Intelligent Control of the Steering for a Powered Wheelchair Using a Microcomputer

Abstract

The research presented in this paper describes a new architecture for controlling powered wheelchairs. A Raspberry Pi microcomputer is considered to assist in controlling direction. A Raspberry Pi is introduced between user input switches and powered wheelchair motors to create an intelligent Human Machine Interface (HCI). An electronic circuit is designed that consists of an ultrasonic sensor array and a set of control relays. The sensors delivered information about obstructions in the surrounding environment of the wheelchair. Python programming language was used to create a program that digitized the user switches output and assessed information provided by the ultrasonic sensor array. The program was installed on a Raspberry Pi and the Raspberry Pi controlled power delivered to the motors. Tests were conducted and results showed that the new system successfully assisted a wheelchair user in avoiding obstacles. The new architecture can be used to intelligently interface any input device or sensor system to powered wheelchair.

One Shot Learning Approach to Identify Drivers

“…The authors are currently investigating other mathematically inexpensive AI techniques applied to powered mobility problems as part of a broader research to use artificial intelligence to share control of a powered-wheelchair between a wheelchair user and an intelligent sensor system [13].…”

“…Sanders et al [10,11] considered rule-based systems to provide steering routes for wheelchairs. Haddad et al [12][13][14] used ultrasonic sensor array readings as inputs to Multi-Criteria Decision Making (MCDM) deciders and mixed the suggested output from the deciders with user desired directions to provide collision free routes for wheelchairs. Haddad and Sanders [15] applied a MCDM method, PROMETHEE II, to suggest a safe route.…”

“…Haddad and Sanders [15] applied a MCDM method, PROMETHEE II, to suggest a safe route. Haddad et al [16,17] utilized microcomputers to create intelligent Human Machine Interfaces (HMI) used to safely steer powered wheelchairs. Haddad and Sanders [2,18] created a deep learning neural network to provide a safe steering direction for a powered wheelchair.…”

Control of a Semi-autonomous Powered Wheelchair

“…Many systems have been created to help with driving and control [1][2][3] and a few have moved from research and development prototypes to become successful and useful devices [4,5]. Semi-autonomous powered wheelchairs [6][7][8] have recently moved out of academic and research institutions [9][10][11] to become available to wheelchair users.…”

Intelligent User Interface to Control a Powered Wheelchair Using Infrared Sensors

“…People with disability often used powered mobility for daily activities [5] and researchers have aimed to improve powered mobility using: sensor systems [6], ultrasonics [7] Multiple Criteria Decision Making and ultrasonic sensors [8][9][10][11], Expert Systems [12,13], Rule Based Systems [14], microcomputers and Human Machine Interfaces [15,16], Deep Learning [2,17] and camera modules [18][19][20][21].…”