Frontier Innovations

New technologies are rapidly developing and deeply transforming our lives and the world, including in humanitarian aid.

The Frontier Innovations programme at the WFP Innovation Accelerator explores game-changing innovations and new technologies such as artificial intelligence, blockchain, autonomous vehicles, edge computing and robotics to support humanitarian and development assistance.

Together with partners like the European Space Agency, the German Space Agency, and Google Research, we examined complex problems from AI-assisted mapping and evaluation of damage after natural and anthropogenic disasters, to how remote-operated vehicles can support the delivery of assistance in complex environments.

Supported humanitarian
and development assistance for
30,000 people
through Frontier Innovations

SPOTLIGHT

WFP-X Moonshot Launchpad

In 2020, WFP Innovation Accelerator launched WFP-X, with the aim of creating 100 moonshot ideas in 100 days. The first cohort took place in Dar es Salam, Tanzania, with 9 local innovators, who then ideated and tested five moonshot ideas. In 2021, local innovators developed prototypes of the five most promising ideas, tested them with residents in Dar es Salaam, and pitched the results to a group of investors, government officials, and NGOs. Given the potential for impact and promising results, two of the prototypes - Forever Food and Mr Bins Gas - received funding from the Embassy of Switzerland for further development.

READ THE FULL STORY HERE

Ahead

For the AHEAD project (Autonomous Humanitarian Emergency Aid Devices), WFP collaborated with German Aerospace Centre and its partners, Sensodrive and Roboception, to develop tele-operated and semi-autonomous driving for amphibious all-terrain vehicles in order to reduce risks for drivers of aid deliveries. The project started in October 2020. Since then, the project has achieved several technological developments. The scope of the project has also increased with the acquisition of two more projects. The VDI/VDE in Germany funded EUR 3 million for the project. The funds would be utilized for development of LiDAR (Light Detection and Ranging) systems that would extend the camera based perception of the AHEAD vehicle with active sensor technology. The new LiDAR system would increase reliability and safety of the tele-operated vehicle.

EXPLORING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (AI) IN THE HUMANITARIAN SPACE

SKAI

A lack of on-the-ground information at the start of a humanitarian crisis is a major obstacle to a quick and effective response. In collaboration with Google Research, SKAI uses AI to analyze satellite images to automatically assess damage post disasters. It drastically speeds up the process of extracting insights from the ground, and enables speedy and accurate emergency response. In 2021, we reached an agreement with Google Research to open-source the machine learning model, and we have been working closely to prepare the source codes to be shared with the broader AI community.

Voice-to-Text AI Phone Survey Tool

Natural Language Processing (NLP) is a very powerful AI technology that helps WFP have a deeper and more instant understanding of the needs of people accessing WFP assistance. This is especially useful when many people served are illiterate. Our goal is not only to reduce the cost of conducting remote surveys, but also to obtain qualitative insights from target populations.

Voice-to-text AI Phone Survey Tool was initially explored as part of the Frontier Innovation programme and has now proceeded to the Sprint programme.

EXPLORING APPLICATIONS OF BLOCKCHAIN IN THE HUMANITARIAN SPACE

Atrium

Atrium is a global collaboration network for learning, engaging and experimenting with blockchain technology. The Atrium provides a safe space for United Nations (UN) staff members to learn about blockchain, engage with microtasks, and discuss on the forum. Underpinning the collaboration platform is a Quorum platform that creates a safe sandbox for blockchain experimentation. We are currently developing version 2.0 and have three nodes (WFP, UNICEF, UNDP) as well as more than 500 members from more than 35 United Nations agencies. It is the largest institutional collaboration platform in the United Nations system.

Blocks for Transport

The Blocks for Transport project aimed to streamline the logistics process and reduce delays in the Djibouti-Ethiopia corridor, one of the key supply chain corridors for WFP in East Africa, by developing a blockchain-enabled supply chain platform. Despite progress made with the development of the second version of the minimum viable product, the project had to be postponed indefinitely due to the emergency situation in the country. We tried to pivot by exploring other potential supply chain corridors where this platform could be used, however, a good use-case did not materialize. This exemplifies the realities of innovation journey within the humanitarian sector - at times, unforeseen events on the ground can disrupt the continuation of a given project.

Smallholder Farmer Procurement platform
Kenya (initial exploration)

We launched the design phase to explore the applicability of using a blockchain-based platform that connects key stakeholders to streamline procurement process of grain from smallholder farmers in Kenya. The project also aimed to provide better market access to smallholder farmers to sell their grain. The initial design and research phase was completed in 2021 and we expect to develop and test in 2022.

The Frontier Innovations programme prioritized the aggregation and dissemination of the knowledge of emerging technologies with the wider community.

The team co-authored a WFP policy brief on “Machine Learning for Damage Assessment” that was included in the Interagency Task Team on Science, Technology and Innovation for the Sustainable Development Goals - a key forum that informs senior United Nations leadership on the impact of rapid technology change on the achievement of the SDGs;
Provided direct testimonies and analysis to the Joint Investigation Unit (JIU) review of blockchain applications in the United Nations system, which helped the United Nations move forward in its exploration of this emerging technology; and
Building on the work co-authoring Assessing Post-Disaster Damage from Satellite Imagery using Semi-Supervised Learning Techniques in 2021, the team spearheaded the first ever workshop series on “Computer Vision for Damage Assessment”, to facilitate discussion for 14 like-minded organizations and partners to co-develop a shared vision and common approach in applying AI and computer vision-powered tools in post-disaster damage assessment, emergency response, and rehabilitation.

Year in Review

2021: A year of impact and growth

Download the report