Challenges for assistive robots

Posted by Dr Praminda Caleb-Solly on 30 June 2017
Photo of robots encountering each other in a kitchen

Flexible technology for changing needs

Gracefully or disgracefully, we all age differently; experiencing the variable effects of growing older at different points in our lives. It could be our eyes, our knees and overall mobility or it could be our minds. Everybody is different and herein lies the major challenge of developing assistive technology; it needs to be flexible enough to cope with changing user needs related to the impact of the ageing process over time.

Responding to complex and changing end-user characteristics

Producing reliable, safe and robust assistive robots in not cheap, at least not at the moment. Yet, despite the well documented positive outcomes of assistive technology, the abandonment rate is particularly alarming with several studies reporting a 25 per cent-33 per cent non-use rate. The contributing factors are complex and range from inadequate consideration of the patient's own opinion to lack of thought about the environment in which the technology is used. For example, while a wheelchair can help a person move around more independently, the technology is impeded by environmental factors such as stairs or other uneven surfaces.

Older adults are not a homogenous group and we need to start being very specific about the clinical characteristics of the people for whom our assistive robots are being designed, and responsive to their changing needs. If we don't, we are in danger of developing technologies that do not adequately address specific physiological, physical and cognitive requirements and are likely to be abandoned by their user. . Working in multi-disciplinary design teams with experienced clinical experts is essential to the successful design, production and evaluation of assistive robotic solutions.

Assess and assist

Currently, an assessment of care needs is carried out by social services, care managers or occupational therapists. Sadly the decreasing number of carers is dramatically affecting the frequency of reassessment, but, this is an area where intelligent robotics systems may be able to help. With the addition of built-in sensors, assistive robots may also be able to conduct assessments of the patient's condition but it is important to understand how the changes may manifest themselves - slower response times, greater interaction errors, weaker grip strength or lower concentration levels?

A joint effort

If changes in condition are detected, the next decision is how should the system adapt, and how could the operation of the system continue to be safe? Who would evaluate its continued functional efficacy and certify its use? These questions point to the need for considering how the training needs of carers and occupational therapists would change. For effective use and adoption of assistive robotic technologies, care-providers and care-givers will need additional skills and understanding of deployment and commissioning of assistive robotic systems. In order to provide the best possible solutions, care professionals will need to be able to relate care plans to assistive technologies which need to provide just the right balance of support, without compromising the patient's own residual abilities.

Dealing with the messiness of real-world environments

A system that learns about its environment and sets its operational parameters to the conditions within which it is working needs to be able to know when these are no longer appropriate. Even for the simplest assistive robot or robotic device, the workspace would be a multi-objective, moving target problem that is underlined by the lack of deterministic behavioural models.

Developing a robust system that can deal with the aspects of changing user needs and their cluttered and noisy environment, will also be compounded by a range of other real-world issues - consider the grandchildren arriving for the afternoon or the curious cat, or something even messier...

I will end with a challenge that all mobile robots could potentially run into - as did this little robot last year.


In the following table, contact information relevant to the page. The first column is for visual reference only. Data is in the right column.

Photo of Dr Praminda Caleb-Solly
Name: Dr Praminda Caleb-Solly, BEng(Honours), MSc., PhD
Job title: Associate Professor in Independent Living Systems and Head of Electronics and Computer Systems
Department: FET - Engineering, Design and Mathematics
Organisation: University of the West of England
Telephone: +44 (0)117 32 83178

Praminda is based at the Bristol Robotics Laboratory, where she is the theme leader for Assisted Living. Praminda’s EPSRC-supported research focuses on the development of assistive robotics technologies. She is currently working on the I-DRESS project, developing a system that will provide assistance with dressing to disabled users or users such as high-risk health-care workers, whose physical contact with the garments must be limited during dressing to avoid contamination.