Hand Gesture Controlled Smart Wheelchair with GPS Tracking and Deep Learning Based Object Detection for Collision Avoidance

As per the World Health Organization, 15% of the total world’s population faces some form of physical impairment. Mobility impairments are one of the most common physical handicaps we have, and wheelchairs are the most commonly used assistive devices for improving different levels of mobility impairments. However, wheelchairs vary in their functionality. Therefore, specific cohorts of users such as individuals with muscular dystrophy and partial paralysis may find it difficult or impossible to operate wheelchairs with manual or powered functionalities, which tend to be the most common. A smart wheelchair is can provide an accessible solution for this cohort of users where it can adapt to the needs of the user using IoT. In this paper, we present a hand gesture-controlled smart wheelchair with gyroscopic stabilization. Furthermore, we introduce collision avoidance functionality for the wheelchair with the help of Deep Learning based object detection methods and an ultrasonic sensor. Apart from these, the proposed wheelchair also provides GPS tracking features.