论文部分内容阅读
针对障碍物分布复杂、存在封闭边界的受限空间,提出一种环境自适应区域栅格化的优化路径规划算法.该算法首先将环境自适应划分为区域栅格,并提出阻碍度指标降低搜索空间的维度以优化区域栅格的划分;然后结合随机变异和定向变异,给出一种可有效平衡搜索效率与精度矛盾的多维变异粒子群优化算法;最后使用最小二乘曲线拟合方法对优化路径予以平滑处理.与非线性递减惯性权值粒子群算法(NDW-PSO)及组合粒子群算法(C-PSO)对比的仿真结果验证了所提出算法的先进性.
In view of the complicated distribution of obstacles and the confined space with closed boundaries, this paper proposes an optimized path planning algorithm for environment adaptive region rasterization. Firstly, the environment is adaptively divided into regional rasters and the obstruction index is lowered to search Dimensional space to optimize the division of the grid; then combined with random variation and directional variation, a multidimensional mutation particle swarm optimization algorithm that can effectively balance the contradiction between search efficiency and accuracy is given; Finally, the least squares curve fitting method is used to optimize And the path is smoothed.The simulation results of NDW-PSO and C-PSO verify the advanced nature of the proposed algorithm.