论文部分内容阅读
对于环境信息完全已知的移动机器人的全局路径规划问题,应用了一种并联的神经网络结构与模拟退火算法相结合的方法,并提出了一种局部路径修正算法,最终得到一条最优的平滑路径。计算机仿真研究表明,该算法计算简单,收敛速度快,规划的路径为一条最短无碰且安全的平滑路径。在计算机仿真验证的基础上,以P3-AT型轮式移动机器人为平台,通过机器人模拟实验验证了该算法的有效性。
For the global path planning of mobile robot with completely known environmental information, a parallel neural network structure and simulated annealing algorithm are combined and a local path correction algorithm is proposed. Finally, an optimal smoothing path. Computer simulation results show that the proposed algorithm is simple in computation and fast in convergence, and the planned path is a shortest non-collision and safe smooth path. On the basis of computer simulation and verification, the effectiveness of this algorithm is verified by a robot simulation experiment using the P3-AT wheeled mobile robot as a platform.