The Assistive Robotic Transport for Youngsters (ARTY) is a smart, robotic wheelchair designed to give the freedom of independent mobility to young children with disabilities, particularly those who are unable to use a regular powered wheelchair safely. ARTY was selected as one of the UK 2012 James Dyson Award Finalists and had a TechReview mention.

The design goal behind ARTY was a wheelchair that could function as both as a daily assistive living tool and a training tool in hospitals and rehabilitation centers. We integrated ideas from robotics technology to prevent unsafe situations—using a combination of sensors, ARTY perceives its environment and makes control decisions to avoid crashes. Although ARTY can be fully autonomous (it can drive itself), we believe a shared-control approach is more natural for most individuals, particularly children who may graduate on to use a regular powered wheelchair.

In addition to the hardware platform, I developed online machine learning methods that enable occupational therapists to teach ARTY how to assist a child drive safely.

Related Publications:

• Learning Assistance by Demonstration: Smart Mobility with Shared Control and Paired Haptic Controllers,
  Harold Soh and Yiannis Demiris
  Journal of Human-Robot Interaction, vol 4, No 3, 2015.
• When and How To Assist: An Iterative Probabilistic Model for Learning Assistance by Demonstration
  Harold Soh and Yiannis Demiris,
  IEEE/RSJ International Conference on Intelligent Robots and Systems, Tokyo, Japan, 2013.
• Towards Early Mobility Independence: An Intelligent Paediatric Wheelchair with Case Studies
  Harold Soh and Yiannis Demiris,
  IROS Workshop on Progress, Challenges and Future Perspectives in Navigation and Manipulation Assistance for Robotic Wheelchairs, Vilamoura, Portugal, 2012.
• Multi-reward policies for medical applications: anthrax attacks and smart wheelchairs
  Harold Soh and Yiannis Demiris
  MedGEC 2011, 7th GECCO Workshop on Medical Applications of Genetic and Evolutionary Computation, 471-478, 2011
• Involving Young Children in the Design of a Safe, Smart Paediatric Wheelchair
  Harold Soh and Yiannis Demiris
  ACM/IEEE HRI-2011 Pioneers Workshop, Lausanne 2011

A Short FAQ

Q: Can ARTY avoid all kinds of obstacles?

ARTY can avoid things that its sensors can see. Recent student led projects have extended ARTY to detect kerbs and drop-offs and provide feedback to the driver via a force-feedback joystick.

Q: Will ARTY be made open-source?

Some elements (those under our copyright) will be made available. I’m currently packaging the obstacle avoidance code into a ROS package that I will make free online. In the meantime, for details, you can refer to my publications.

Q: Can I take ARTY for a test drive?

It depends. Do send me an email at haroldsoh@imperial.ac.uk and we’ll see what we can set up.

Q: Is ARTY available commercially?

Not yet. ARTY is a research project within the Personal Robotics Lab at Imperial College London. It is currently being field tested at a London hospital.