Predicting Movement and Balance to Prevent Falls

Dr. Kennedy and team have developed a wearable sensor approach that can predict a person’s walking path and how their body will move—especially balance and torso sway—in real time. Using data from body-worn sensors and advanced modeling, their work shows it is possible to anticipate how a person will navigate their environment and how their stability may change along the way. This enables early identification of potential trip hazards and moments of instability before a fall occurs.

Building on this foundation, the team’s research also incorporates scene-aware trajectory prediction, allowing movement to be understood in the context of the surrounding environment. Together, these advances support the development of real-time fall warning systems, as well as next-generation assistive technologies such as responsive lower-limb exoskeletons. This work has important implications for improving mobility, safety, and independence, particularly for older adults and individuals at risk of falls.

Publication: W. Wang, M. Raitor, S. Collins, C. K. Liu and M. Kennedy, “Trajectory and Sway Prediction Towards Fall Prevention,” 2023 IEEE International Conference on Robotics and Automation (ICRA), London, United Kingdom, 2023, pp. 10483-10489, doi: 10.1109/ICRA48891.2023.10161361., https://ieeexplore.ieee.org/document/10161361

(Preprint) EgoNav: Egocentric Scene-aware Human Trajectory Prediction: https://arxiv.org/abs/2403.19026