Aiming at the problem of spacecraft attitude maneuver planning under multiple mandatory pointing constraints and prohibited pointing constraints, based on pigeon-inspired optimization (PIO), we proposed an improved policy gradient reinforcement learning (RL) algorithm (PIOPGRL). First, we establish an angle-based attitude constraint model, and then, we establish the reward function of RL based on the model. Then, the fitness function is used to replace the policy evaluation function, so PIOPGRL is integrated with RL. The PIOPGRL algorithm uses the PIO algorithm to solve the policy gradient, significantly reduces the amount of calculation and accelerating the convergence speed. The simulation results show that the spacecraft attitude maneuvering path planning method based on PIO-improved RL (PIOPGRL) has better planning results and lower cost of maneuver than the classical PGRL algorithm, which can solve the problem of spacecraft attitude maneuver planning under multiple pointing constraints perfectly.

To solve the problem of spacecraft attitude manoeuvre planning under dynamic multiple mandatory pointing constraints and prohibited pointing constraints, a systematic attitude manoeuvre planning approach is proposed that is based on improved policy gradient reinforcement learning. This paper presents a succinct model of dynamic multiple constraints that is similar to a real situation faced by an in-orbit spacecraft. By introducing return baseline and adaptive policy exploration methods, the proposed method overcomes issues such as large variances and slow convergence rates. Concurrently, the required computation time of the proposed method is markedly reduced. Using the proposed method, the near optimal path of the attitude manoeuvre can be determined, making the method suitable for the control of micro spacecraft. Simulation results demonstrate that the planning results fully satisfy all constraints, including six prohibited pointing constraints and two mandatory pointing constraints. The spacecraft also maintains high orientation accuracy to the Earth and Sun during all attitude manoeuvres.

针对航天器编队重构的路径规划问题,考虑燃料消耗和碰撞概率等约束条件,以及基本鸽群算法存在的问题,提出一种基于混沌初始化和高斯扰动的自适应鸽群(C GA-PIO)算法.为了得到多样性和覆盖性更好的鸽群初始值,采用Tent Map混沌模型进行鸽群初始化操作;在地图和指南针算子阶段,为提高全局搜索能力,引入了自适应的权重因子和学习因子更新个体的位置和速度;在地标算子阶段,为避免算法陷入局部最优,将高斯扰动加入到鸽群中心位置.仿真实验结果表明:CGAPIO算法与基本鸽群算法和粒子群算法相比,提高了全局搜索能力,避免了局部最优,规划得到的路径更加平滑,各航天器碰撞概率较低,编队重构消耗的总燃料至少减少了12％.

The orbit planning problem of spacecraft clusters under complex conditions is a hotspot and a difficult goal in the current aerospace field. This paper studies the orbital optimal programming problem of distributed cluster spacecraft in the process of formation transformation and proposes the population evolution algorithm adaptive population variation pigeon-inspired optimization (APVPIO). Based on the core evolutionary algorithm and evolutionary stagnation, there is a high tendency to fall into the local optimal solution in the classical PIO algorithm. The fitness function of the classical PIO algorithm is studied and improved with the orbit planning problem. Finally, the simulation based on the adaptive population variation algorithm is performed. The results reveal that the APVPIO algorithm has better planning results, deeper population evolution depth, and a faster convergence speed in comparison to the classical PIO algorithm and particle swarm algorithm (PSO) algorithm. Hence, it has the potential to meet the complexity requirement of spacecraft clusters and orbital planning problems.