Cooperative Game-Based Approximate Optimal Control of Modular Robot Manipulators for Human-Robot Collaboration.

Area of Science:

• Robotics
• Control Systems
• Artificial Intelligence

Background:

• Human-robot collaboration (HRC) faces challenges in estimating human motion intention and optimizing performance for modular robot manipulators (MRMs).
• Accurate human motion intention estimation is crucial for seamless and safe HRC.
• Performance optimization is key to efficient collaborative tasks.

Purpose of the Study:

• To propose a cooperative game-based approximate optimal control method for MRMs in HRC tasks.
• To develop a human motion intention estimation method using only robot position measurements.
• To optimize the control of MRM systems for enhanced HRC performance.

Main Methods:

• A harmonic drive compliance model-based method for human motion intention estimation using robot position data.
• Transformation of the optimal control problem into a cooperative game using a cooperative differential game strategy.
• Application of the adaptive dynamic programming (ADP) algorithm with critic neural networks to solve the Hamilton-Jacobi-Bellman (HJB) equation and identify joint cost functions.
• Lyapunov theory to prove the ultimately uniformly bounded (UUB) trajectory tracking error.

Main Results:

• A novel method for estimating human motion intention in MRMs during HRC tasks.
• Successful transformation of the optimal control problem into a cooperative game framework.
• Development of an ADP-based approach to find Pareto optimal solutions for joint cost functions.
• Experimental validation demonstrating the effectiveness and advantages of the proposed control method.

Conclusions:

• The proposed cooperative game-based approximate optimal control method effectively addresses challenges in HRC-oriented MRMs.
• The method enhances human motion intention estimation and optimizes system performance.
• Experimental results confirm the superiority of the developed approach for HRC tasks.