share the results of our latest field tests in this new report (PDF); Spanish version here. For high-resolution photos of the field tests, please follow this link. Videos below.

F8. Frankie 8

Our cargo drone flights were carried out in collaboration with the Peruvian Ministry of Health and local doctors. The field-tests themselves were coordinated by our local WeRobotics lab: Peru Flying Labs. Anti-venom was flown from the town of Contamana to the more remote village of Pampa Hermosa about 40 kilometers away. A regular boat (canoe) takes up to 6 hours to complete the journey. Our drone took around 35 minutes.

At night, we flew the drone back to Contamana with blood samples. While cargo drone projects typically use very expensive technology, WeRobotics prefers to use affordable and locally repairable solutions instead. Behind the scenes footage of the actual cargo drone flown in the Amazon is available in the video below.

Thanks to the success of our first drone deliveries, we’ve been invited back by the Ministry of Health and local doctors to carry out additional field tests. This explains why our Peru Flying Labs team is back in the Amazon this very week to carry out additional drone deliveries. We’re also gearing up to carry out deliveries across a distance of more than 100km using affordable drones. In parallel, we’re also working on this innovative Zika-control project with our Peru Flying Labs; drawing on lessons learned from our work in the Amazon Rainforest.

We’ll be giving a free Webinar presentation on all our efforts in Peru on Wednesday, February 22nd at 11am New York time / 4pm London. Please join our email-list for more information.

PeruDec16-0525-IMG_8379 (1)

To support our local Flying Labs teams in Peru, Nepal and/or Tanzania with donations, kindly contact Peter Mosur (peter@werobotics.org). For media inquiries on the Amazon Rainforest project and WeRobotics, please contact Dr. Patrick Meier (patrick@werobotics). Ministry of Health officials and other local partners are also available for interviews.


About WeRobotics

The mission of WeRobotics is to scale the positive impact of social good projects through the use of appropriate robotics solutions. We do this by creating robotics labs (Flying Labs) that transfer professional skills and robotics solutions locally. We have Flying Labs in Asia (Nepal), Africa (Tanzania), and South America (Peru). WeRobotics is funded by the Rockefeller Foundation, which enabled the recent project in the Amazon rain-forest with our Peru Flying Labs.

First Ever Cargo Drone Deliveries in Amazon Rainforest

Cross-posted from WeRobotics

The Amazon is home to thousands of local indigenous communities spread across very remote areas. As a result, these sparsely populated communities rarely have reliable access to essential medicines and public health services. Local doctors in the region report an average of 45 snakebites per month and no rapid access to anti-venom meds, for example. We recently traveled to the rainforest to learn more about these challenges and to explore whether cargo drones (UAVs) could realistically be used to overcome some of these challenges in a sustainable manner. We’re excited to share that our cargo drone flights in the Amazon rainforest were a big success!

pic-1

This unique and successful pilot project was a big team effort including our Peru Flying Labs Coordinator Juan Bergelund, UAV del Peru and the Peruvian Ministry of Health along with some of Peru’s leading public health experts. We carried out both day and night autonomous flights between local health hub Contamana and the remote village of Pampa Hermosa around 40 kilometers away. The drones delivered life-saving anti-venom medicines as well as blood samples. The flights took around 35 minutes compared to traditional riverboat transportation, which can take up to 6 hours.

pic-2
We have already been asked by multiple local authorities in the region to carry out additional flights in coming months. These flights will test the aerial delivery of medical supplies across 100+ kilometers. A detailed review of our recent flight tests will be released in early January along with high definition pictures and videos. Our Peru Flying Labs will also be working on this Zika reduction project in Peru using cargo drones. For media enquiries, please contact Dr. Patrick Meier (patrick@werobotics) and Juan Bergelund (juan@werobotics). Ministry of Health officials and other partners are also available for interviews.

pic-3

In the meantime, we wish to sincerely thank all our outstanding partners and colleagues in Peru for their invaluable support and partnership over the past two weeks. We are very excited to continue our good work together in coming months and years.


