Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
Assessing the Joint Misalignment Effects in Rehabilitation Robotics: A Case Study
Submission Author:
jonathan campo , SP
Co-Authors:
jonathan campo, Gabriel Patti Sanches Coelho, Adriano Siqueira
Presenter: Gabriel Patti Sanches Coelho
doi://10.26678/ABCM.COBEM2023.COB2023-1784
Abstract
Context: Designing safe control strategies for robotic rehabilitation requires addressing challenges such as modeling human behavior, ensuring safety operation, robustness and adaptability. A common assumption in these control strategies is that robot and human limbs share the same movement and joint rotation during assistance, enabling a unified and simplified model for system modeling. However, human joint kinematics are relatively complex to emulate precisely by robotic devices; this is due to body segment variability, ligaments and tendons stretching, and inherent migration of joint centers. Therefore, understanding the physical human-robot interaction (HRI) turns out critical when dealing with undesired forces produced by misalignment between the robot and the wearer’s body, which must be minimal to prevent injury, discomfort and transmission looses of the robot-assisted joint torques. GAP: Even though proper force transducers can be positioned at the user-robot interface, quantifying joint misalignment remains difficult. An essential consideration for sensors in HRI systems is that high joint misalignment during motion might shift the contact area where the interaction force is measured, producing variability and unpatterned behavior. Consequently, an HRI force sensor must be validated with joint misalignment information to provide reliable measurements. Purpose: The objective of this work is to identify the misalignment of a human-robot interaction system for robotic rehabilitation purposes and study the effects of misalignment and interaction forces in different possible misalignment scenarios. Experiments were conducted using a robotic knee exoskeleton for walking trials with various misalignment configurations. Custom 3D-printed embedded HRI sensors, based on Force Sensing Resistors (FSR), were developed and placed at different contact points at the interface of the human leg and the knee exoskeleton to quantify HRI forces and identify misalignment levels. Results and Conclusions: The results showed a strong relation between misalignment level, the walking speed and the robot device slippage due to gravity and natural user movements during the physical HRI. It was observed that the duration of phases in the gait cycle was significantly affected by misalignment. These outcomes have implications for the design of modeling-based controllers. Inaccuracies in system modeling could result from a misunderstanding of misalignment, potentially impacting the performance of such controllers. By identifying and studying the effects of misalignment in HRI systems for rehabilitation purposes, safety, effectiveness, and user comfort can be enhanced, ultimately improving the rehabilitation experience and controller performance.
Keywords
Rehabilitation Robotics, Human-robot interaction (HRI), Joint misalignment
