Al-Dujailii Ayad Q, Hasan Alaq F, Humaidi Amjad J, Al-Jodah Ammar
Electrical Engineering Technical College, Middle Technical University, Baghdad, 10022, Iraq.
Technical Engineering College, Middle Technical University, Baghdad, Iraq.
Heliyon. 2024 Mar 31;10(9):e28911. doi: 10.1016/j.heliyon.2024.e28911. eCollection 2024 May 15.
In this study, Active Disturbance Rejection Control (ADRC) has been designed for motion control of knee-joint based on exoskeleton medical robot. The extended state observer (ESO) is the main part of ADRC structure, which is responsible for estimating both actual states and system uncertainties. The proposed control scheme has adopted two versions of observers as disturbance estimators: linear extended state observer (LESO) and nonlinear extended state observer (NESO). The efficacy of proposed ADRC is strongly related to the performance of used ESO. As such, a comparison study has been conducted to evaluate the performance of two ADRCs in terms of disturbance-rejection capability and robustness to variation in system parameters under two version of ESO (LSO and NLESO). In order to enhance the dynamic performance of ADRC, Particle Swarm Optimization (PSO) algorithm has been used to optimally tune the design parameters of control scheme. To show the effectiveness of proposed LESO-based ADRC and NLESO-based ADRC, numerical simulation have been conducted. The proposed controllers have tested for an uncertain exoskeleton-knee system, where a 20% change in parameters was permitted over their nominal values. The results indicate that the ADRC algorithm based on LESO outperforms the one based on NESO in terms of disturbances rejection ability and robustness to parameters' variations.
在本研究中，基于外骨骼医疗机器人设计了用于膝关节运动控制的自抗扰控制（ADRC）。扩展状态观测器（ESO）是ADRC结构的主要部分，负责估计实际状态和系统不确定性。所提出的控制方案采用了两种观测器作为干扰估计器：线性扩展状态观测器（LESO）和非线性扩展状态观测器（NESO）。所提出的ADRC的有效性与所用ESO的性能密切相关。因此，进行了一项比较研究，以评估在两种ESO（LSO和NLESO）版本下，两种ADRC在干扰抑制能力和对系统参数变化的鲁棒性方面的性能。为了提高ADRC的动态性能，采用粒子群优化（PSO）算法对控制方案的设计参数进行了优化调整。为了验证所提出的基于LESO的ADRC和基于NLESO的ADRC的有效性，进行了数值模拟。所提出的控制器针对一个不确定的外骨骼膝关节系统进行了测试，该系统允许参数在其标称值的基础上有20%的变化。结果表明，基于LESO的ADRC算法在干扰抑制能力和对参数变化的鲁棒性方面优于基于NESO的算法。
