DFKI Researchers Develop Advanced Smart Wheelchair Prototypes
Christian Mandel and Serge Autexier, researchers at the German Research Center for Artificial Intelligence (DFKI) in Bremen, Germany, have co-led a team that developed prototype sensor-equipped electric wheelchairs. These wheelchairs are designed to navigate rooms filled with potential obstacles. Earlier this month, the DFKI researchers presented their work at the CSUN Assistive Technology Conference in Anaheim, California. The development marks a significant step in smart wheelchair technology, aiming to enhance mobility for individuals with reduced mobility.
The research team conducted experiments utilizing two identical smart wheelchairs. Each unit was equipped with two lidars, a 3D camera, odometers, user interfaces, and an embedded computer. Furthermore, the researchers tested a novel safety system that integrated sensor data from the wheelchairs with data from sensors located within the room. These room-based sensors included drone-mounted color and depth cameras, providing a comprehensive view of the environment.
Autonomous Navigation and User Interaction Explored
In contrast to semiautonomous mode, where the user operates the wheelchair with a joystick, the autonomous mode utilizes the open-source ROS2 Nav2 navigation system. This mode allows for control via natural-language input, enabling users to issue commands such as, “Please drive me to the coffee machine.” The wheelchairs also incorporated simultaneous localization and mapping (SLAM) maps and local obstacle-avoidance motion controllers to ensure safe and efficient navigation.
One scenario tested by Mandel and his team involved the user initiating a command by pressing a key on the wheelchair’s human-machine interface, speaking the command, and then confirming or rejecting the instruction through the same interface. Once a command was confirmed, the mobility device would guide the user along a predetermined path to the destination. During this guidance, sensors would actively detect obstacles and adjust the wheelchair’s movement accordingly to avoid them.
Braze Mobility Enhances Wheelchair Safety
Braze Mobility, a company based in Toronto, is contributing to the advancement of wheelchair safety by developing blind-spot sensors for electric wheelchairs. These sensors are designed to detect obstacles within areas that are typically difficult for a wheelchair user to see. By detecting these blind spots, the technology aims to prevent collisions and improve overall situational awareness for the user.
The Braze sensors offer a versatile solution, as they can be added to any wheelchair. This addition transforms a standard wheelchair into a smart wheelchair by providing multimodal alerts to the user. This approach focuses on supporting existing user capabilities rather than attempting to replace them, reflecting a collaborative design philosophy between user and technology.
Future of Smart Wheelchairs and AI Integration
Christian Mandel estimates that smart wheelchairs will be ready for the mainstream marketplace within 10 years. He is driven by the inspiration that led him into the field years ago, recalling early experiences developing a smart wheelchair controllable with a head joystick. Mandel’s observation that individuals with severe handicaps could navigate narrow passages effectively highlighted the need for such technology while also emphasizing the capabilities of wheelchair users.
Pooja Viswanathan from Braze Mobility notes that the REXASI-PRO system, while currently beyond the reach of present-day smart wheelchair technologies, holds significant importance for the long term. The REXASI-PRO project, which stands for Reliable and Explainable Swarm Intelligence for People With Reduced Mobility, is recognized for its strengths in intelligent navigation and advanced sensing. This system aims to build a wheelchair capable of interpreting and responding to complex environments in a more autonomous manner. The approach of integrating trustworthy and explainable AI is considered paramount for any mobility technology where safety, reliability, and user confidence are critical.
The challenges in smart wheelchair development include cost, which remains a major barrier, as funding systems are often not designed to support advanced add-on intelligence without clear evidence of value and safety. Reliability is another hurdle; a smart wheelchair must function effectively in messy, variable conditions of daily life, not just ideal scenarios. Additionally, the human factors dimension, encompassing diverse user needs, means that a one-size-fits-all solution is rarely feasible. The increased complexity of smart wheelchairs necessitates more sensing, requiring careful management of communication and synchronization within the system. In the near term, the primary challenge is designing better partnerships between the user and the technology, rather than solely replacing user input with AI. This article was supported by the IEEE Foundation and a Jon C. Taenzer fellowship grant.
✨ Intelligent Curation Note
This article was processed by AI Universe’s Intelligent Curation system. We’ve decoded complex technical jargon and distilled dense data into this high-impact briefing.
Estimated time saved: ~1 minutes of reading.
Tools We Use for Working with AI:
