Admittance-Based Adaptive Cooperative Control for Multiple Manipulators With Output Constraints

Area of Science:

• Robotics
• Control Systems Engineering
• Artificial Intelligence

Background:

• Cooperative manipulation by multiple robots presents challenges in trajectory tracking and constraint handling.
• Existing adaptive control methods often lack guaranteed global stability or struggle with physical limitations.

Purpose of the Study:

• To develop a novel adaptive control methodology for cooperative manipulation of a rigid object by multiple manipulators.
• To ensure precise trajectory tracking while adhering to physical and environmental constraints.

Main Methods:

• Utilized an admittance model to generate online reference trajectories for individual manipulators.
• Employed an integral barrier Lyapunov function to manage system constraints.
• Integrated adaptive neural networks (NNs) for approximating manipulator dynamics uncertainties.
• Introduced a switching function to achieve global closed-loop stability.

Main Results:

• The proposed method successfully generated online reference trajectories for cooperative object transport.
• Integral barrier Lyapunov functions effectively addressed physical and environmental constraints.
• Adaptive NNs with a switching function ensured global stability, outperforming conventional methods.

Conclusions:

• The novel adaptive control methodology offers a robust solution for cooperative robot manipulation.
• The approach demonstrates enhanced stability and constraint satisfaction in complex robotic tasks.
• Simulation results validate the efficacy of the proposed admittance-based control strategy.