Inverse Kinematics for Upper Limb Compound Movement Estimation in Exoskeleton-Assisted Rehabilitation

Robot-Assisted Rehabilitation (RAR) is important for treating patients with nervous system injuries. Accurately estimating the angles of the patient's limb joints during RAR is crucial for assessing their progress. The prevalent method approximates limb joint angles with the exoskeleton's, but this is inaccurate. Motion capture systems (MOCAPs) are more accurate but impractical. The Extended Inverse Kinematics Posture Estimation (EIKPE) method models the limb and Exoskeleton as differing parallel kinematic chains and has shown promise with single DOF movements. This paper assesses EIKPE with elbow-shoulder compound movements.

• 1
  EIKPE renders a good numerical approximation of the actual posture during compound movement execution, especially for the shoulder joint angles.
• 2
  EIKPE provides significantly better estimations than using Exoskeleton joint angles directly, particularly for GH joint angles.
• 3
  The RMSE errors of EIKPE are 50 to 60% lower than those of the Exoskeleton in estimating GH joint angles.