Community drone mapping

Places where climate change will affect most people are cities, where nearly 4.6 billion people live. These effects will be felt primarily by poor communities and informal settlements. The design of effective and equitable climate change solutions requires community data and community knowledge. Community mapping—who is where, their climate risks and vulnerabilities—is an essential component of community-driven urban adaptation to climate change. Communities with good data—and equipped with the tools and knowledge to create data on an ongoing basis—are better able to identify climate risks, solve problems, and advocate for support. This kind of data does not exist for many cities or is not accessible to the people most impacted. 

Remote sensing data such as aerial imagery is a key source for mapping and community data. And, advances in AI to process and analyze remote sensing data are accelerating mapping efforts. However, the tools and processes remain too complicated and too costly for most local communities to collect, process and use high resolution remote sensing data for mapping and the regulatory barriers are difficult to surmount.

Drone imagery as a source of imagery has massive potential for mapping in places without access to recent or affordable satellite imagery. The resolution of drone imagery is much better than open satellite imagery and this opens up huge opportunities for  use cases that require more detail. It can be collected and processed locally so the communities themselves are in control of data generation. While drones are available at reasonable prices from commercial manufacturers, there are barriers to their widespread use. Low-cost drones are often deliberately made difficult to use for mapping by manufacturers to drive purchasers to expensive models. Software to coordinate drone use is costly, mostly proprietary, and not well-adapted to local conditions. In addition,  government agencies often impose regulatory requirements to address privacy and security risks, limiting access to the technology and local economic development by making it expensive for most local people to obtain fly permissions.

For example, Dar es Salaam, Monrovia and Bamako have been mapped with high-resolution aerial imagery a few years ago, but due to their rapid pace of change, it is no longer useful. Cities such as Nairobi, Kinshasa, Lagos, and many more haven’t been imaged, although there is demand from international agencies and local land ministries.

Drone imagery with a resolution of 5 -10 cm per pixel, gathered by contractors using expensive equipment costs between $300-$900/km². Open satellite imagery can have as good as 30cm GSD and is often available for $25/km² so the drone imagery is only competitive where the higher resolution is essential. So the many benefits of having higher resolution imagery are only available when a stakeholder is willing to finance the drone imagery. By contrast, high-quality drone imagery with a 5cm GSD can be gathered by local people with inexpensive drones for $75-$150/km², making it highly competitive with satellite imagery given the much greater resolution and detail, especially in urban areas.

HOT’s  solution is to build low-cost drones that are fit for mapping urban areas, especially informal settlements, and enable drone operators to coordinate their work. HOT will build locally-led and participatory approaches for creating and accessing imagery to create high quality and high value datasets, which local actors can use to identify climate vulnerabilities and design solutions.  Some examples of effective data use by cities are:
-Disaster risk assessments (eg, flooding, landslides, earthquakes, tsunamis);
- Change detection related to urbanization and populated areas, watercourse boundaries and vegetation cover due to climate change;
- Assessment of clean energy options such as potential for solar panels;
- Monitor infrastructure maintenance, including waste management and drainage for flood resilience.

Our solution will help create an inclusive ecosystem of local people with inexpensive, locally-appropriate drones and effective software tools and processes to contribute to a global free and open aerial imagery repository that benefits everyone. It  consists of several parts:

• Develop and distribute low cost drones with open designs and software suited for mapping and optimized for data collection. 

• Support and develop open and free software for local people to operate drones safely and effectively, including automated flight control.

• Develop a platform to allow local drone operators to coordinate flights to collect consistent high quality imagery for entire cities flown using low-cost equipment. 

  • Mechanisms for people to earn a living contributing to needed imagery sets

  • Develop  an open aerial imagery repository in which anyone can share and process imagery for mapping.

Local people in for example community settings can effectively produce high-quality aerial imagery projects using locally-appropriate equipment and skills. In many cases they can earn a living by doing so, while contributing to impactful datasets that improve development, disaster risk reduction, and resilience.

HOT works through four regional Open Mapping Hubs to bring our work closer to the communities we serve: Asia Pacific, East & Southern Africa, West & Northern Africa, Latin American & the Caribbean. We work across 94 priority countries that have high levels of disaster risk and/or high levels of multidimensional poverty. Vulnerability varies across a country  so we map priority areas determined by local communities and partners.

