Classification of reaching and gripping gestures for safety on walking aids

Abstract

This paper proposes a solution to increase safety when using robotic walking aids. Walking aids are an important tool as long as they can be safely operated, but elderly people often discard them due to the fear of falling. The authors propose a safety system that locks the walker and giving the appropriate feedback when the user is inadequately gripping the walker's grips. It is based on a low-cost optical hand tracker that allows the system to perceive how safely and efficiently a user reaches and grips the handles of the walking aid. This is an important requirement, especially on those scenarios where the walker operation requires bodyweight support on the upper limbs. Adequate interaction with the walker reduces the risk of falling and gives some extra confidence to the user. Experimental results with 10 volunteers provided strong evidence that the proposed system is able to distinguish between correct and incorrect grips. The proposed experimental setup can be easily integrated in different contexts, making it unconstrained to the physical setup.