Categories
Articles

Shared control for wheelchairs

To help aging populations with mobility, researchers are developing robotic wheelchairs. Typically, control switches between the user of the wheelchair and the robot when tasks become difficult or dangerous. However, users sometimes become frustrated when losing control of the wheelchair and complete autonomy may lead to the patient losing certain capabilities (which are not practiced […]

To help aging populations with mobility, researchers are developing robotic wheelchairs. Typically, control switches between the user of the wheelchair and the robot when tasks become difficult or dangerous.

However, users sometimes become frustrated when losing control of the wheelchair and complete autonomy may lead to the patient losing certain capabilities (which are not practiced any more). Therefore, it is important to provide the right amount of help to the patient: no more, no less.

For this purpose, Urdiales et al. implement a control strategy where the robot and user continuously share control of the wheelchair. This is done by combining commands sent by the user using a joystick with commands computed by a potential field that ensures that the robot is repulsed from obstacles while being attracted to a goal. Both commands receive a weight based on how efficient the user and robot are at a given task. The resulting command is used to control the wheelchair.

The wheelchair, augmented with odometry and a frontal Hokuyo laser URG04-RX for localization and obstacle detection, was tested in a rehabilitation hospital by 30 users with different degrees of cognitive and physical disabilities. Users were asked to go through a door, proceed down a hallway, turn around in the hallway and come back. In other tests, users simply needed to go down a hallway.

All patients were successful at completing the navigation task using shared control. Furthermore, shared control improved performance of both the human and robot, helped patients learn how to use electric wheelchairs and avoided all collisions. Interestingly, shared control tends to equalize performance among patients with different disabilities, meaning the control is able to adapt to each patient’s needs.

In the future, Urdiales et al. plan on testing their system in more complex, human-like scenarios and with patients who have more severe disabilities.

By Sabine Hauert

Sabine Hauert is Assistant Professor in Robotics at the University of Bristol in the UK. Her research focusses in designing swarms that work in large numbers (>1000), and at small scales (<1 cm). Profoundly cross-disciplinary, Sabine works between Engineering Mathematics, the Bristol Robotics Laboratory, and Life Sciences. Before joining the University of Bristol, Sabine engineered swarms of nanoparticles for cancer treatment at MIT, and deployed swarms of flying robots at EPFL.

Sabine is also President and Co-founder of Robohub.org, a non-profit dedicated to connecting the robotics community to the world.

As an expert in science communication with 10 years of experience, Sabine is often invited to discuss the future of robotics and AI, including in the journals Nature and Science, at the European Parliament, and at the Royal Society. Her work has been featured in mainstream media including BBC, CNN, The Guardian, The Economist, TEDx, WIRED, and New Scientist.