Scholarlink

<p>This article introduces an innovative robotic rehabilitation device designed for both right and left-hand wrists and forearms, featuring three active degrees of freedom (Pronation/Supination (P/S), Abduction/Adduction (AB/AD), and Flexion/Extension (F/E)) and two passive degrees of freedom (DOF). The device's unique design, incorporating two passive DOF, enhances adaptability to wrist rotation axes and ensures user safety by preventing lateral and shear forces. Its modular mechanism allows for seamless integration with other rehabilitation robots. Given the inherent complexities in the design and structure of robotic systems, including dynamic and parametric uncertainties and non-linear behaviors, this research explores an alternative approach for inverse kinematics (IK) analysis. The Takagi–Sugeno-Kang (TSK) method is utilized instead of traditional analytical and geometric techniques. To optimize the TSK fuzzy system hyperparameters, a novel hybrid approach is developed, combining Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) across three distinct scenarios. Simulation results reveal that the hybrid PSO-GA algorithm, which executes both PSO and GA twice per iteration in conjunction with the TSK fuzzy system, achieves superior performance. It attains the lowest Root Mean Square Error (RMSE) values of 0.1564, 0.1395, and 0.4117 degrees for P/S, AB/AD, and F/E, respectively, in IK analysis. In comparison, the GA-optimized TSK fuzzy system exhibits higher RMSE errors of 4.0087, 4.0264, and 6.1220 degrees for the same DOFs. This study highlights the effectiveness of the hybrid optimization strategy in enhancing the precision and adaptability of TSK fuzzy system parameters for joint coordinate space analysis in wrist and forearm rehabilitation robots.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Innovative Integration of Meta-Heuristic Algorithms with Adaptive TSK Fuzzy Systems for Inverse Kinematics in a New Wrist and Forearm Rehabilitation Exoskeleton

  • 期刊论文

  • 发表日期：

    2025年1月3日

查看全文
  • Behzad Saeedi,
  • Majid Sadedel,
  • Majid Mohammadi Moghaddam

摘要

This article introduces an innovative robotic rehabilitation device designed for both right and left-hand wrists and forearms, featuring three active degrees of freedom (Pronation/Supination (P/S), Abduction/Adduction (AB/AD), and Flexion/Extension (F/E)) and two passive degrees of freedom (DOF). The device's unique design, incorporating two passive DOF, enhances adaptability to wrist rotation axes and ensures user safety by preventing lateral and shear forces. Its modular mechanism allows for seamless integration with other rehabilitation robots. Given the inherent complexities in the design and structure of robotic systems, including dynamic and parametric uncertainties and non-linear behaviors, this research explores an alternative approach for inverse kinematics (IK) analysis. The Takagi–Sugeno-Kang (TSK) method is utilized instead of traditional analytical and geometric techniques. To optimize the TSK fuzzy system hyperparameters, a novel hybrid approach is developed, combining Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) across three distinct scenarios. Simulation results reveal that the hybrid PSO-GA algorithm, which executes both PSO and GA twice per iteration in conjunction with the TSK fuzzy system, achieves superior performance. It attains the lowest Root Mean Square Error (RMSE) values of 0.1564, 0.1395, and 0.4117 degrees for P/S, AB/AD, and F/E, respectively, in IK analysis. In comparison, the GA-optimized TSK fuzzy system exhibits higher RMSE errors of 4.0087, 4.0264, and 6.1220 degrees for the same DOFs. This study highlights the effectiveness of the hybrid optimization strategy in enhancing the precision and adaptability of TSK fuzzy system parameters for joint coordinate space analysis in wrist and forearm rehabilitation robots.