Learning-based control approaches for service robots on cloth manipulation and dressing assistance: a comprehensive review.
Olivia Nocentini, Jaeseok Kim, Zain Muhammad Bashir and 1 others
PMID 36329473WHAT IT FOUND
Robots are learning to help with dressing through supervised learning, reinforcement learning, and imitation, but they remain confined to labs.
They struggle with unfamiliar clothes and cannot yet safely interact with people, so there is no clinical evidence to change practice.
Key findings
01Dressing assistance robots are currently tested in pre-defined laboratory conditions and cannot yet handle the variability of real-world human movement or unfamiliar clothing.
02Most studies use mannequins or simulations, meaning there is no feedback on the physical forces applied to a real person, which is a critical safety gap.
03There are currently no legal or safety regulations for social robots, preventing their unsupervised use in environments with people.
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
The review covers literature up to 2019, so it may not reflect the most recent rapid advances in AI. The included studies are largely engineering prototypes tested on mannequins or in simulations, not clinical trials with patients. The review does not assess the clinical effectiveness or safety of these devices for people with disabilities, only the technical success of the robot's actions.
Declared interests
The paper acknowledges support from non-U.S. government sources but does not list specific commercial conflicts of interest.
The easy way to misread this
Do not interpret the reported success rates of these robots as clinical evidence that they are safe or effective for patient care. Most results come from controlled lab tests on mannequins or simulations, and the authors explicitly state that legal and safety regulations for social robots do not yet exist.