计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (17): 162-168.DOI: 10.3778/j.issn.1002-8331.1903-0387

• 模式识别与人工智能 • 上一篇    下一篇

改进遗传算法在移动机器人路径规划中的应用

孙波,姜平,周根荣,卢易天   

  1. 南通大学 电气工程学院,江苏 南通 226019
  • 出版日期:2019-09-01 发布日期:2019-08-30

Application of Improved Genetic Algorithm in Path Planning of Mobile Robots

SUN Bo, JIANG Ping, ZHOU Genrong, LU Yitian   

  1. School of Electrical Engineering, Nantong University, Nantong, Jiangsu 226019, China
  • Online:2019-09-01 Published:2019-08-30

摘要: 提出了一种应用于机器人路径规划的改进自适应遗传算法。在遗传算法的选择操作中引入模拟退火思想,以此来提高算法的全局搜索能力;对交叉、变异算子自调整策略进行改进,以提高算法的收敛速度;将规划出的路径做平滑优化处理,并根据路径与障碍物间的距离进行不同速度段的划分;在适应度函数中加入安全行驶速度和转弯次数等多个规划指标,使规划出的路径更加安全高效。仿真实验表明,改进后的算法实现效率好、安全可靠性高,规划出的路径也更加符合实际情况。

关键词: 路径规划, 改进遗传算法, 模拟退火, 自调整策略, 平滑处理, 多指标规划

Abstract: An improved adaptive genetic algorithm for robot path planning is proposed. Firstly, the simulated annealing idea is introduced in the selection operation of genetic algorithm to improve the global search ability of the algorithm. Secondly, the crossover and mutation operator self-adjustment strategy is improved to improve the convergence speed of the algorithm. Then, the planned path is smoothed and optimized, and different speed segments are divided according to the distance between the path and the obstacle. Finally, multiple planning indicators such as safe driving speed and turning times are added to the fitness function to make the planned path safer and more efficient. Simulation experiments show that the improved algorithm has good implementation efficiency, high security and reliability, and the planned path is more in line with the actual situation.

Key words: path planning, improved genetic algorithm, simulated annealing, self-adjusting strategy, smooth processing, multi-indicator planning