In this article, a deep reinforcement learning based three-dimensional path following control approach is proposed for an underactuated autonomous underwater vehicle (AUV). To be specific, kinematic control laws are employed by using the three-dimensional line-of-sight guidance and dynamic control laws are employed by using the twin delayed deep deterministic policy gradient algorithm (TD3), contributing to the surge velocity, pitch angle and heading angle control of an underactuated AUV. In order to solve the chattering of controllers, the action filter and the punishment function are built respectively, which can make control signals stable. Simulations are carried out to evaluate the performance of the proposed control approach. And results show that the AUV can complete the control mission successfully.

Underwater robots demonstrate significant potential for hull biofouling removal. However, achieving uniform and damage-free cleaning remains a persistent challenge. The fixed arrangement of cleaning mechanisms, combined with the inherent non-uniformity of cavitation jet energy distribution, frequently results in inconsistent removal depths, leading to local over-cleaning or under-cleaning. To address this, this paper proposes an optimization framework to coordinate the robot’s motion with its cleaning mechanism. First, the flow field dynamics of the cavitation nozzle are elucidated using the Stress-Blended Eddy Simulation (SBES) turbulence model. Based on the Computational Fluid Dynamic (CFD) data, a Gaussian mapping model is constructed to quantify the relationship between jet erosion efficiency and robotic motion parameters. Furthermore, to resolve the multi-objective coverage parameter optimization problem, an improved hybrid metaheuristic algorithm—the Composite Cycloid Subtraction-Average-Based Optimizer (CCSABO)—is introduced to determine the optimal synchronization of forward and lateral velocities. Numerical simulations demonstrate the framework’s robustness across various fouling thickness scenarios and nozzle parameters. Notably, the CCSABO algorithm achieves a coverage rate of 99% and minimizes the uniformity index to 0.011, demonstrating superior consistency compared to traditional PSO and GWO methods. This improvement effectively mitigates the risk of hull damage while ensuring cleaning quality.