Abstract
To assist home care for the elderly, we've finished a smart walker in previous study, with the features including destination track navigation, obstacle detection, self-positioning, follow-up, and wireless inductive charging. In this study, we further finish the user's intention detection function on the smart walker to control the smart walker. We install 3 pressure sensors on both sides of smart walker handles respectively to extract the user's force application, which are taken as input signal. Through AdaBoost classifier, it finishes input signal recognition and controls of the smart walker moving forward/backward directly. The user doesn't need to press the switch or click on the panel, but apply force on the handle directly, which makes the operation easier. The experiment validates that the accuracy can achieve higher than 98%, which can simplify the operation effectively and achieve the practical purpose.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
