基于改进遗传算法的餐厅服务机器人路径规划

发布时间：2018-06-19 05:02

本文选题：遗传算法 + 餐厅服务机器人　；参考：《计算机应用》2017年07期

【摘要】：针对遗传算法(GA)易产生早熟现象和收敛速度慢的问题,提出了一种基于传统遗传算法(TGA)的改进遗传算法——HLGA,用于实际餐厅服务机器人的路径规划。首先,通过基于编辑距离的相似度方法对拟随机序列产生的初始种群进行优化;其次,采用自适应算法的改进交叉概率和变异概率调整公式,对选择操作后的个体进行交叉、变异操作;最后,计算具有安全性评价因子函数的个体适应度值,进一步对比、迭代得到全局最优解。理论分析和Matlab仿真表明,与TGA和基于个体相似度改进的自适应遗传算法(ISAGA)相比,HLGA的运行时间分别缩短了6.92 s和1.79 s,且规划的实际路径更具有安全性和平滑性。实验结果表明HLGA在实际应用中能有效提高路径规划质量,同时缩小搜索空间、减少规划时间。
[Abstract]:Aiming at the problems of premature convergence and slow convergence rate of genetic algorithm (GA), an improved genetic algorithm (HLGA) based on traditional genetic algorithm (TGA) is proposed, which can be used in path planning of practical restaurant service robot. Firstly, the initial population of quasi random sequences is optimized by using the similarity method based on editing distance. Secondly, the improved crossover probability and mutation probability adjustment formula of adaptive algorithm is used to cross the selected individuals. Finally, the individual fitness value with the security evaluation factor function is calculated, and the global optimal solution is obtained by further comparison. Theoretical analysis and Matlab simulation show that the running time of HLGA is 6.92 s and 1.79 s shorter than that of TGA and improved adaptive genetic algorithm (ISAGA) based on individual similarity, and the actual path planning is more secure and smooth. The experimental results show that HLGA can effectively improve the quality of path planning, reduce the search space and reduce the planning time in practical applications.
【作者单位】：中国计量大学质量与安全工程学院;
【基金】：国家公益性行业科研项目(201410028-02)~~
【分类号】：TP18;TP242

【参考文献】