基于T细胞效应的多机器鱼环境免疫协作探测  

Immune collaboration environmental detection for multi-robot fish based on T cell effect

作　　者：江亚峰[1,2,3] 王舜 张亮 刘维[4] 袁明新 JIANG Ya-feng;WANG Shun;ZHANG Liang;LIU Wei;YUAN Ming-xin(School of Mechanical and Power Engineering,Jiangsu University of Science and Technology,Zhangjiagang 215600,China;School of Mechanical&Power Engineering,Suzhou Institute of Technology,Jiangsu University of Science and Technology,Zhangjiagang 215600,China;Zhangjiagang Industrial Technology Research Institute,Jiangsu University of Science and Technology,Zhangjiagang 215600,China;T-SEA Marine Technology Co.,Ltd.,Zhangjiagang 215600,China)

机构地区：[1]江苏科技大学机电与动力工程学院,江苏张家港215600 [2]江苏科技大学苏州理工学院机电与动力工程学院,江苏张家港215600 [3]张家港江苏科技大学产业技术研究院,江苏张家港215600 [4]中科探海(苏州)海洋科技有限责任公司,江苏张家港215600

出　　处：《舰船科学技术》2025年第4期154-162,共9页Ship Science and Technology

基　　金：工信部高技术船舶项目([2019]360);张家港市产业链创新产品攻关计划(ZKC2206);张家港市产学研预研资金(ZKYY2253,ZKYY2328)。

摘　　要：为提高多机器鱼在未知复杂水域中的探测质量和效率,提出了基于T细胞效应的环境免疫协作探测模型和算法。首先基于Jerne的独特型免疫网络假设构建多机器鱼基本免疫协作探测网络模型;然后根据已探测区域的环境信息,借鉴生物T细胞效应,构建多机器鱼的互联耦合免疫协作探测网络模型;最后进行算法设计和实验测试。实验结果表明,与基于免疫机理的完全探测算法和基于生物熵的免疫协作探测算法相比,文中算法的覆盖率平均提高了7.6%,重复探测率平均降低了32.8%,平均探测步数平均降低了11.6%,平均重复探测步数平均降低了24.7%,验证了未知复杂环境中多机器鱼互联耦合免疫协作探测网络模型的有效性。To enhance the detection quality and efficiency of underwater multi-robot fish in unknown complex environ-ments.A model and algorithm for immune collaborative environment detection based on T cell effector function are pro-posed.Firstly,a basic immune collaborative detection network model for multi-robot fish is constructed based on Jerne's idiotype immune network hypothesis.Then,an interconnected coupled immune detection network model for multi-robot fish is constructed on the basis of biological T cell effect.Finally,algorithm design and experimental testing are conducted.The experimental results indicate that,compared to the fully detection algorithm based on immune mechanism and the immune cooperative detection algorithm based on biological entropy,the algorithm proposed in this paper has an average increase in the coverage rate of 7.6%,a decrease in the average repeated detection rate of 32.8%,a decrease in the average detection steps of 11.6%,and a decrease in the average repeated detection steps of 24.7%.This validates the effectiveness of the im-mune collaborative detection network model of multi-robot fish in unknown complex environments.

关 键 词：多机器鱼 环境探测 免疫网络模型 独特型免疫网络假设 T细胞效应 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置] TP391.41[自动化与计算机技术—控制科学与工程]