Collision Avoidance: A Game Theoretic Approach
Steven Snyder, Naira Hovakimyan · 2015
This paper presents a game theoretic approach to collision avoidance of multirotor unmanned aerial vehicles (UAVs). The approach is intended for use in a cooperative mission system that includes trajectory generation, path-following, and time-coordination. Under the given conditions, a priori bounds are provided for the deviation from the path due to collision avoidance. The trajectory generator can then take this bound into account, along with the path-following and time-coordination bounds, in order to guarantee that cooperative vehicles using these algorithms will maintain some pre-specified separation in space-time. Simulation results demonstrate the approach with UAVs considered as both the cooperative agents and the obstacles.