About WeRobotics

The mission of WeRobotics is to scale the positive impact of social good projects through the use of appropriate robotics solutions. We do this by creating robotics labs (Flying Labs) that transfer professional skills and robotics solutions locally. We have Flying Labs in Asia (Nepal), Africa (Tanzania) and South America (Peru). WeRobotics is funded by the Rockefeller Foundation, which enabled the recent project in the Amazon rainforest with our Peru Flying Labs.

The Most Comprehensive Study on Drones in Humanitarian Action

In August 2015, the Swiss humanitarian organization FSD kindly hired me as a consultant to work on the EU-funded Drones in Humanitarian Action program. I had the pleasure of working closely with FSD and team during the past 16 months. Today represents the exciting culmination of a lot of hard work by many dedicated individuals.

Today we’re launching our comprehensive report on “Drones in Humanitarian Action: A Guide to the Use of Airborne Systems in Humanitarian Crises.” The full report is available here (PDF). Our study draws on extensive research and many, many consultations carried out over a year and a half. The report covers the principle actors & technologies along with key applications and case studies on both mapping and cargo drones. Note that the section on cargo delivery is drawn from a larger 20+ page study I carried out. Please contact me directly if you’d like a copy of this more detailed study. In the meantime, I want to sincerely thank my fellow co-authors Denise Soesilo, Audrey Lessard-Fontaine, Jessica Du Plessis & Christina Stuhlberger for this productive and meaningful collaboration.

screenshot-2016-12-02-09-03-23

The report and case studies are also available on the FSD Website.

Aerial Robotics and Agriculture: Opportunities for the Majority World

The majority of studies and articles on the use of drones/UAVs for agriculture seem to focus on examples and opportunities in the US, Europe or Japan. These reports talk about the needs for large scale aerial surveys over massive farms, machine learning algorithms for automated crop detection, and the development of sophisticated forecasting models to inform decisions at the very micro level. This is the realm of precision agriculture. But what about small-holder and family farms in the Majority World? Do flying robots make sense for them? Yes, in some cases, but not in the same way that this technology makes sense for large farms in highly industrialized countries.

First things first, smallholder and family farms won’t have as much need for long-range fixed-wing UAVs as ranchers do in the US. According to this FAO study (PDF), smallholder farms typically cover less than 0.02 square kilometers. Secondly, these farms do not necessarily need access to very high-resolution, orthorectified mosaics or fancy 3D models. Mosaics and 3D models require data processing software; and software requires a computer to run said software, not to mention having time to learn how to use said software. Without software, a farmer could still upload her aerial images to the cloud for processing but that requires a reliable and relatively fast Internet connection. Also, data processing means having to store that data before and after processing. So now the farmer needs software, a computer and a hard disk (or two for backup). 

I’m not suggesting that very high-resolution orthorectified mosaics, 3D models and multi-spectral sensors cannot add value to smallholder and family farms. Of course they can. Farmers have always needed accurate as well as up-to-date information on their crops and on the environmental conditions of the land on which their crops grow. I’m just suggesting a more practical approach to begin with. Simply getting a live video feed from a bird’s eye view can already reveal patterns that show everything from irrigation problems to soil variation and even pest and fungal infestations that aren’t apparent at eye level.

At the end of the day, farming is an input-output problem. Local farmers can get a live video feed from the sky to subsequently reduce their inputs—water and pesticides—while maintaining the same output. But lets unpack that a bit. Once a farmer detects an irrigation problem from the video feed, they don’t need an other piece of robotics tech to intervene. They can easily see where the drone is hovering and simply walk over to the area with basic tools to intervene as needed. Smallholder and family farms do not have access to variable-rate sprayers and other fancy tractor tools that can take precision data and respond with precision interventions. So very high-res mosaics and 3D models may add little value in the context of smallholder farms in developing countries.

rededge-cam-nepal-rice-fields

Of course, some farmers may prefer to pay consultants or local companies to carry out these aerial surveys instead of leasing or owning a drone and carrying out the surveys themselves. In fact, some companies actually found it too “tedious to teach farmers how to use the drones they had made. Instead, they decided to focus on providing the much-needed service of mapping out farms and sites” (1). In stark contrast, my team & I at WeRobotics have really enjoyed training local partners in Nepal and Tanzania. We don’t find it tedious at all but rather highly rewarding. Building local capacity around the use of appropriate robotics solutions goes to the heart of our mission.

