摘要
针对传统移动机器人路径规划方法存在的不足,基于独特型免疫网络理论,提出一种改进的免疫网络算法(MINA)用于移动机器人路径规划问题.该算法采用一种新的抗体对称均匀变异成熟机制和子群稳定判定策略,减少了算法对抗体克隆规模的过于敏感性,大大降低了算法的计算量;为真正体现免疫网络动态调节机制,增加抗体的多样性,提出了一种基于抗体亲和度和浓度的选择方法.仿真实验结果表明,该算法能使移动机器人在较复杂环境下快速找到一条优化路径,与同类算法相比具有一定优越性,是一种有效的移动机器人路径规划算法.
Aimed at the deficiencies of traditional mobile robots path planning methods, a modified immune network algorithm （ MINA） for mobile robot path planning based on idiotypic immune network theory is proposed in the paper. In order to overcome the shortcomings of the traditional opt-aiNet, such as the heavy computational cost and too sensitive to the clone sizes of antibody, a new antibody symmetrical mutation maturation mechanism and subpopulation stabilization determination strategy are used. In order to really embody the dynamic adjustment mechanism of immune network and maintain the diversity of population, an immune selection mechanism based on density and fitness is devised. The simulation experimental results show that the new algorithm can make the mobile robot to rapidly find the optimal path in complex environment. Compared with other algorithms, MINA has certain advantages, and is an effective mobile robot path planning algorithm.
出处
《小型微型计算机系统》
CSCD
北大核心
2014年第6期1437-1440,共4页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61174013)资助
江苏高校优势学科建设工程项目资助
关键词
独特型免疫网络
对称变异成熟机制
移动机器人
路径规划
idiotypic immune network
symmetrical mutation maturation mechanism
mobile robot
path planning
