WFP is taking early steps to harness the latest artificial intelligence technology to enhance our emergency response capabilities.

Unmanned Aerial Vehicles (UAVs), widely known as drones, are aircrafts that can fly remotely with no on-board pilot. The exceptional mobility and accessibility associated with this evolving technology is attracting increasing interest from humanitarian organisations.

The use of UAVs in humanitarian contexts could bring many benefits, including higher data accuracy, faster data collection, and safer monitoring systems. With the benefits, however, come several challenges, ranging from legal issues, ethical procurement and partnerships, and data protection, to transparency, privacy and community perception.

Many disaster-prone countries, which could benefit from the deployment of UAVs in when responding to emergencies, have no regulations or guidelines on the use of this technology. Like other sectors of humanitarian response, the use of UAVs needs pre-defined agreements and coordination. With awareness and understanding of the potential benefits of UAVs increasing, it is vital to facilitate and support the humanitarian community by working more efficiently and effectively in a coordinated matter with all actors.

As the global lead of the UN’s Emergency Telecommunications Cluster (ETC), the World Food Programme (WFP) has been approached by different UAV actors to lead coordination of this services. With the support of the Government of Belgium, WFP is designing UAV coordination solutions for implementation within the broader humanitarian emergency preparedness and response community. The coordination model will be implemented in a total of five high risk countries before the end of 2017.

Use of Artificial Intelligence

While UAVs provide a platform to efficiently collect remotely sensed image data, the data must be processed and analysed in order to extract meaningful information. This image analysis is normally achieved through visual interpretation by an experienced remote sensing analyst; a manual approach requires a very high workload to be carried out. It means that data often remains unused, because of time constraints. Automating or partially automating some of the interpretation with machine learning techniques would greatly reduce the time required to carry out the analysis and help get better information into the hands of emergency coordinators faster.

As part of the current pilot project Rapid UAVs for Data Analysis in Emergencies (RUDA), the machine learning software will be integrated with the data collection and analysis workflows, and be implemented at first instance in five countries prone to natural disasters, in which the UAV coordination model is being developed.

Since enhanced emergency response can be useful for many humanitarian actors, WFP invites interested organisations to reach out and explore opportunities for collaboration.