平行导引律优化的多机器人免疫围捕算法  

Multi-Robot Immune Hunting Algorithm Optimized By Parallel Guidance Law

作　　者：袁明新[1,2] 黄艇 高云强 刘锁东 刘维 YUAN Ming-xin;HUANG Ting;GAO Yun-qiang;LIU Suo-dong;LIU Wei(School of Mechanical Engineering,Jiangsu University of Science and Technology,Zhenjiang 212003,China;T-SEA Marine Technology Co.,Ltd.,Zhangjiagang 215600,China)

机构地区：[1]江苏科技大学机械工程学院,江苏镇江212003 [2]中科探海(苏州)海洋科技有限责任公司,江苏张家港215600

出　　处：《控制工程》2023年第1期177-185,共9页Control Engineering of China

基　　金：国家自然科学基金资助项目(61105071);张家港市产学研预研资金项目(ZKCXY2253)。

摘　　要：为了提高多机器人协作围捕效率,提出了基于平行导引律的多机器人免疫网络协作围捕算法。首先,将围捕者的运动策略作为抗体,逃逸者和目标区域的位置信息作为抗原,通过抗体和抗原的刺激和抑制来构建协作围捕免疫网络;然后,基于平行导引律定义了平行导引律调节因子;最后,通过抗体浓度自适应选择来完成围捕任务。数值测试表明,与其他算法相比,免疫围捕算法不仅保证了围捕成功率,而且所需步数平均减少21%,转角平均减少13%;而仿真平台的测试对比结果表明,围捕算法所需时间平均减少7%,转角平均减少24%,从而验证了基于平行导引律的多机器人免疫围捕算法的有效性。In order to improve the multi-robot cooperative hunting efficiency,an immune network cooperative hunting algorithm for multi-robots based on parallel guidance law is proposed.Firstly,the movement strategy of the catcher was used as antibody,and the location information of the escapee and target area was used as antigen,and the location information between the escaper and the target area as antigen,an immune network for cooperative hunting is constructed through the stimulation and suppression between the antigen and antibody.Then,a parallel guidance law adjustment factor is defined based on the parallel guidance law.Finally,the hunting task is completed by adaptively selecting the antibody concentration.Numerical test results show that,compared with other immune algorithms,the proposed immune network cooperative hunting algorithm guarantees the hunting success rate,and reduces the hunting steps by an average of 21% and the rotation angle by 13%.The test comparison results based on the simulation platform show that the hunting time of the proposed hunting algorithm is reduced by an average of 7%,and the turning angle is reduced by an average of 24%,which verifies the effectiveness of the multi-robot immune hunting algorithm based on the parallel guidance law.

关 键 词：多机器人 协作围捕 免疫网络算法 抗体浓度 平行导引律 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]