Event-Sampled Output Feedback Control of Robot Manipulators Using Neural Networks

Area of Science:

• Robotics
• Control Systems
• Artificial Intelligence

Background:

• Robot manipulators require precise trajectory tracking for various applications.
• Traditional control methods often struggle with system nonlinearities and disturbances.
• Event-triggered control offers potential for reduced computational load and improved efficiency.

Purpose of the Study:

• To develop an adaptive neural network-based output feedback control scheme for robot manipulator trajectory tracking.
• To investigate two configurations for implementing the neural network observer and controller.
• To ensure the stability and boundedness of the closed-loop system signals.

Main Methods:

• Utilized adaptive neural networks (NNs) for both observer and controller functions.
• Reconstructed robot joint velocities using a nonlinear NN observer from position measurements.
• Employed Lyapunov stability analysis to derive event-triggering conditions and NN weight update rules.
• Designed two distinct controller configurations based on observer placement.

Main Results:

• Demonstrated successful trajectory tracking for the robot manipulator.
• Showcased that all signals in the closed-loop system remain locally uniformly ultimately bounded despite disturbances.
• Validated the efficacy of the proposed event-triggered adaptive neural network control through simulations.

Conclusions:

• The proposed event-triggered adaptive neural network control framework effectively enables robot manipulators to track trajectories.
• The two proposed configurations provide viable options for implementing the observer and controller.
• The Lyapunov stability analysis confirms the robustness and boundedness of the system under disturbances.