Many African cities, particularly coastal cities in East Africa, are highly vulnerable to pluvial and fluvial flooding. Example in Tanzania, the World Bank has documented massive increases in flood risk due to urbanization (loss of absorption capacity), poor drainage and solid waste management, deforestation, erosion, sea level rise, and changing precipitation patterns. Similar patterns are seen in other coastal cities. Cities such as Freetown in Sierra Leone face disastrous landslide risks. 

Drone and aerial imagery and Digital Elevation Model products are useful to quantify and mitigate such risks, and have been used in places like Dar es Salaam to guide and inform resilience efforts. Making such data more affordable and available helps cities and the people in them adapt to climate change and become more resilient.

We have extensive experience working with local stakeholders such as government and international agencies, as well as communities, non profit, civil societies and local business. With this initiative, we aim to create economic opportunities for people in low income countries. We focus on enabling marginalized people to get decent work in their own communities.

From 2010 to 2023, HOT has grown from a small, dedicated group of advocates and innovators with a big vision, to a global movement uniting over 650,000 volunteers and dozens of mapping communities all over the world who are committed to using the power of open mapping to make a difference. We  have strong proximity to the communities we serve through our 4 Regional Mapping Hubs. Our hubs engage with local mapping communities to understand local challenges and data needs, facilitate knowledge exchanges, distribute funding, and provide training and support in order to massively scale local mapping in 94 countries. Our data sets were downloaded more than 50,000 times from the UN’s Humanitarian Data Exchange in 2023.

HOT has revolutionized disaster response and expectations for data in the humanitarian sector. A global army of remote mappers coordinated by HOT or self-directed now springs into action after any crisis event so that an up-to-date map exists to support relief efforts. Our investments in training and movement-building mean that hundreds of thousands of individuals around the world have skills and tools to map and generate value for themselves and for their communities. The result is local ownership of data. 

In addition to managing hundreds of disaster response mapping activations, we have significant field program implementation experience. We are actively partnering with local organizations around open mapping to support climate-ready cities so that open map data can be used to plan for and mitigate the impacts of extreme weather and climate events.

In Monrovia, Liberia, we are working with Slum Dwellers International and local partners to map 52 informal settlements and the homes of more than 70,000 people to ensure that city adaptation plans respond to the risks these residents face. The result will also be a wealth of open geospatial data accessible to the communities, government, and other stakeholders for the first time.

HOT is also an innovator in technology, constantly developing and refining new tools to make mapping more accessible to all. Our tech tools, all available free-of-charge, lower barriers to mapping. We are developing a streamlined, full end-to-end, user-friendly open mapping workflow and provide technical assistance while applying our solutions with communities.

We do not work alone but further expand community building and engagement for open mapping technology, to scale the technology movement for open mapping. HOT’s partnerships span government, civil society, and the private sector. Each brings specific skills and abilities that enrich the mapping ecosystem. Local partnerships are a vital part of how HOT works. We recognize the value of local expertise and knowledge and seek to amplify this through collaboration. Much of our training and technical support efforts are with local partners. Our technology partners are also based in priority countries (for example, OMDTZ in Dar es Salaam and Naxa Technologies in Kathmandu) and understand how to build tools for low-resource settings.

We have conducted a number of pilot projects using early, bespoke drone mapping  tools. In most cases, they are successful but they require more expensive equipment, or higher levels of technical knowledge than is usually available in vulnerable places.

We have built a workflow for people with inexpensive (~$1,000), locally-available drones to fly and collect data with quality that is equivalent to the outputs of expensive (~$20,000) drones. This includes:

•  A set of flight planning parameters

• Training materials

• Example datasets

• A methodology to collect Ground Control Points using low-cost, high-precision GPS receivers (inexpensive drones do not have onboard kinematically-corrected GPS)

• A processing pipeline using OpenDroneMap to convert the images into orthophotos and Digital Elevation Models

Using this workflow, we have conducted the following pilot projects:

In Janakpur, Nepal, we mapped one of the 24 administrative areas of the city (approximately 20km² and an area home to ~60,000 people) using drones and field mapping tools in partnership with the municipality. They  now want to do the other 23 areas, which will require more automation of the process.

