Hybrid system design for formations of autonomous vehicles

摘要

Cooperative control of multiple unmanned aerial vehicles (UAVs) poses significant theoretical and technical challenges. Recent advances in sensing, communication and computation enable the conduct of cooperative multiple-UAV missions deemed impossible in the recent past. We are interested in solving the Formation Reconfiguration Planning (FRP) problem which is focused on determining a nominal state and input trajectory for each vehicle such that the group can start from the given initial configuration and reach its given final configuration at the specified time while satisfying a set of given inter- and intra- vehicle constraints. Each solution of a FRP problem represents a distinct reconfiguration mode. When coupled with formation keeping modes, they can form a hybrid automaton of formation maneuvers in which a transition from one formation maneuver to another formation maneuver is governed by a finite automaton. This paper focuses on the implementation of the optimized hybrid system approach to formation reconfiguration for a group of 1 real and 3 virtual UAVs. Experimental results performed in the Richmond Field Station by using a helicopter-based Berkeley Aerial Robot are presented.