基于零空间行为融合的多智能体编队控制综述  

A review of multiagent formation control based on the null-space-based behavioral fusion algorithm

作　　者：葛宇 廖煜雷 王博[1,3] 李晔[1,3] 史长亭[4] 刘海波[4] 贾琪 GE Yu;LIAO Yulei;WANG Bo;LI Ye;SHI Changting;LIU Haibo;JIA Qi(National Key Laboratory of Autonomous Marine Vehicle Technology,Harbin Engineering University,Harbin 150001,China;Tianjin Navigation Instruments Research Institute,Tianjin 300131,China;Sanya Nanhai Innovation and Development Base,Harbin Engineering University,Sanya 572000,China;College of Computer Science and Technology,Harbin Engineering University,Harbin 150001,China)

机构地区：[1]哈尔滨工程大学智能海洋航行器技术全国重点实验室,黑龙江哈尔滨150001 [2]天津航海仪器研究所,天津300131 [3]哈尔滨工程大学三亚南海创新发展基地,海南三亚572000 [4]哈尔滨工程大学计算机科学与技术学院,黑龙江哈尔滨150001

出　　处：《哈尔滨工程大学学报》2024年第8期1442-1450,共9页Journal of Harbin Engineering University

基　　金：国家自然科学基金项目(52071097);海南省自然科学基金项目(522MS162);水下机器人技术重点实验室研究基金项目(2021JCJQ-SYSJJ-LB06910).

摘　　要：针对单智能体在民用和军用领域无法完成现实场景下复杂任务的问题,本文综述了零空间行为融合算法。阐述零空间行为融合算法的设计思想与数学模型,介绍零空间行为融合算法的文献计量分析。综述了零空间算法的研究进展,包括编队速度饱和、行为转换不连续、行为优先级分配和多种算法结合方面的改进工作,并分析当前研究存在的问题。研究表明:零空间行为融合算法在多智能体编队控制领域,提高算法的自适应性、考虑动力学特性、异构编队和行为函数优化是该算法未来的发展方向。Addressing the issue that a single intelligent agent cannot accomplish complex tasks during realistic scenarios in civil and military domains,this paper reviews the null-space-based behavioral(NSB)fusion algorithm.Initially,the design concept and mathematical model of the NSB fusion algorithm are described.Then,a bibliometric analysis of the algorithm is presented.The emphasis then shifts to the research progress of the algorithm,including improvements in formation-speed saturation,discontinuous behavioral transitions,behavioral priority allocation,and the combination of multiple algorithms.Finally,the paper analyzes issues in current research,highlighting that future development directions in multi-intelligent-agent formation control are the improvement of the adaptability of the algorithm,considering dynamics characteristics,addressing heterogeneous formations,and optimizing the behavior function.

关 键 词：零空间行为 编队控制 多智能体 逆运动学 基于行为 速度饱和 行为转换 行为优先级优化 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]