Communication-Free MPC-Based Neighbors Trajectory Prediction for Distributed Multi-UAV Motion Planning

Abstract

In an environment with multiple static obstacles, UAVs usually communicate with each other to avoid collisions during trajectory planning. However, such communication may become infeasible or unreliable due to interference or jam in practice. This paper introduces a neighbors trajectory prediction algorithm based on model predictive control (MPC), which enables each UAV to predict the motion behavior of its neighbors without communication. By solving the MPC model of its neighbors, an UAV can predict their trajectories and then avoid collision with them in the future. To prove the practicability, we integrate the proposed algorithm into distributed model predictive control (DMPC) framework to realize multi-UAV trajectory planning without communication and with static obstacles. The performance of our method is verified by simulation experiments in two scenes.