Convergence-guaranteed time-varying RRT path planning for profiling floats in 4-dimensional flow

This paper presents an extension to the Rapidly-exploring Random Tree (RRT) al- gorithm applied to drifting autonomous un- derwater vehicles. The proposed algorithm is able to plan paths that guarantee con- vergence in the presence of time-varying ocean dynamics. The method utilizes 4- Dimensional ocean model prediction data as an evolving basis for expanding the tree from the start location to the goal. The perfor- mance of the proposed method is validated through Monte-Carlo simulations. Results illustrate the importance of the temporal variance in path execution and demonstrate the convergence performance of the proposed methods.