In New Kru Town in Monrovia, Liberia, we have flown and mapped 2 km² of the city (home to approximately 35,000 people in ~6000 homes) at 5 cm resolution using a local operator with a DJI Mavic 2 Pro  in support of the local YMCA and the NGO Slum Dwellers International (SDI). SDI has requested that we continue mapping other areas of Monrovia, potentially as much as 75 km², which we are now beginning, still using semi-manual methods to coordinate and execute the flights.

In Dar es Salaam, Tanzania, we mapped 80km² of the Msimbazi River Basin for flood resilience as part of a World Bank project. This was done using a Sensefly Ebee, an expensive, heavy drone which is not appropriate for local community use. However, it showed the potential and appetite for such imagery. We are now working with the World Bank to shift this same type of work from expensive drones operated by foreign consultants to inexpensive drones operated by local people (pilot beginning in May 2024).

In Karangasem Regency, Bali, Indonesia, we worked with the local disaster management agency (Badan Penanggulangan Bencana Daerah or BPBD) to collect imagery and conduct field mapping in a remote village affected by landslides where the population is often augmented by thousands of people coming to the temple complex. The resulting imagery was used to digitize every structure in the village for a disaster risk assessment in the field. The flying was done using inexpensive, locally-available drones but required a great deal of expert piloting due to the extreme terrain, limiting the ability of the local agency to replicate and scale the effort. As  our tooling becomes more advanced, we anticipate BPBD will be able to fly more villages independently.

Also in Karangasem, Bali, BPBD was able to operate a locally-owned, inexpensive drone using our methodology to map approximately 600 homes near their office. Since the terrain is flat and manageable, they are able to continue doing this work on their own, albeit with more manual flight planning than is ideal.

We have received funding for two parts of the solution and are starting the implementation of this in Q4 2024:

• $250,000 CAD from Grand Challenges Canada to develop a mapping-optimized, community-appropriate drone  

• $240,000 USD from the World Bank to implement our Drone Tasking Manager and pilot it in St. Lucia and Dominica in the Caribbean

We want to accelerate development of this initiative with additional resources. The problem we are solving is highly technical, and  requires solutions that function in environments (heat, dust, poor supply chains, poor connectivity, low digital literacy, and occasionally hostile regulatory regimes) where traditional technology does not work well. As a result, we seek knowledgeable, enthusiastic collaborators, contributors, and sparring partners to assist with the technical aspects of our work. We see MIT Solve as a venue  to identify a range of capable and creative minds interested in partnering with us. 

We also consider that it’s valuable for more people to hear about the problem of aerial imagery for open mapping in vulnerable contexts, and to get excited about the solutions we are working on. MIT Solve is  a great platform to bring attention to the issue and attract minds and resources to work on it; the people who pay attention to this event are exactly the kind of people we need to collaborate and work with.

No other entity has made a serious attempt to create tools specifically for local people in low-income countries to gather aerial data using locally-appropriate tools. There exist a number of initiatives to crowdsource local imagery in poor areas, but all require operators with expensive drones, access to processing software and the specialized skills to use it, and provide little guidance or coordination. 

 A Canadian startup appears to be building a product that emphasizes simplifying the toolchain for individuals to create imagery and coordinate their efforts, but it is proprietary, requires payment by credit card (difficult for people in low-income settings to organize) and for the moment targets only Canadian users. Our solution is open, global, tailored for citizens of low-income countries, and squarely addresses the challenges faced by such citizens to participate in the creation of imagery. Local empowerment is at the heart of this innovation, it’s intended impact, and the expected change in the market/landscape.

The blockers for local people to create impactful aerial imagery are gaps in the toolchain, regulatory regimes, and open knowledge base. The conditions for success are present: data consumers are eager to pay for aerial imagery data, and the current consumer technology available to citizens of low-income countries is in theory capable of providing this data. Our theory of change is simple: if we remove the obstacles by creating the knowledge and tools for local people to satisfy the existing demand, generally beginning in places where the regulatory regimes permit such work, people will succeed, grow their knowledge and resources, and expand their activities. Neighboring regions will take note, and hopefully realize that a local drone industry is a positive thing, leading them to reduce the regulatory blockers. 

Empower people to create needed data, and they will do so. Show that this is beneficial, and others will realize this and allow it.

