Rapid 3D Trajectory Planning Under State Constraints Using Receding Horizon PSO Algorithm

Abstract

For long-range Unmanned Aerial Vehicle (UAV) such as boost-glide hypersonic vehicle and unmanned combat aircraft, there is a rather large planning space for its flight trajectory due to the wide range of movement and long flight time. In this paper, a novel trajectory planning method is proposed based on Receding Horizon Particle Swarm Optimization (RHPSO) algorithm to realize rapid 3D trajectory planning despite a large planning space. First, the motion equations of mass center considering the rotation and oblateness of the Earth are established, and the change rates of attack angle, velocity roll angle and throttle setting are selected as decision variables which leads to the optimized results including the state constraints of trajectory. Then, by introducing receding horizon optimization strategy, the global optimization problem for the whole flight period is decomposed into local optimization problems within multiple receding horizon optimization windows, which reduces the computational complexity and realizes online rapid trajectory planning. In addition, in order to reduce blind search and speed up the optimization, a guidance strategy is proposed for the initialization of particle swarm, and in order to reduce the possibility of falling into local optimum caused by receding horizon optimization, a global heuristic strategy for specifying global heuristic feature points is hence proposed. The simulation results show that this method can effectively realize rapid 3D trajectory planning for long-range UAVs.