A time-evolving topology based obstacle avoidance algorithm for multi-UAV formation

During avoidance of obstacles, an Unmanned Aerial Vehicle (UAV) team confronts with varying communication distances and intermittent visibilities among the member nodes, leading to a time-evolving communication topology. In this paper, therefore, we present a time-evolving based avoiding algorithm for a teamed Unmanned Aerial Vehicle (UAV) system in a two-dimension environment with dynamic obstacles. In one snapshot of the time-varying topology, especially, each member node computes out a convex polygon-based hull of the next-step positions set by making distributed consensus with neighbor nodes. With a centralized approach, the team determines the largest convex region by using these obtained convex hulls within a two-dimension geometric space, where each robot will locally compute the optimal parameters for its next proper position within the resulted convex region. From the simulation results, for a dynamic clutter environment, the proposed approach presents obviously less communication overheads, less time cost and scalable with the formation size.