The UN SDG indicators on climate action includee indicators that are applicative to our solution:

• “Strengthen resilience and adaptive capacity to climate-related hazards and natural disasters in all countries”: Proportion of local governments that adopt and implement local disaster risk reduction strategies in line with national disaster risk reduction strategies

• “Promote mechanisms for raising capacity for effective climate change-related planning and management in least developed countries and small island developing States, including focusing on women, youth and local and marginalized communities” Number of least developed countries and small island developing States with nationally determined contributions, long-term strategies, national adaptation plans and adaptation communications, as reported to the secretariat of the United Nations Framework Convention on Climate Change

For above 2 indicators set by the UN, we do apply a variation and, we measure our progress in terms of how many communities and/or organizations are actively using our solution to make a positive change, as a key metric.

The core technology includes three parts:
- Building of low cost drones: assembly of a lightweight 3 fixed-focal length cameras, vibration damping, and data storage (not usable for military uses)
- Developing open software that enables coordinated flying.
This includes (a) a Drone operator portal, (b) Flight control, (c) Processing and data management tools
-Develop an open imagery platform to allow local drone operators to coordinate flights to collect consistent high quality imagery for entire cities flown using low-cost equipment. We have the building blocks in place with OAM (https://openaerialmap.org/) but this solution would include many more data processing capabilities, next to data sharing capabilities

We have been testing pieces of the solution and HOT has been working partly on similar solutions in the past. Projects like Tasking Manager (coordinated remote mapping for good; see tasks.hotosm.org and Field Mapping Tasking Manager (coordinated mobile mapping) are tools we have developed in the past and we know we can manage the coordination layer.
Building at type of “drone tasking manager” would in different ways build on similar concepts.

And, we know commercial software exists for parts of our solution, so we know that the biggest challenges will not be technical, rather ethical and legal.

HOT has a dedicated Technology & Data team of about 15 staff. Two of those staff are driving the entire solution while several staff members actively contribute in terms of product management and software engineering.

Hot has a long term partnerships with whom we aim to implement this solution. These include amongst others OpenDroneMap, Naxa Technologies, Flying Labs etc.. Next we work with a set of contractors that are experts in specific domains (e.g. an expert on drone flight planning etc.).

To test and improve iteratively, we work with our regional hubs that are connected to local communities. 

Hot’s technology & Data team have been working since 2010 on open source solutions in the mapping space. We have been long time using  satellite imagery as main source of remote sensing data, but since a few years we are more and more using drones. On the particular solution we are proposing now, we started about 1 year ago.

At HOT we have a Gender, Diversity, & Inclusion framework and have regular updates and activities regarding those.

Our leadership consists of 5 women, 3 man. Our CEO and Deputy CEO are both women. In our leadership team we have 1 leader from Senegal, one from Uruguay, one from Nepal and one from Kenia. 

Our organisation includes more than 50% women and more than 50% is based in our regional hubs. 

We have open community working groups where everyone is welcome to join and is heard

The reason we exist is exactly to ensure that missing data cannot be a cause for human suffering, we aim to include everyone, especially the least vulnerable communities in this goal.

We aim to create with this solution a revenue stream for local drone pilots. Local people in for example community settings can effectively produce high-quality aerial imagery projects using locally-appropriate equipment and skills. In many cases they can earn a living by doing so, while contributing to impactful datasets that improve development, disaster risk reduction, and resilience.

We would need diversified imagery customers. Beyond humanitarian and development agencies, potential markets include environmental agencies (monitoring the effects of climate change is a pretty solid use-case), governmental infrastructure agencies (transport and urban planning ministries, for example), private sector (the insurance industry is perhaps even more keen on vulnerability and resilience data than the aid sector), and so on. While we intend to focus on the humanitarian and development customers that we already know well, as the project unfolds we will be open to new markets for community drone imagery.

We think there's enormous potential revenue in aerial imagery. The drone manufacturing project and our aspirations to build a drone tasking manager are both intended to get us to a place where we and our associated communities can generate very valuable imagery, both as a foundation for our community and humanitarian mapping, but also as a revenue generator.

We have a very diversified funding strategy at HOT. We focus on high net worth individuals, Foundations as well as corporates and multinational institutions that are willing to support our work. We believe the best way to do so is to actually make an impact and tell those stories to people that care about our work. So far, we have received funding for two parts of the solution and are starting the implementation of this in Q4 2024:

• $250,000 CAD from Grand Challenges Canada to develop a mapping-optimized, community-appropriate drone  

• $240,000 USD from the World Bank to implement our Drone Tasking Manager and pilot it in St. Lucia and Dominica in the Caribbean