An innovative robotic walker for mobility assistance and lower limbs rehabilitation

Abstract

Robotic walkers are assistive robotic devices that provide mobility assistance to frail or elderly users, also providing a significant potential for lower limb rehabilitation. This paper presents the ISR-AIWALKER robotic walker, an instrumented robotic platform that has been developed to conduct research on rehabilitation and assistive robotics. The contributions of this work are: (i) an intuitive and safe HMI, in combination with (ii) a robot-assisted navigation module for safe mobility assistance, (iii) and a system to analyze gait patterns as a potential diagnostic tool for rehabilitation purposes. In our work, we introduce a novel vision-based HMI boasting hand gripping pattern analysis through machine learning (ML) techniques, to ensure the user is adequately gripping the walker's handles. Navigation wise, we developed a method using utility theory and a rapidly exploring random tree technique to find the safest path according to the user's intent. On the user's gait analysis side, using computer vision techniques and ML algorithms, we model the user's lower limbs' kinematics, providing a motion capture tool and also an automatic system to identify changes in the gait pattern. Experiments with volunteers evidenced satisfactory results when evaluating the proposed modules.