Automation of User Pace-Adjusted Treadmill: Comparison of Control Methods based on LRF Sensors

Neurological disorders affect gait and posture, leading to significant difficulties in daily life. Traditional gait assessment lacks objective and quantifiable data, hindering comprehensive evaluation. To address these limitations, interactive and objective systems, such as smart walkers and specialized treadmills, have been developed utilizing advanced technologies to gather detailed gait data. This study focuses on evaluating two control strategies (A controller based on adaptive estimation of gait parameters and a PI controller-based systems) to automate a conventional treadmill, utilizing LRF sensors and the user's walking speed for personalized system control. The results demonstrate high adaptability of the control models to gait variations, surpassing reported walking speeds in the literature using these algorithms. These findings are relevant for enhancing user interaction and experience in rehabilitation applications and serious games, which can enable improvements in rehabilitation and human-machine interaction. Our results show that the strategy employing adaptive filters exhibits superior responsiveness, enabling adaptation to different treadmill speeds with greater ease.