移动机器人的一种烟花爆炸式新免疫规划算法  

New Fireworks Explosive Immune Planning Algorithm for Mobile Robots

作　　者：叶兆莉[1,2] 袁明新[1,2] 程帅[1,2] 王琪[1,2] 

机构地区：[1]江苏科技大学机械工程学院,江苏镇江212003 [2]江苏科技大学机电与汽车工程学院,江苏张家港215600

出　　处：《计算机仿真》2013年第3期323-326,375,共5页Computer Simulation

基　　金：国家自然科学基金资助项目(61105071);江苏科技大学人才引进项目(35271004);张家港校区青年基金项目(112110106);江苏省高校青蓝工程优秀青年骨干教师资助项目

摘　　要：针对移动机器人路径规划避障难和搜索路径等问题,要求机器人从起点到终点能搜索一条最优无碰路。为解决上述问题,提出了一种新的烟花爆炸式免疫算法(FEIA)。在免疫遗传算法(IGA)基础上,引入烟花爆炸机制进行种群更新,即在算法进化过程中,当种群达到预设爆炸代数时,从种群中提取若干较优个体和若干较差个体,将较优个体进行邻域扩展,并对扩展结果与较差个体择优进行种群重组。函数优化结果表明,与其它算法相比,FEIA收敛速度更快,搜索精度更高,且能有效地解决早熟收敛问题。而路径规划结果表明,在不同复杂环境中,FEIA能实现机器人的最优路径搜索及避障,显示出较强的搜索能力和鲁棒性。To solve difficulties in obstacle avoidance and weak convergence in planning path in mobile robot path planning, a new type of fireworks explosive immune algorithm（FEIA） was put forward. Developed on the basis of immune genetic algorithm （IGA） , the population was updated by introducing fireworks explosion mechanism when the population was evolved to certain generations, and some better and worse individuals were found. Then the better ones in its neighborhood were expanded to get the expanded individuals which were compared with the worse according to their fitness, and finally the better was saved to the next optimization. Function optimization results show that FEIA convergences faster than other optimization algorithm, and can effectively solve the problem of premature convergence and is easy to reach a higher precision. The results of path planning simulation show that FEIA shows strong search a- bility and robustness, and can realize the robot optimal path search and obstacle avoidance in different complex envi- ronments.

关 键 词：移动机器人 烟花爆炸式进化 路径规划 免疫遗传算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]