Path Planning for Autonomous Underwater Vehicles based on an Improved Artificial Jellyfish Search Algorithm in Multi-obstacle Ocean Current Environment

Path planning for autonomous underwater vehicles (AUVs) is a key research focus in the marine domain, requiring consideration of the underwater environment's complexity and the efficiency of the planning algorithms. Firstly, this paper uses the grid method to construct a realistic ocean terrain for the underwater environment and incorporates obstacles and ocean currents. Also, the AUV's movement strategy and posture are illustrated and constrained by a visual search strategy. Secondly, an improved artificial jellyfish search (IJS) algorithm is to overcome the drawbacks of low convergence accuracy and adaptability of the basic algorithm. The IJS algorithm determines the optimal path point by optimizing the objective function. This objective function not only takes into account factors such as path navigation time and path safety but also designs an ocean current disturbance model, which facilitates the AUV to successfully reach the target point while avoiding obstacles and strong side currents. Finally, the optimal smoothed path is output using a cubic spline midpoint interpolation method. Multiple simulation experiments are carried out on the constructed environment model. The comparison results show that the IJS algorithm with a short running time has the optimal time cost and ocean current penalty cost for the planned path. In addition, the IJS algorithm is also shown to be adaptable in the field of multi-AUV movements.