Autonomous operation of Unmanned Aerial Vehicles (UAVs) in complex, uncertain, highly-dynamic and often hazardous environments poses a fundamental problem for the use of such systems across a range of domains, from defence to environmental monitoring.

This session showcases recent and ongoing research into the use of cutting-edge AI methods, deployed at the ‘edge’ on low size, weight and power hardware, to improve the resilience of real-world UAVs and their ability to operate in such challenging environments.

This work spans a full range of scales, from enabling individual fixed-wing UAVs to carry out highly-dynamic, aerobatic manoeuvres all the way up to allowing rapid and resilient in-mission replanning for multi-UAV systems that need to adapt to unforeseen changes.