T-STAR: Time-Optimal Swarm Trajectory Planning for Quadrotor Unmanned Aerial Vehicles
Honghao Pan, Mohsen Zahmatkesh, Fatemeh Rekabi-Bana, Farshad Arvin, Junyan Hu
Abstract
This paper introduces a time-optimal swarm trajectory planner for cooperative uncrewed aerial vehicle (UAV) systems, designed to generate collision-free trajectories for flocking control in cluttered environments. To achieve this goal, model predictive contour control is utilised to generate time-optimal trajectories for each UAV. By demonstrating the differential flatness dynamic equations, the system state constraints are simplified, the algorithm’s complexity is reduced, and the overall stability is improved. Additionally, flocking control is achieved among multiple UAVs by applying virtual repulsive and attractive forces. Furthermore, an event-triggered trajectory deconflict strategy for trajectory replanning is considered to resolve multiple trajectory conflicts. Comparative experiments with baseline methods have confirmed that the proposed planner can generate faster and safer trajectories than conventional methods.