Collision-free Trajectory Generation for UAV Swarm Formation Rendezvous

Abstract

This paper proposes offline and online solutions to deal with a collision-free trajectory generation problem for unmanned aerial vehicle (UAV) swarm formation rendezvous. This method consists of two stages: 1) finding a geometric path based on improved ant colony optimization (ACO) with elite strategy and using minimum-snap algorithm with safe corridors to obtain feasible original trajectories so that the spatial-temporal constraints can be satisfied; 2) incorporating the zeroing control barrier function (ZCBF) constraints and the previous trajectories into the optimal problem to meet safety-critical requirement both for static obstacles and inter-UAV avoidance. The proposed method is tested in numerical simulation and the results demonstrate the efficacy in both dynamic constraints and safety constraints for the swarm.