More independence for disabled young people
Case study
Developments in artificial intelligence powered technologies and sensor-based machine learning are growing in their technical capabilities. UK researchers are applying these innovations to wheelchair design, offering more disabled people hope of independent mobility.
 |
David Sanders, Professor of Systems Engineering, University of Portsmouth, UK |
| Literati award won | Outstanding Paper 2021 |
| Winning research | Simple rules to modify pre-planned paths and improve gross robot motions associated with pick & place assembly tasks |
The problem
About 10% of people with a disability need a wheelchair, but conventional options are unable to cater to people with certain types of physical, mental or spatial disabilities. Traditional wheelchair research assumes a certain level of user cognitive ability and doesn’t fully consider those with complex needs. Intellectual barriers to independent mobility need to be better considered to find mobility solutions that work for more people.
According to the World Health Organization, mobility is essential for better health and improved quality of life. But how can the individual needs of wheelchair users be better accommodated? And how can independent mobility become a possibility for those who don’t have it?
The overall impact has been gratifying to see. New intelligent and user-friendly navigation, communication and control systems for powered wheelchairs have made a significant and positive impact on the lives of users
The research
When Professor David Sanders, Professor of Systems Engineering at the University of Portsmouth drafted his thesis on machine learning in 2011, the techniques were applied to industrial robots undertaking pick and place tasks in manufacturing lines. The basic idea was that machines could learn about themselves and improve their behaviour. The robots would learn and then use simple rules to improve their movements and make themselves faster. That original work presented real time methods that allowed robots to continue working while they automatically calculated more efficient paths for future movements. The concept that machines could learn about themselves and improve their behaviour was applied to mobile robots and tele-operated robots. From there, Professor Sanders together with his co-author Dr Giles Tewkesbury applied the idea to powered wheelchairs in collaboration with the Chailey Heritage Foundation and Sussex Community NHS Foundation Trust.
In the decade that has since passed, artificial intelligence (AI) capabilities have rapidly advanced, and the software initially designed for the factory floor has the capability to transform the lives of thousands of wheelchair users daily. AI technology gives the wheelchair user control, and sensor technology informs obstacle avoidance systems to minimise accidents. There is room for constant innovation and new ideas and applications continue to be researched.
Professor Sanders and Dr Tewkesbury are now investigating the use of artificial intelligence and engineering principles to bring self-mobility to people previously considered too intellectually impaired. This is the first time that has been attempted.
The impact towards healthier lives
The technology has helped more than 1,500 children and young people with complex physical disabilities gain independent mobility, in addition to people with multiple sclerosis, arthritis, stroke, paraplegia, orthopaedic impairment, cerebral palsy and those with diabetes related blindness and limb loss. The sophistication of the technology has made it possible for many of these people to use powered wheelchairs for the very first time.
Professor Sanders’s work has been formally recognised and was awarded the IET International Innovation Awards in 2020 for both ‘Outstanding Innovation in Digital Health and Social Care’ and ‘Excellence in Creating a Smarter World’. Building on this work, Professor Sanders and Dr Tewkesbury will conduct further sensor research and design shared control systems and AI. The aim is for AI algorithms to predict user patterns and assess user ability to drive, then blend user signals with intelligent sensors and environment monitoring. This will provide safe mobility with clear cause and effect. In time, generic systems will automatically detect and identify a user, as well as adjust themselves to match their functionality and capabilities. This new technology will allow users to share mobility systems, making the technology more accessible and economically viable.
The potential impact of these innovations extends beyond the wheelchair. The AI control and obstacle avoidance systems are also being explored in defence and automotive industries, and it is anticipated that it will be possible to apply the technology to the fields of tracking, driving, robotics, and navigation.
The background story
Listen to Professor David Sanders at the University of Portsmouth explain why his award winning research into improved digital healthcare has been an enormous team effort over the last 25 years and why making a difference is so important to the work they do.