AI-Powered Wheelchairs: Navigating the Future of Mobility

A new wave of research is focused on developing AI-powered wheelchairs capable of navigating complex environments, offering increased independence for users with severe disabilities. While wheelchair users often demonstrate remarkable skill in maneuvering tight spaces, advancements in artificial intelligence are aiming to further enhance their mobility, and safety.

The Rise of Smart Wheelchairs

Researchers at the German Research Center for Artificial Intelligence (DFKI) in Bremen, Germany, are leading the charge in developing prototype sensor-equipped electric wheelchairs. Led by Christian Mandel, a senior researcher at DFKI, the team is exploring both semiautonomous and fully autonomous navigation systems. IEEE Spectrum reports on these developments.

Semiautonomous vs. Autonomous Control

“Semiautonomous is the shared control system where the person sitting in the wheelchair uses the joystick to drive,” explains Mandel. AOL highlights this distinction. “Fully autonomous is controlled by natural-language input. You say, ‘Please drive me to the coffee machine.’ ”

Technical Components and Functionality

The DFKI wheelchairs are equipped with a suite of sensors and computing power, including:

• Two laser scanners
• A depth-imaging camera
• Odometry sensors on the front drive wheels
• A screen and button-based input device
• An emergency stop switch
• A computing unit

These components function together to create a system that can map the environment using Simultaneous Localization and Mapping (SLAM) technology, avoid obstacles, and respond to user commands. The wheelchairs utilize the open-source ROS2 Nav2 navigation system for autonomous control.

Safety and User Interaction

The research team has also focused on safety, integrating sensor data from both the wheelchair and external sources, such as drone-based cameras. Users can interact with the system by pressing a key on the human-machine interface or speaking a command, then confirming or rejecting the instruction. The system then guides the user to their destination, adjusting for obstacles along the way.

Challenges and Considerations

Despite the promising advancements, several challenges remain before smart wheelchairs turn into widely available. Pooja Viswanathan, CEO & founder of Braze Mobility, emphasizes the importance of affordability and reliability. “Cost remains a major barrier,” she says. “Funding systems are often not designed to support advanced add-on intelligence unless there is remarkably clear evidence of value and safety.”

Louise Devinge, a biomedical research engineer from IRIS, points out the increasing complexity of these systems and the need for careful management of communication and synchronization between sensors. “The more sensing, computation, and autonomy you add,” she says, “the harder it becomes to ensure robust performance across the full range of real-world environments that wheelchair users encounter.”

Future Outlook

Christian Mandel anticipates that smart wheelchairs will be ready for the mainstream market within the next 10 years. The REXASI-PRO project, while currently advanced, is seen as a valuable step towards building more intelligent and autonomous wheelchairs. The focus is shifting towards designing better partnerships between the user and the technology, rather than attempting to replace the user’s skills entirely.

Mandel reflects on his early work developing a head-joystick controlled wheelchair, realizing the inherent abilities of wheelchair users. “At that point in time, I realized that even persons that had severe handicaps [traveling through] a narrow passage, they did very, very well. And then I realized, okay, there is this need for this technology, but never underestimate what [wheelchair users] can do without it.”