Role Differentiation Mechanism for Aerial Group Search Planning within Compound Constraints
摘要
Cooperative search through multiple unmanned aerial vehicles (UAVs) is a crucial engagement for UAV systems. The mission effectiveness could be enhanced through multi-robot collaborations, whereas the collaborative constraints impose limitations on individual behaviors. The compound constraints, including connectivity maintenance and obstacle avoidances, can restrict the action selections of an individual UAV and deteriorate the global search performance accordingly. This paper proposes a decision-making approach for multi-UAV system by considering role differentiation scheme between informing and relay UAVs to address the problems above. The informers are allowed to switch to the relay nodes when no feasible actions could be found to sustain their search process, while the relays can release the compound constraint on the informers’ actions by providing larger communication distance. A sequential auction-based decision-making framework is designed to formulate the coordination and search behaviors intra and inter the two subgroups. Dubins curves are used to generate local search paths. Simulation results validate the higher task successes of the proposed method over some existing homogeneous methods, showing its promising tradeoff between explorations, connectivity, and sparse obstacle avoidances.