Researchers have developed a multi-sensor method to recognize the ambulation states of electric wheeled walker users, addressing a critical safety gap for the growing elderly population. By combining pressure sensors on the handles with a LiDAR sensor tracking lower-limb motion, the system classifies user states—normal walking (forward, left/right turn), falling (forward, left/right), and disengagement—and triggers adaptive control strategies accordingly.
Key takeaways
- Pressure sensors capture upper-limb force changes on the handles; denoising via Kalman filtering extracts motion intent with lower latency than moving-average filtering.
- LiDAR tracks leg positions using least-squares fitting to estimate lower-limb speed, enabling precise state classification.
- The fusion approach improves real-time monitoring accuracy and control stability, with potential to prevent falls and support clinical rehabilitation.
The work highlights the importance of considering user-specific factors over environmental ones in state recognition, and addresses challenges in low-speed turning and sudden speed changes. This integrated sensing strategy offers a practical path to enhance the safety and independence of elderly walker users.
Source: 《机器人》期刊 (robot.sia.cn) · Published 2026-03-25 · “轮式助行器使用者推行状态识别研究”