摘要
针对基座浮动刚性空间机器人的轨迹跟踪控制问题,基于自适应神经网络、性能函数以及障碍函数提出了一种新的全局自适应控制策略。首先,对于机器人系统任意初始状态,所设计的控制器使得轨迹跟踪误差在预定义时间内收敛,并始终约束在给定的性能函数边界范围内,从而保证期望瞬态与稳态性能;然后,基于单参数更新技术设计自适应神经网络逼近空间机器人动力学模型不确定性以减小计算负担;最后,通过Lyapunov稳定性分析判据证明系统所有状态的一致有界性。数值仿真结果表明:所提方法不仅能使得空间机器人的轨迹跟踪精度达到期望性能约束要求,还能保证被控系统的全局稳定性。
Targeting the trajectory tracking control issue of a base-floating rigid space robot,a novel global adaptive control strategy was proposed based on an adaptive neural network,a performance function,and a barrier function.First,for any initial state of the space robot system,the designed controller made the trajectory tracking error converging in a predefined time and was always constrained to be within the boundaries of the given performance function,so as to guarantee the desired transient and steady state performances.Then,the adaptive neural network was designed based on a single-parameter updating technique to approximate the uncertainty of the space robot dynamics model,thus reducing the computational burden.Finally,the consistent boundedness of all states of the system was proved by Lyapunov stability analysis criterion.Numerical simulation results showed that the proposed method not only made the trajectory tracking accuracy of the space robot reaching the desired performance constraints requirements,but also ensured the global stability of the controlled system.
作者
刘程果
李俊阳
LIU Chengguo;LI Junyang(College of Mechanical and Vehicle Engineering,Chongqing University,Chongqing 400044,China;State Key Laboratory of Mechanical Transmission for Advanced Equipment,Chongqing University,Chongqing 400044,China)
出处
《载人航天》
CSCD
北大核心
2024年第6期805-814,共10页
Manned Spaceflight
基金
国家自然科学基金面上项目(52375041)
中央高校基本科研业务费项目(2023CDJXY-021)
航空科学基金项目(202000020Q9001)。
关键词
空间机器人
自适应控制
全局稳定性
预设性能
space robot
adaptive control
global stability
prescribed performance
