What is the challenge that is being addressed?

Disaster response involves the consideration of an abundance of time-varying information as well as the coordination of a substantial number of first responder assets as efficiently as possible. The complexity of response operations can be even more accentuated according to the scale of the disaster and weather conditions. For the above reasons, many first responder organizations leverage unmanned aerial vehicles (UAVs) in rescue missions, especially in impervious areas.
The deployment of more UAVs naturally increases mission efficiency, e.g. reduced search time in a Search-and-Rescue (SAR) mission. Current operations involve at least two drone operators for each drone: one being the pilot, while the other manoeuvres the mission payload (i.e. camera). Therefore, increasing the UAV fleet size requires scaling up the size of the drone operator team, which is extremely challenging. Besides, increased personnel also complicates communication and mission planning, making assignments of tasks and operating areas to drones error prone.  

What is the AI solution the project plans to implement?

Deploying more UAVs in view of increasing mission efficiency is only viable if the coordination of the unmanned vehicles is highly automatized. Therefore, an automated mission planning system will be highly beneficial in this setting. Coupled with asynchronous communication techniques developed for robotics, plan execution and monitoring can also be carried out in a more flexible manner, enabling planning and acting in a close loop and thereby increasing also mission autonomy even in dynamic environments.