Hybrid Control for Uav-assisted Search and Rescue

We develop a decentralized hybrid controller for fixed-wing UAVs assisting a manned helicopter in a United States Coast Guard search and rescue mission. The UAVs assist the manned helicopter by providing an expanded sensor footprint using onboard cameras. We consider two UAVs, one flying on either side of the helicopter, with constant velocity and maximum turn rate constraints. Tracking the helicopter around sharp corners will be difficult due to these constraints and the difference in path lengths for the two UAVs. To solve this problem, we propose a hybrid controller that allows the UAVs to swap positions in an attempt to improve the tracking and ground coverage performance of the formation. We discuss tracking control, the position swapping algorithm and collision avoidance. Simulation results demonstrate improved search efficiency and aircraft safety. keywordsunmanned aerial vehicles, waypoint tracking, hybrid control, conflict resolution INTRODUCTION The search and rescue scenario includes two fixed-wing UAVs tracking a manned helicopter flying in a standard United States Coast Guard (USCG) search pattern [1]. It is desired to maintain an aircraft formation that provides continuous sensor coverage of the ground although the helicopter is more maneuverable than the UAVs. The following assumptions are made: 1. UAVs fly at a constant altitude and airspeed and have a limited turn rate. ∗Address all correspondence to this author. 2. The helicopter has a much smaller minimum turn radius than the UAVs. Its maximum speed is less than or equal to the airspeed of the UAVs and it flies at a different altitude. 3. The search and rescue flight pattern executed by the helicopter consists of sharp turns that the UAVs will not be able to execute. 4. There is two-way communication between the UAVs, and the positions of all aircraft are known using GPS. Each UAV is equipped with an autopilot such as the Piccolo from Cloud Cap Technologies. This system controls the UAV’s non-linear flight dynamics, receiving commands such as desired altitude and yaw rate from a higher-level application. The search and rescue controller produces a desired heading angle (ψdes) for each UAV, and then commands the desired yaw rate to the autopilot. The yaw rate command is generated by a simple proportional controller shown in Eqn. (1), where ψ is the heading angle and ψ̇ is the yaw rate command.