Deep Learning architecture to assist with steering a powered wheelchair

Malik Haddad; David Sanders

doi:10.1109/TNSRE.2020.3031468

Deep Learning architecture to assist with steering a powered wheelchair

Malik Haddad, David Sanders

Research output: Contribution to journal › Article › peer-review

427 Downloads (Pure)

Abstract

This paper describes a novel Deep Learning architecture to assist with steering a powered wheelchair. A rule-based approach is utilized to train and test a Long Short Term Memory (LSTM) Neural Network. It is the first time a LSTM has been used for steering a powered wheelchair. A disabled driver uses a joystick to provide desired speed and direction, and the Neural Network provides a safe direction for the wheelchair. Results from the Neural Network are mixed with desired speed and direction to avoid obstacles. Inputs originate from a joystick and from three ultrasonic transducers attached to the chair. The resultant course is a blend of desired directions and directions that steer the chair to avoid collision. A rule-based approach is used to create a training and test set for the Neural Network system and applies deep learning to predict a safe route for a wheelchair. The user can over-ride the new system if necessary.

Original language	English
Article number	9225000
Pages (from-to)	2987-2994
Number of pages	7
Journal	IEEE Transactions on Neural Systems and Rehabilitation Engineering
Volume	28
Issue number	12
Early online date	15 Oct 2020
DOIs	https://doi.org/10.1109/TNSRE.2020.3031468
Publication status	Published - 1 Dec 2020

Keywords

Deep Learning
Wheelchair
Neural Network
Rule-based
Disabled
Steer
RCUK
EPSRC
EP/S005927/1

Access to Document

10.1109/TNSRE.2020.3031468

Deep Learning Architecture
© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Accepted author manuscript (Post-print), 890 KB

Dataset for 'Deep Learning Architecture to Assist with Steering a Powered Wheelchair'.
Haddad, M. (Creator), University of Portsmouth, 18 Nov 2020
DOI: 10.17029/432fa5e2-eff4-47f6-99a9-df5123cd8bee
Dataset

File

Using artificial intelligence to share control of a powered-wheelchair between a wheelchair user and an intelligent sensor system
Sanders, D. (PI), Gegov, A. (CoI) & Haddad, M. (Team Member)
Engineering and Physical Sciences Research Council
1/11/18 → 1/11/22
Project: Research

File

Top two per cent of researchers in the world
Sanders, D. (Recipient), 2 Nov 0001
Prize: National/international honour

Cite this

@article{76f3e5e385a64134a0895afd33495ac3,

title = "Deep Learning architecture to assist with steering a powered wheelchair",

abstract = "This paper describes a novel Deep Learning architecture to assist with steering a powered wheelchair. A rule-based approach is utilized to train and test a Long Short Term Memory (LSTM) Neural Network. It is the first time a LSTM has been used for steering a powered wheelchair. A disabled driver uses a joystick to provide desired speed and direction, and the Neural Network provides a safe direction for the wheelchair. Results from the Neural Network are mixed with desired speed and direction to avoid obstacles. Inputs originate from a joystick and from three ultrasonic transducers attached to the chair. The resultant course is a blend of desired directions and directions that steer the chair to avoid collision. A rule-based approach is used to create a training and test set for the Neural Network system and applies deep learning to predict a safe route for a wheelchair. The user can over-ride the new system if necessary.",

keywords = "Deep Learning, Wheelchair, Neural Network, Rule-based, Disabled, Steer, RCUK, EPSRC, EP/S005927/1",

author = "Malik Haddad and David Sanders",

year = "2020",

month = dec,

day = "1",

doi = "10.1109/TNSRE.2020.3031468",

language = "English",

volume = "28",

pages = "2987--2994",

journal = "IEEE Transactions on Neural Systems and Rehabilitation Engineering",

issn = "1534-4320",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "12",

}

TY - JOUR

T1 - Deep Learning architecture to assist with steering a powered wheelchair

AU - Haddad, Malik

AU - Sanders, David

PY - 2020/12/1

Y1 - 2020/12/1

N2 - This paper describes a novel Deep Learning architecture to assist with steering a powered wheelchair. A rule-based approach is utilized to train and test a Long Short Term Memory (LSTM) Neural Network. It is the first time a LSTM has been used for steering a powered wheelchair. A disabled driver uses a joystick to provide desired speed and direction, and the Neural Network provides a safe direction for the wheelchair. Results from the Neural Network are mixed with desired speed and direction to avoid obstacles. Inputs originate from a joystick and from three ultrasonic transducers attached to the chair. The resultant course is a blend of desired directions and directions that steer the chair to avoid collision. A rule-based approach is used to create a training and test set for the Neural Network system and applies deep learning to predict a safe route for a wheelchair. The user can over-ride the new system if necessary.

AB - This paper describes a novel Deep Learning architecture to assist with steering a powered wheelchair. A rule-based approach is utilized to train and test a Long Short Term Memory (LSTM) Neural Network. It is the first time a LSTM has been used for steering a powered wheelchair. A disabled driver uses a joystick to provide desired speed and direction, and the Neural Network provides a safe direction for the wheelchair. Results from the Neural Network are mixed with desired speed and direction to avoid obstacles. Inputs originate from a joystick and from three ultrasonic transducers attached to the chair. The resultant course is a blend of desired directions and directions that steer the chair to avoid collision. A rule-based approach is used to create a training and test set for the Neural Network system and applies deep learning to predict a safe route for a wheelchair. The user can over-ride the new system if necessary.

KW - Deep Learning

KW - Wheelchair

KW - Neural Network

KW - Rule-based

KW - Disabled

KW - Steer

KW - RCUK

KW - EPSRC

KW - EP/S005927/1

U2 - 10.1109/TNSRE.2020.3031468

DO - 10.1109/TNSRE.2020.3031468

M3 - Article

SN - 1534-4320

VL - 28

SP - 2987

EP - 2994

JO - IEEE Transactions on Neural Systems and Rehabilitation Engineering

JF - IEEE Transactions on Neural Systems and Rehabilitation Engineering

IS - 12

M1 - 9225000

ER -

Deep Learning architecture to assist with steering a powered wheelchair

Abstract

Keywords

Access to Document

Fingerprint

Datasets

Dataset for 'Deep Learning Architecture to Assist with Steering a Powered Wheelchair'.

Projects

Using artificial intelligence to share control of a powered-wheelchair between a wheelchair user and an intelligent sensor system

Prizes

Top two per cent of researchers in the world

Cite this