Improving mobility using exosuits - a feasibility study on patients with incomplete spinal cord injury
Final Report Abstract
The ten months internship was a great success regarding the direct but also indirect research and personal development objectives. For the first phase it was planned to evaluate a soft exoskeleton, the Myosuit, in sit to stand transitions and level walking. After these evaluations with healthy users, we planned to test assistance with incomplete spinal cord injury patients. After revising the textile interface and developing a novel anti-gravity control, we successfully performed Myosuit experiments in sit to stand and stand to sit transitions. We were able to assist hip and knee extension with up to 26% of the natural knee moment and up to 35% of the knee power. To be able to transfer the anti-gravity control approach to walking, questions had to be solved regarding stance and swing detection on level ground and slopes but also regarding the passive elastic ligaments that should assist hip and knee flexion. Further, the Myosuit actuation unit had to be revised to improve assistance efficiency and mechanical robustness. All steps were required to ensure a secure execution of walking experiments with healthy users and patients. Regarding stance and swing detection we recorded a large walking reference dataset. Based on this data we extracted a fundamental concept for stance and swing detection based on real or virtual lower limb segment angular velocities. We found that multiple segments are able to reliably detect stance with our approach but further constraints are required to detect swing. To evaluate the passive ligaments, a wearable testbed with exchangeable rest length and stiffness was developed. Experiments are currently performed to investigate how such ligaments influence human walking kinematics and kinetics and if they are able to support push off. The new actuation unit is currently updated in cooperation with different companies. The Updated controller, the ligament structure, the sensor setup, the textile interface and the actuation unit will be combined in spring 2018 to perform first level walking studies on patients. As I had the chance to contribute to all areas of the Myosuit development, I improved my knowledge for many technical aspects. Further I increased my knowledge regarding patient requirements and product development due to the steps initiated by my colleagues before and after founding of the MyoSwiss AG and pre-experiments with different patient populations. I led the control developments and its evaluations in the project team and believe that the developed concepts might still be not perfect (as not tested in patients) but they provide a fundamental basement for a very robust and reliable control for the Myosuit users. After finishing my internship at ETH, it is planned to further investigate on lightweight exoskeletons for elderly and mobility impaired. Questions should be answered regarding the optimization of assistance patterns and actuation principles for different user groups, environments and movement tasks.
Publications
- The myosuit: Bi-articular anti-gravity exosuit that reduces hip extensor activity in sitting transfers. Frontiers in Neurorobotics, 11:57, 2017
K Schmidt, JE Duarte, M Grimmer, AS Puchades, H Wei, C Schmidt, and R Riener
(See online at https://doi.org/10.3389/fnbot.2017.00057)
