控制理论(社会学)
非线性系统
扰动(地质)
可逆矩阵
终端(电信)
终端滑动模式
滑模控制
计算机科学
控制工程
模式(计算机接口)
观察员(物理)
国家观察员
控制(管理)
工程类
数学
人工智能
物理
古生物学
纯数学
操作系统
生物
电信
量子力学
作者
Tangzhong Song,Lijin Fang,Yue Zhang
摘要
Abstract This paper delves into the problem of high‐precision tracking control for robotic systems, specifically addressing challenges posed by model uncertainties and external disturbances. To this end, a new nonsingular fast terminal sliding mode surface (NFTSM) with an unified structure is developed to circumvent singularity, without segmenting sliding manifold into multiple segments, which can improve steady‐state accuracy and reduce the complexity of the sliding surface. Following that, a Lyapunov stable controller characterized by a fast convergence law has been developed to stabilize the tracking errors to zero with bounded time. To address model uncertainties and external disturbances, a novel nonlinear disturbance observer (DO) is devised. The DO only has one adjusted parameter and can effectively estimate lumped disturbances to provide a feedforward compensation. Crucially, there is no requirement for upper bounds of disturbances and their derivatives during the design steps. Finally, the effectiveness and advantages of the proposed scheme are verified by extensive simulations and experiments conducted on a 7‐DOF robot manipulator. The results sufficiently illustrate that the proposed DO‐based sliding mode control scheme has high tracking accuracy, good robustness, and disturbance‐rejection abilities.
科研通智能强力驱动
Strongly Powered by AbleSci AI