Robust walking control of a lower limb rehabilitation exoskeleton coupled with a musculoskeletal model via deep reinforcement learning.
Shuzhen Luo, Ghaith Androwis, Sergei Adamovich and 3 others
PMID 36935514WHAT IT FOUND
A computer simulation showed a new exoskeleton controller can keep a virtual patient walking without falling, even when simulating paralysis or weakness.
No real patients were tested, so this does not yet change clinical practice.
Key findings
01The study is a simulation of a lower-limb exoskeleton controller, not a test on humans.
02In the simulation, the controller maintained balance and kept the center of pressure within the stable region for virtual patients with paralysis, weakness, or hemiparesis.
03The authors plan to test this controller on the physical hardware with real patients in the future.
STILL TO COME
How it was doneWhat they found
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
No real patients or healthy controls were tested; all results are from computer simulations. The exoskeleton hardware was not used to carry a human, so safety and comfort in the real world are unknown. The simulation used a simplified model of muscle weakness (halving force capacity) which may not match the complex spasticity or coordination deficits seen in actual stroke or spinal cord injury patients. The reference walking motion was manually created and had more knee flexion than natural human walking to increase stability.
Declared interests
The work was supported by the U.S. Government (Non-P.H.S.) and the National Institutes of Health (Extramural).
The easy way to misread this
Do not assume this exoskeleton is ready for clinical use or that it has been proven safe for patients. The paper reports only simulation results; the controller has not yet been deployed on the physical hardware with any human participants.