Supervisory controller design for a robot-assisted reach-to-grasp rehabilitation task

Area of Science:

• Robotics
• Rehabilitation Engineering
• Control Systems

Background:

• Rehabilitation robotic devices require precise control for effective patient assistance.
• Automating trajectory generation for complex tasks like reach-to-grasp is a significant challenge.
• Existing control systems may lack the adaptability needed for diverse rehabilitation needs.

Purpose of the Study:

• To develop a high-level supervisory controller for rehabilitation robotic devices.
• To enable automatic generation of reference trajectories for low-level controllers.
• To facilitate reach-to-grasp movements in rehabilitation tasks.

Main Methods:

• Designed an event-driven, asynchronous discrete event system (DES) as the supervisory controller.
• Utilized a finite automaton to model the supervisory controller's behavior.
• Conducted extensive simulations of the supervisory controller in a reach-to-grasp rehabilitation scenario.

Main Results:

• The proposed supervisory controller successfully generated reference trajectories for the low-level controller.
• Simulations demonstrated the controller's capability in executing reach-to-grasp tasks.
• The system showed feasibility and convenience for rehabilitation robotics applications.

Conclusions:

• The developed supervisory controller is a feasible and convenient solution for automating trajectory generation in rehabilitation robotics.
• The discrete event system approach provides an effective framework for controlling robotic rehabilitation devices.
• This controller can enhance the functionality and applicability of robotic devices in rehabilitation settings.