Research on AUV Trajectory Tracking Control Based on Improved Genetic Algorithm

In order to enhance the trajectory tracking performance and stability of controllers for autonomous underwater vehicles (AUVs), a nonlinear model predictive control scheme incorporating improved genetic algorithm (IGA) is proposed. When considering practical operational constraints, including environmental disturbances and actuator saturation, an auxiliary nonlinear control law was constructed using Lyapunov-Based backstepping methodology. The designed controller integrates this nonlinear control law with the model predictive control (MPC) framework, incorporating inequality constraints into the predictive control formulation to ensure closed-loop stability while avoiding system linearisation. The traditional manual parameter tuning process was replaced by an enhanced genetic algorithm, which was used to optimize the controller coefficients. This enhancement resulted in an improvement in tracking performance. Validation was conducted through simulations under realistic AUV hydrodynamic parameters and disturbance conditions. The experimental results demonstrate that, in comparison with conventional MPC solutions, the proposed approach achieves a 58.3% reduction in average tracking error and maintains stable trajectory tracking under 25% parametric perturbations, thus confirming its superior control precision and robustness.