Development and Evaluation of a Sensor Glove to Detect Grasp Intention for a Wearable Robotic Hand Exoskeleton
Restoring hand function in people suffering from neuromotor hand impairment is a crucial step towards regaining independence. Wearable robotic hand orthoses are a promising approach to support that aim by providing grasp assistance in daily life. For successful independent use of such an assistive device, a robust and intuitive method to detect the user's intention to grasp is crucial. However, current solutions often fail to meet these two requirements. In this work, we present a novel sensor glove to detect the user's grasp intention for the actuated hand exoskeleton RELab tenoexo by integrating soft flexible sensors into a conventional glove. Through contact detection with the object to grasp, the user input can be directly embedded into the regular movement. Sensor characterization and performance evaluation with three able-bodied subjects have shown that the developed sensor glove works reliably and can accurately detect the user's grasp intentions with true positive and true negative rates of 93.1% and 97.6%, respectively. A pilot test with a possible end-user with spinal cord injury (SCI) underlined the potential benefits of the approach and helped to identify aspects, that require further improvements. The study demonstrated the general feasibility of this method and paves the way for inherently intuitive intention detection in wearable assistive devices.
WOS:000636920600002
2020-01-01
978-1-7281-5907-2
New York
Proceedings of the IEEE RAS-EMBS International Conference on Biomedical Robotics and Biomechatronics
19
24
REVIEWED
Event name | Event place | Event date |
New York City, NY | Nov 29-Dec 01, 2020 | |