This explains why we’re creating local robotics labs—we call them Flying Labs—to transfer the skills and technologies that our partners need to scale the positive impact of their local efforts. So we’re especially keen to work with smallholder and family farms so they can use robotics solutions to improve their yields. They could lease small drones from the labs for a nominal fee, and I’m willing to bet that some savvy young men and women working on these farms will be keen to learn a new set of skills that could lead to an increase in income. We’re also keen to work with local drone consultants or local companies to enable them to expand their services to include agriculture. The key, either way, is to design and deliver effective trainings to local farmers, consultants and/or companies while providing each with long-term support through the Flying Labs. 


Thanks to colleagues from WeRobotics for feedback on an earlier version of this postI’m keen to receive additional input from iRevolution readers. 

What Happens When the Media Sends Drone Teams to Disasters?

Media companies like AFP, CNN and others are increasingly capturing dramatic aerial footage following major disasters around the world. These companies can be part of the solution when it comes to adding value to humanitarian efforts on the ground. But they can also be a part of the problem.

screenshot-2016-10-26-04-51-44

Media teams are increasingly showing up to disasters with small drones (UAVs) to document the damage. They’re at times the first with drones on the scene and thus able to quickly capture dramatic aerial footage of the devastation below. These media assets lead to more views and thus traffic on news websites, which increases the probability that more readers click on ads. Cue Noam Chomsky’s Manufacturing Consent: The Political Economy of the Mass Media, my favorite book whilst in high school.

Aerial footage can also increase situational awareness for disaster responders if that footage is geo-located. Labeling individual scenes in video footage with the name of the towns or villages being flown over would go a long way. This is what I asked one journalist to do in the aftermath of the Nepal Earthquake after he sent me dozens of his aerial videos. I also struck up an informal agreement with CNN to gain access to their raw aerial footage in future disasters. On a related note, I was pleased when my CNN contact expressed an interest in following the Humanitarian UAV Code of Conduct.

In an ideal world, there would be a network of professional drone journalists with established news agencies that humanitarian organizations could quickly contact for geo-tagged video footage after major disasters to improve their situational awareness. Perhaps the Professional Society of Drone Journalists (PSDJ) could be part of the solution. In any case, the network would either have its own Code of Conduct or follow the humanitarian one. Perhaps they could post their footage and pictures directly to the Humanitarian UAV Network (UAViators) Crisis Map. Either way, the media has long played an important role in humanitarian disasters, and their increasing use of drones makes them even more valuable partners to increase situational awareness.

The above scenario describes the ideal world. But the media can (and has) been part of the problem as well. “If it bleeds, it leads,” as the saying goes. Increased competition between media companies to be the first to capture dramatic aerial video that goes viral means that they may take shortcuts. They may not want to waste time getting formal approval from a country’s civil aviation authority. In Nepal after the earthquake, one leading company’s drone team was briefly detained by authorities for not getting official permission.

screenshot-2016-10-26-05-07-05

Media companies may not care to engage with local communities. They may be on a tight deadline and thus dispense with getting community buy-in. They may not have the time to reassure traumatized communities about the robots flying overhead. Media companies may overlook or ignore potential data privacy repercussions of publishing their aerial videos online. They may also not venture out to isolated and rural areas, thus biasing the video footage towards easy-to-access locations.

So how do we in the humanitarian space make media drone teams part of the solution rather than part of the problem? How do we make them partners in these efforts? One way forward is to start a conversation with these media teams and their relevant networks. Perhaps we start with a few informal agreements and learn by doing. If anyone is interested in working with me on this and/or has any suggestions on how to make this happen, please do get in touch. Thanks!

Why Robots Are Flying Over Zanzibar and the Source of the Nile

An expedition in 1858 revealed that Lake Victoria was the source of the Nile. We found ourselves on the shores of Africa’s majestic lake this October, a month after a 5.9 magnitude earthquake struck Tanzania’s Kagera Region. Hundreds were injured and dozens killed. This was the biggest tragedy in decades for the peaceful lakeside town of Bukoba. The Ministry of Home Affairs invited WeRobotics to support the recovery and reconstruction efforts by carrying out aerial surveys of the affected areas. 

