A Comparison Between Conventional and User-Intention-Based Adaptive Pushrim-Activated Power-Assisted Wheelchairs

Area of Science:

• Robotics
• Biomechanics
• Human-Computer Interaction

Background:

• Users of pushrim-activated power-assisted wheelchairs (PAPAW) have varying assistance needs based on activity and personal preference.
• Developing adaptive PAPAW controllers is crucial for accommodating these individual differences.

Purpose of the Study:

• To integrate a user intention estimation framework into a PAPAW system.
• To develop personalized adaptive controllers that adjust power assistance based on identified user intentions.

Main Methods:

• Collected wheelchair propulsion kinetics (pushrim forces) during various daily activities.
• Trained random forest classification models to estimate user intentions (e.g., moving forward, turning, braking).
• Implemented and tested the intention estimation framework in real-time on a laboratory-developed PAPAW.

Main Results:

• Real-time user intention predictions closely matched offline model performance.
• The power-assist ratio was dynamically adjusted based on identified user intentions.
• Propulsion effort was significantly reduced for participants using the adaptive PAPAW controller.

Conclusions:

• Adaptive intention-based controllers are effective in reducing wheelchair propulsion effort.
• User preferences for controller performance vary across individuals and maneuvers, highlighting the need for customized solutions.
• Personalized adaptive controllers are essential for optimizing PAPAW functionality for diverse users and activities.