【摘要】 空间机器人每次携带的燃料有限,提高空间机器人的工作效率以及延长其在轨寿命研究具有重要意义,分析了空间机器人多空间站访问问题。为了弥补传统路径规划方法容易陷入局部极小点的问题,提出利用基于蚁群算法的Hopfield神经网络来解决空间机器人多空间站访问问题。仿真实验结果表明,基于蚁群算法的Hopfiled神经网络用于多空间站访问问题,收敛速度要比Hopfield神经网络快,且比Hopfield神经网络易于跳出局部极点,该算法有利于解决多空间站路径规划问题。更多还原

【Abstract】 As each carrying fuel of space robot is limited,it is very important to improve the efficiency of space robot and extension of its in-orbit life.This paper analyzed multi space station path planning problem.To improve the easier occurring of stagnation behavior,used Hopfield neural network based on ant system in multi-station path planning problem.Simulation result shows that in contrast with conventional Hopfield neural network,convergence rate of this algorithm is faster than Hopfield neural network and can obtain good path.The algorithm will help to solve the path planning of multi space station.更多还原