Bounded barrier Lyapunov function-based neural network control of space robot with output tracking error constraints...

Area of Science:

• Robotics
• Control Systems Engineering
• Artificial Intelligence

Background:

• Space robots face challenges like output tracking error constraints, input saturation, and joint input dead zones.
• Existing control strategies often struggle to address these issues simultaneously, impacting performance and stability.

Purpose of the Study:

• To propose a novel control strategy for space robots that simultaneously addresses output tracking error constraints, input saturation, and joint input dead zones.
• To enhance the control accuracy and stability of space robot systems.

Main Methods:

• Development of a novel Bounded Barrier Lyapunov Function (BBLF) method for output tracking error constraints.
• Design of an adaptive control scheme incorporating BBLF and an assistant system for input saturation.
• Utilization of a radial basis function neural network (RBFNN) to approximate uncertainties and compensate for joint input dead zones.

Main Results:

• The proposed BBLF-NN control strategy effectively manages output tracking error constraints.
• The strategy successfully addresses input saturation and compensates for joint input dead zones.
• Numerical simulations demonstrate the algorithm's efficacy in improving control accuracy and stability.

Conclusions:

• The BBLF-NN control strategy provides a comprehensive solution for multiple challenging issues in space robot control.
• This approach significantly enhances the robustness, accuracy, and stability of space robot operations.
• The validated algorithm offers a promising advancement for autonomous space missions.