Adaptive Human Force Scaling via Admittance Control for Physical Human-Robot Interaction

Area of Science:

• Robotics
• Human-Robot Interaction
• Control Systems

Background:

• Collaborative manipulation tasks require robots to adapt their assistance levels.
• Existing admittance controllers often lack adaptability to dynamic human intentions.

Purpose of the Study:

• To design an adaptive admittance controller for robots in human-robot collaboration.
• To improve task performance by adaptively scaling robot contribution based on human movement intention.

Main Methods:

• Movement intention estimation using fuzzy logic based on human force and object velocity.
• Adaptive gain adjustment in the admittance controller without altering the time constant.
• Validation through a physical human-robot interaction (pHRI) experiment using Fitts' reaching task.

Main Results:

• Identified an optimal admittance time constant for maximizing human force amplification.
• Demonstrated a desirable admittance gain profile for effective co-manipulation.
• Showcased improved overall task performance through adaptive control.

Conclusions:

• The proposed adaptive admittance controller effectively enhances collaborative manipulation.
• Interpreting human movement intention allows for dynamic and optimized robot assistance.
• This approach offers a promising direction for more intuitive and efficient human-robot collaboration.