Baldwin效应的自适应免疫网络规划算法  

Adaptive Immune Network Algorithm Using Baldwin Effect for Path Planning

作　　者：袁明新[1] 王孙安[1] 吴灿阳[1] 陈乃建[1] 

机构地区：[1]西安交通大学机械工程学院,西安710049

出　　处：《西安交通大学学报》2009年第5期85-89,113,共6页Journal of Xi'an Jiaotong University

摘　　要：针对复杂环境下的机器人路径规划问题,提出了一种新的免疫网络规划算法.借鉴独特型网络假设,将机器人环境作为抗原,机器人行为作为抗体,通过抗原抗体的激励和抑制来构建免疫网络.为了进一步提高算法的网络搜索能力,根据Baldwin效应构建了抗体生命力更新算子,并对抗体生命力衰减系数进行了自适应调整.仿真实验结果表明,新算法具有自组织和自学习的特点,在收敛性能、规划能力方面有很大提高,能够较好地解决复杂环境下的路径规划问题.To realize path planning in complicated environments, a new immune network algorithm for path planning is presented. Inspired by the mechanism of idiotypic network hypothesis,an immune network is constructed with the stimulation and suppression between the antigen and antibody by taking the environment and robot behavior as antigen and antibody respectively. To further improve the searching capability of proposed algorithm, an updating operator for antibody vitality is provided according to Baldwin effect, and the attenuation coefficient of antibody vitality is adaptively adjusted. The simulation results show that the proposed algorithm is characterized by self-organizing and self-learning, and the convergence performance and planning capacity are remarkably improved.

关 键 词：BALDWIN效应 免疫网络 路径规划 衰减系数 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]