基于势场法优化的蚁群免疫网络路径规划研究.pdfVIP

第21 卷第15 期 系 统 仿 真 学 报© Vol. 21 No. 15 2009 年8 月 Journal of System Simulation Aug., 2009 基于势场法优化的蚁群免疫网络路径规划研究 袁明新，王孙安，李昆鹏，陈乃建 （西安交通大学机械工程学院，西安 710049 ） 摘 要：借鉴独特型网络假设和蚂蚁觅食行为的有关机理，提出了一种蚁群免疫网络路径规划算 法。将机器人环境作为抗原，机器人行为策略作为抗体，通过抗原抗体的激励和抑制构建免疫网 络，同时利用蚁群算法进行网络搜索，改善了免疫网络的最优规划。为了进一步提高算法收敛速 度，将人工势场法的规划结果作为先验知识，通过疫苗抽取和接种实现网络初始化。仿真实验表 明，新算法具有收敛速度快，规划路径短以及自学习的优点。 关键词：路径规划；免疫网络；蚁群算法；人工势场 中图分类号：TP24 文献标识码：A 文章编号：1004-731X (2009) 15-4686-05 Ant Colony and Immune Network Algorithm Based on Optimization of Potential Field Method for Path Planning YUAN Ming-xin, WANG Sun-an, LI Kun-peng, CHEN Nai-jian (School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an 710049, China) Abstract: Inspired by the mechanisms of idiotypic network hypothesis and ant finding foods, an ant colony and immune network algorithm (AC-INA) for path planning was proposed. Taking the environments surrounding the robot and action strategies as antigens and antibodies respectively, an immune network was constructed by the stimulation and suppression between the antigen and antibody, and the path planning was carried out by the search of ant colony algorithm, which improved the optimal planning of immune network. To further quicken the convergence speed of AC-INA, the path planning results of potential field method were taken as the prior knowledge, and the network was initialized by the vaccine extraction and inoculation. Simulation results indicate that the proposed algorithm (AC-INA) is characterized by high