Nonlinear motor-mechanism coupling tank gun control system based on adaptive radial basis function neural network...

Area of Science:

• Control Systems Engineering
• Robotics
• Mechanical Engineering

Background:

• Tank gun control systems (TGCS) face challenges from inherent nonlinearities such as friction, parameter uncertainty, and modeling errors.
• These nonlinearities complicate the precise spatial pointing control required for effective operation.
• Existing control strategies may struggle to adequately compensate for these complex dynamics.

Purpose of the Study:

• To develop and validate an advanced control system for the spatial pointing of a motor-mechanism coupling tank gun.
• To address and mitigate the impact of nonlinearities within the tank gun control system.
• To enhance the tracking accuracy, adaptability, and robustness of the tank gun's stabilization.

Main Methods:

• Modeling the mechanical and motor dynamics of a marching tank gun.
• Establishing a motor-mechanism coupling dynamics model using the principle of equivalent torque.
• Designing a computed torque controller incorporating a radial basis function neural network (RBFNN) for uncertainty estimation.
• Employing a modified adaptive algorithm for RBFNN weight estimation and a compensation controller for observer error.

Main Results:

• The proposed control system effectively manages the nonlinearities inherent in the tank gun control system.
• Simulation results demonstrate superior tracking accuracy under various operational conditions.
• The system exhibits strong adaptability and robustness, crucial for dynamic environments.

Conclusions:

• The developed computed torque controller, enhanced with RBFNN and adaptive algorithms, successfully addresses TGCS nonlinearities.
• The proposed control strategy significantly improves the spatial pointing control performance of tank guns.
• The validated effectiveness indicates potential for enhanced military targeting and stabilization systems.