2016-10-10-08-14-57-hdr

The mission of WeRobotics is to build local capacity for the safe and effective use of appropriate robotics solutions. We do this by co-creating local robotics labs that we call Flying Labs. We use these Labs to transfer the professional skills and relevant robotics solutions to outstanding local partners. Our explicit focus on capacity building explains why we took the opportunity whilst in Kagera to train two Tanzanian colleagues. Khadija and Yussuf joined us from the State University of Zanzibar (SUZA). They were both wonderful to work with and quick learners too. We look forward to working with them and other partners to co-create our Flying Labs in Tanzania. More on this in a future post.

Aerial Surveys of Kagera Region After The Earthquake

We surveyed multiple areas in the region based on the priorities of our local partners as well as reports provided by local villagers. We used the Cumulus One UAV from our technology partner DanOffice to carry out the flights. The Cumulus has a stated 2.5 hour flight time and 50 kilometer radio range. We’re using software from our partner Pix4D to process the 3,000+ very high resolution images captured during our 2 days around Bukoba.

img_6753

Above, Khadija and Yussuf on the left with a local engineer and a local member of the community on the right, respectfully. The video below shows how the Cumulus takes off and lands. The landing is automatic and simply involves the UAV stalling and gently gliding to the ground. 

We engaged directly with local communities before our flights to explain our project and get their permissions to fly. Learn more about our Code of Conduct.

img_6807

Aerial mapping with fixed-wing UAVs can identify large-scale damage over large areas and serve as a good base map for reconstruction. A lot of the damage, however, can be limited to large cracks in walls, which cannot be seen with nadir (vertical) imagery. We thus flew over some areas using a Parrot Bebop2 to capture oblique imagery and to get closer to the damage. We then took dozens of geo-tagged images from ground-level with our phones in order to ground-truth the aerial imagery.

img_6964

We’re still processing the resulting imagery so the results below are simply the low resolution previews of one (out of three) surveys we carried out.

ortho1_bukoba

Both Khadija and Yussuf were very quick learners and a real delight to work with. Below are more pictures documenting our recent work in Kagera. You can follow all our trainings and projects live via our Twitter feed (@werobotics) and our Facebook page. Sincerest thanks to both Linx Global Intelligence and UR Group for making our work in Kagera possible. Linx provided the introduction to the Ministry of Home Affairs while the UR Group provided invaluable support on the logistics and permissions.

img_6827

Yussuf programming the flight plan of the Cumulus

img_6875

Khadija is setting up the Cumulus for a full day of flying around Bukoba area

img_6756

Khadija wants to use aerial robots to map Zanzibar, which is where she’s from

img_6787

Community engagement is absolutely imperative

img_6791

Local community members inspecting the Parrot’s Bebop2

From the shores of Lake Victoria to the coastlines of Zanzibar

Together with the outstanding drone team from the State University of Zanzibar, we mapped Jozani Forest and part of the island’s eastern coastline. This allowed us to further field-test our long-range platform and to continue our local capacity building efforts following our surveys near the Ugandan border. Here’s a picture-based summary of our joint efforts.

2016-10-14-09-09-48

Flying Labs Coordinator Yussuf sets up the Cumulus UAV for flight

2016-10-13-14-44-27-hdr

Turns out selfie sticks are popular in Zanzibar and kids love robots.

2016-10-14-10-01-25

Khairat from Team SUZA is operating the mobile air traffic control tower. Team SUZA uses senseFly eBees for other projects on the island.

2016-10-15-09-03-10

Another successful takeoff, courtesy of Flying Labs Coordinator Yussuf.

2016-10-15-11-11-20

We flew the Cumulus at a speed of 65km/h and at an altitude of 265m.

2016-10-15-13-11-13

The Cumulus flew for 2 hours, making this our longest UAV flight in Zanzibar so far.

2016-10-15-10-38-51-hdr

Khadija from Team SUZA explains to local villagers how and why she maps Zanzibar using flying robots.

2016-10-15-17-26-23

Tide starts rushing back in. It’s important to take the moon into account when mapping coastlines, as the tide can change drastically during a single flight and thus affect the stitching process.

The content above is cross-posted from WeRobotics.

