Admittance Control Scheme Comparison of EXO-UL8: A Dual-Arm Exoskeleton Robotic System

Abstract

In physical rehabilitation, exoskeleton assistive devices aim to restore lost motor functions of a patient suffering from neuromuscular or musculoskeletal disorders. These assistive devices are classified as operating in one of two modes: (1) passive mode, in which the exoskeleton passively moves its joints through the full range (or a subset) of the patient's motion during engagement, or (2) assist-as-needed (AAN) mode, in which the exoskeleton provides assistance to the joints of the patient, either by initiating the movements or assisting the patient's movements to complete the task at hand. Achieving high physical human-robot interaction (pHRI) transparency is an open problem for multiple degrees-of-freedom (DOFs) redundant exoskeletons. Using the EXO-UL8 exoskeleton, this study compares two multi-joint admittance control schemes (hyper parameter-based, and Kalman Filter-based) with comfort optimization to improve human-exoskeleton transparency. The control schemes were tested by three healthy subjects who completed reaching tasks while assisted by the exoskeleton. Kinematic information in both joint and task space, as well as force-and torque-based power exchange between the human arm and exoskeleton, are collected and analyzed. The results show that the preliminary Kalman Filter-based control scheme matches the performance of the existing hyper parameter-based scheme, highlighting the potential of the Kalman Filter-based approach for additional performance.

A Parallel, 2-DOF Exoskeleton for the Human Shoulder: Device Characterization and Preliminary Results on Healthy Subjects

“…It is important to note that the exoskeleton lacks a controller that involves human input in the activation of the exoskeleton's functions. Admittance control has proven to be effective in predicting the onset of user motion and has been implemented in a number of exoskeletons (37,38). Continuum-based actuators impart forces on a large surface area, making it impractical to measure forces through conventional sensors.…”

Wearable upper limb robotics for pervasive health: a review

“…Both EE type robotic devices and exoskeletons have been used in randomized controlled trials (RCT) for subacute and chronic stroke rehabilitation. Examples of commercialized rehabilitation exoskeletons include the Armeo Spring, Armeo Power, and Myomo and those of research ones include Harmony [108], NESM [107], HEXO [162], NTUH-II [163], Aalborg University Exoskeleton [106], ALEx [164], EXO-UL8 [165], FELXO-Arm1 [166], CleverARM [167], etc, with typical ones shown in figure 2. In addition, concomitant therapies have also been adopted to enhance rehabilitation including virtual reality [168] and conventional stroke therapies.…”

Review on Patient-Cooperative Control Strategies for Upper-Limb Rehabilitation Exoskeletons

“…The EXO-UL8 is a dual-arm exoskeleton that covers all the main movements of a human’s upper limb. The robot supports the motion of the shoulder, elbow, and wrist through seven non-backdrivable joints, and an additional joint operates the handgrip Shen et al (2019) . In a previous version, the EXO-UL7 was actuated through cable-driven actuation mechanisms.…”

“…Precisely, torques applied by the human to the exoskeleton joints are estimated from the F/T sensors, then, through an admittance model, reference trajectories are generated and operated by the inner low-level PID control loops. Friction and gravity compensation is added as feedforward terms to the low-level controller Shen et al (2019) . The core concept of the EXO-UL8 controller is to generate motion in response to human-applied forces to improve backdrivability and reduce the user-perceived weight of the robot.…”