An efficient motion planning based on grid map: Predicted Trajectory Approach with global path guiding

This paper presents a Predicted Trajectory Approach (PTA) for global motion planning of an underactuated unmanned surface vehicle (USV). Different from the conventional path planning algorithms which mostly focus on the shortest, safest or smoothest paths, PTA takes all the dynamic constraints of the USV into account, and finds the global trajectory that can be traced precisely via its own maneuvering system. In present strategy, the predicted trajectories produced by the mathematical model of the USV are decomposed into a series of waypoints on grid map. Then an efficient trajectory can be found for the USV by performing the A* search method. To improve the search efficiency and accuracy, the PTA is enhanced with global path guiding (GPTA). When searching for feasible trajectories, GPTA can employ the global path obtained by Theta* to accurately estimate the heuristic cost of the objective function, rather than using the Euclidean distance directly as PTA does. Numerical simulations in various scenarios are conducted to compare the performance of the PTA, GPTA, Trajectory Units (TUs) and hybrid path planning algorithm combining global and local path planning (HGL). The results demonstrate that GPTA has higher safety and navigation efficiency for motion planning of the underactuated USV.

Automated summary

This paper introduces a Predicted Trajectory Approach (PTA) for global motion planning of an underactuated unmanned surface vehicle (USV), which takes into account all dynamic constraints and uses A* search method and global path guiding (GPTA) to efficiently find the trajectory.