Optimal Path Planning Algorithm of AUV State Space Sampling Based on Improved Cost Function

In order to solve the problem of autonomous underwater vehicle (AUV) trajectory planning for local obstacle avoidance, this paper proposes an optimal trajectory planning algorithm based on improved cost function of AUV state space sampling. This algorithm samples the end states of the s-direction and q-direction in the Frenet coordinate system with the initial position, speed, direction, acceleration, and reference path of the AUV, and uses a fifth-order polynomial to generate a series of alternative trajectories. And combined with the potential field strength to optimize the selection of the traditional cost function, by calculating the end state of the minimum trajectory of the cost function in each sampling period as the starting state of the next period, iteratively obtain the trajectories in two directions and couple with the time parameter t get the optimal trajectory. Simulation results show that the algorithm can be used to plan a set of trajectories that can efficiently avoid underwater emerging obstacles and meet the actual braking constraints of AUV.