Human-centric Assessment of Rehabilitation Robots


Robotic rehabilitation exoskeletons are particularly valuable in therapy because they leverage robotic devices' unique potential for accurate and repeatable movements, and quantitative measurement in position and force domains. In addition to coordinated movement capabilities and functional workspace requirements such as range of motion (ROM) and torque required for ADL, a rehabilitation robot must possess quantitative measurement capabilities for evaluation, which requires high quality position sensing, good backdrivability, and backlash-free operation. Quantitative measurement capabilities and backdrivability, also termed device transparency, have been inferred from hardware-centric metrics such as inertia, static friction and viscous friction. These robotic device characterizations are useful for comparisons of the transparency of established devices, but do not quantify the effect the devices have on the kinematic movement properties. A direct comparison between movement measured by a robotic rehabilitation device and a no robot movement condition is required to validate a device as a reliable measurement tool. This comparison should take the form of generalizable characterization methods and metrics for motion-based device transparency assessments. Kinematic characteristics of coordinated movements are a prime target for investigations, since they can be measured in a non-invasive way, allowing for validation of an exoskeleton as a measurement device.

Currently, we are interested in understanding the suitability of robots for rehabilitation and as an assessment device. This research project is focused on evaluating the effect of wearing the robot on coordinated, multijoint movements in a human-centric manner. The RiceWrist, RiceWrist-S, MAHI Exo-II, and the READAPT exoskeletons, detailed further in Robot Assisted Rehabilitation - Spinal Cord Injury (SCI) ( and Design and Development of a Cybernetic Exoskeleton for Hand-Wrist Rehabilitation (  are undergoing testing during multi-joint movements to assess their effect on healthy human movement, and validate their use as measurement tools. Future work on this project will include the development of general metrics for human-centric assessment of rehabilitation robots.