Control algorithm and evaluation of a powered wheeled walker for tracking a user

This study proposed a user-tracking control algorithm for a powered wheeled walker and examined how it improves the tracking performance of a user. Our newly developed powered wheeled walker based on the mechanism of infant walker and powered wheelchair has enhanced walking stability and needs less physical power to be moved. The walker is able to smoothly track user's various movements (i.e. forward, backward translation, and rotation) using pelvis motion sensors determining walking direction and stereo cameras analyzing lower extremity motion. In particular, weighting factors of controllers using pelvis motion sensor and stereo cameras can be suitably adjustable online. The experimental testing was conducted to examine walker's tracking performance, and the results from electromyography testing showed that the walker based on user-tracking control algorithm significantly decreased the magnitude of exerted muscle strength in both upper and lower extremity compared to conventional passive wheeled walker. Thus, we were able to successfully verify that the proposed walker can automatically track a user with gait impairment for safe walking without a helper.