Design And Implementation Of An Autonomous Electric Wheelchair Using Directions-Based Navigation And Smart Mobility Control

Abstract

This paper presents the development of a smart electric wheelchair system designed to assist individuals with mobility challenges by offering three control methods. The system includes a manual mode that allows users to control speed and direction via Bluetooth, a human-following mode that utilizes a camera for tracking and following a designated person, and an autonomous mode that enables navigation based on directional signboards. Convolutional Neural Networks (CNN) are employed for human detection, while cosine similarity is used to interpret movement commands from signboards. The primary goal is to empower physically disabled individuals by leveraging evolving technologies. The proposed system is cost-effective, easy to construct, and adaptable, and has been implemented and tested in real-time under all operating modes.