Policy Learning for Social Robot-Led Physiotherapy

Carl Bettosi, Lynne Baillie, Susan Shenkin, Marta Romeo · 2025

Read Policy Learning for Social Robot-Led Physiotherapy by Carl Bettosi, Lynne Baillie, Susan Shenkin, Marta Romeo (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-11297.txt.

Description

Carl Bettosi and three co-authors explore robot-led physiotherapy in this 2025 robotics paper. A friendly social robot could guide patients through exercise sessions alone. It must adapt instructions to each person's effort and progress. Real patient data for training are scarce.

The team recruited thirty-three expert healthcare practitioners. They acted as stand-in patients. Their interactions with the robot fed a patient behavior model. The model generates exercise performance measures. It also generates self-reported exertion scores. A reinforcement learning policy trained in simulation against this model. It learned to adjust instructions to each person's tolerance. It handled changing performance too. It also handled patients at different recovery stages. Their exercise plans differed as well.

Catalog record

CREATORCarl Bettosi, Lynne Baillie, Susan Shenkin, Marta Romeo
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE29 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-11297

Browse the full library · About the Tower of Babel