Adaptive Tracking Control of Surface Vessel Using Optimized Backstepping Technique.

Area of Science:

• Marine Engineering
• Control Systems Theory
• Robotics

Background:

• Surface vessel control systems often exhibit complex, nonlinear dynamics.
• Traditional backstepping control can be challenging to optimize for these systems.
• Solving the Hamilton-Jacobi-Bellman equation for optimal control is computationally intensive.

Purpose of the Study:

• To develop a novel tracking control approach for surface vessels.
• To integrate optimization principles into the backstepping control design.
• To address the challenges of optimizing complex vessel dynamics.

Main Methods:

• Developed an optimized backstepping (OB) control technique.
• Modeled surface vessel dynamics in a second-order strict feedback form.
• Employed an actor-critic reinforcement learning (RL) strategy for virtual and actual controls.

Main Results:

• The optimized backstepping approach effectively achieves tracking control.
• Actor-critic RL successfully optimized both virtual and actual control subsystems.
• Simulation results validated the proposed control strategy's effectiveness.

Conclusions:

• The optimized backstepping control offers a robust solution for surface vessel tracking.
• Reinforcement learning provides a viable method to overcome complexities in optimization-based control.
• The proposed method demonstrates superior performance in simulations for surface vessel control.