Using Sound and Artificial Intelligence to Detect Human Rights Violations

Video continues to be a powerful way to capture human rights abuses around the world. Videos posted to social media can be used to hold perpetrators of gross violations accountable. But video footage poses a “Big Data” challenge to human rights organizations. Two billion smartphone users means almost as many video cameras. This leads to massive amounts of visual content of both suffering and wrong-doing during conflict zones. Reviewing these videos manually is a very labor intensive, time consuming, expensive and often traumatic task. So my colleague Jay Aronson at CMU has been exploring how artificial intelligence and in particular machine learning might solve this challenge.

blog1pic

As Jay and team rightly note in a recent publication (PDF), “the dissemination of conflict and human rights related video has vastly outpaced the ability of researchers to keep up with it – particularly when immediate political action or rapid humanitarian response is required.” The consequences of this are similar to what I’ve observed in humanitarian aid: At some point (which will vary from organization to organization), time and resource limitations will necessitate an end to the collection, archiving, and analysis of user generated content unless the process can be automated.” In sum, information overload can “prevent human rights researchers from uncovering widely dispersed events taking place over long periods of time or large geographic areas that amount to systematic human rights violations.”

jayblog1

To take on this Big Data challenge, Jay and team have developed a new machine learning-based audio processing system that “enables both synchronization of multiple audio-rich videos of the same event, and discovery of specific sounds (such as wind, screaming, gunshots, airplane noise, music, and explosions) at the frame level within a video.” The system basically “creates a unique “soundprint” for each video in a collection, synchronizes videos that are recorded at the same time and location based on the pattern of these signatures, and also enables these signatures to be used to locate specific sounds precisely within a video. The use of this tool for synchronization ultimately provides a multi-perspectival view of a specific event, enabling more efficient event reconstruction and analysis by investigators.”

Synchronizing image features is far more complex than synchronizing sound. “When an object is occluded, poorly illuminated, or not visually distinct from the background, it cannot always be detected by computer vision systems. Further, while computer vision can provide investigators with confirmation that a particular video was shot from a particular location based on the similarity of the background physical environment, it is less adept at synchronizing multiple videos over time because it cannot recognize that a video might be capturing the same event from different angles or distances. In both cases, audio sensors function better so long as the relevant videos include reasonably good audio.”

Ukrainian human rights practitioners working with families of protestors killed during the 2013-2014 Euromaidan Protests recently approached Jay and company to analyze videos from those events. They wanted to “ locate every video available in their collection of the moments before, during, and just after a specific set of killings. They wanted to extract information from these videos, including visual depictions of these killings, whether the protesters in question were an immediate and direct threat to the security forces, plus any other information that could be used to corroborate or refute other forms of evidence or testimony available for their cases.”

jayblog2

Their plan had originally been to manually synchronize more than 65 hours of video footage from 520 videos taken during the morning of February 20, 2014. But after working full-time over several months, they were only able to stitch together about 4 hours of the total video using visual and audio cues in the recording.” So Jay and team used their system to make sense of the footage. They were able to automatically synchronize over 4 hours of the footage. The figure above shows an example of video clips synchronized by the system.

Users can also “select a segment within the video containing the event they are interested in (for example, a series of explosions in a plaza), and search in other videos for a similar segment that shows similar looking buildings or persons, or that contains a similar sounding noise. A user may for example select a shooting scene with a significant series of gunshots, and may search for segments with a similar sounding series of gunshots. This method increases the chances for finding video scenes of an event displaying different angles of the scene or parallel events.”

Jay and team are quick to emphasize that their system “does not  eliminate human involvement in the process because machine learning systems provide probabilistic, not certain, results.” To be sure, “the synchronization of several videos is noisy and will likely include mistakes—this is precisely why human involvement in the process is crucial.”

I’ve been following Jay’s applied research for many years now and continue to be a fan of his approach given the overlap with my own work in the use of machine learning to make sense of the Big Data generated during major natural disasters. I wholeheartedly agree with Jay when he reflected during a recent call that the use of advanced techniques alone is not the answer. Effective cross-disciplinary collaboration between computer scientists and human rights (or humanitarian) practitioners is really hard but absolutely essential. This explains why I wrote this practical handbook on how to create effective collaboration and successful projects between computer scientists and humanitarian organizations.