信任驱动的自适应协调控制算法及应用  

Application of Trust Driven Adaptive Cooperative Control Algorithm

作　　者：桂欣冬 吉鸿江 范玲玲 刘世达 Gui Xindong;Ji Hongjiang;Fan Lingling;Liu Shida(School of Automation,Beijing Information Science&Technology University,Beijing 100192,China;China Shipbuilding IT Co.,Ltd.,Beijing 100861,China;School of Electrical and Control Engineering,North China University of Technology,Beijing 100144,China)

机构地区：[1]北京信息科技大学自动化学院,北京100192 [2]北京中船信息科技有限公司,北京100861 [3]北方工业大学电气与控制工程学院,北京100144

出　　处：《系统仿真学报》2021年第8期1809-1817,共9页Journal of System Simulation

基　　金：国家自然科学基金(61803036,61703019);北京信息科技大学师资补充与支持计划(5111911129);北京市教委科研计划一般项目(KM201911232015)。

摘　　要：多机械臂执行协同任务时,由于作业环境、装配条件、系统扰动等诸多非线性因素影响,其末端位置难以精确跟踪目标并保证一致,导致协同工作时产生较大误差。为解决多机械臂位置协同控制问题,通过研究新颖的信任机制,包括信任值(自信任值和互信任值)自适应更新机制和Fisher信息权重(协方差)更新机制,提出一种基于多智能体的自适应位置协调控制算法,当系统存在干扰时,实现多机械臂末端位置一致跟踪到目标位置,提升自动化协同装配任务的控制精度和执行效率。仿真结果验证了所设计方法的有效性。When a multi-manipulator performs cooperative task,it’s end position is difficult to accurately track the target and ensure the consistency influenced by the nonlinear factors such as working environment,assembly condition and system disturbances,resulting in large errors during collaborative work.To solve multi-manipulators position coordinated control problem,a multi-agent based adaptive position coordination control algorithm is proposed by researching on the novel trust mechanism,including adaptive update mechanism of trust value(self-trust and mutual-trust)and Fisher information weighted(covariance)update mechanism.In automated collaborative assembly tasks,the precise consensus positioning of end-effector is achieved with improvements in the control accuracy and efficiency.The simulation results verify the effectiveness of the proposed method.

关 键 词：信任驱动 多智能体系统 自适应协调控制 多机械臂协同 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]