Optimal Spacecraft Formation Reconfiguration with Collision Avoidance Using Particle Swarm Optimization

Abstract

This paper presents an energy-optimal trajectory planning method for spacecraft formation reconfiguration in deep space environment using continuous low-thrust propulsion system. First, we employ the Legendre pseudospectral method (LPM) to transform the optimal reconfiguration problem to a parameter optimization nonlinear programming (NLP) problem. Then, to avoid the computational complexity for calculating the gradient information caused by traditional optimization methods, we use particle swarm optimization (PSO) algorithm to solve the NLP problem. Meanwhile, in order to avoid the collision between any pair of Legendre-Gauss-Lobatto (LGL) points, we insert some test points in the region where collision may happen most likely. What’s more, the collision avoidance constraints are also checked at these test points. Finally, numerical simulation shows that the energy-optimal trajectories for spacecraft reconfiguration could be generated by the method we proposed in a relative short time, so that it could be adopted on-board for practical spacecraft formation problems.

DOI: http://dx.doi.org/10.5755/j01.itc.41.2.854