一种混合优化的全局路径规划算法

针对全局路径规划研究中遗传算法存在搜索范围广而导致收敛速度慢的问题,本文提出一种混合优化的全局路径规划方法,完成对图像读取、处理后使用A*算法预处理缩小可行区域从而提高收敛速度。所提出的混合优化规划方法主要优化遗传算法的初始种群,在不影响最终路线的情况下,缩小初始种群的搜索范围,提高算法进行全局路径规划的速度,快速有效的规划出全局路线。另外本文给出一种评价体系对规划结果进行定量的避障评价,评价结果能够以数值形式对规划结果进行综合评价,评价结果显示通过混合优化算法规划出的路径具有更佳的安全性。

A hybrid optimization algorithm for global path planning

In order to solve the problem of slow convergence due to the wide search range of genetic algorithm in global path planning, this paper proposes a hybrid optimization global path planning method, which can reduce the feasible area and improve the convergence speed after reading and processing the image. The proposed hybrid optimization planning method mainly optimizes the initial population of genetic algorithm, reduces the search scope of the initial population, improves the speed of global path planning, and quickly and effectively plans the global path without affecting the final path. In addition, this paper presents an evaluation system for quantitative obstacle avoidance evaluation of planning results. The evaluation results can comprehensively evaluate the planning results in numerical form. The evaluation results show that the path planned by hybrid optimization algorithm has better security.