Unmanned Ground Vehicle in Unstructured Environments Applying Improved A-star Algorithm

Abstract

The research objective of this paper is to explore the path planning problem of Unmanned Ground Vehicles (UGV) in unstructured environments. Firstly, we establish a kinematic model based on the UGV, and after reasonable assumptions, only consider the longitudinal motion of the UGV. Next, provide the overall control architecture of the vehicle mainly including the information of an unstructured map and a path planner. Then, based on the rasterization of the 2D map, the A-star algorithm is applied to obtain the path, and have been improved by our team. Finally, in the robot emulation environment, the effectiveness and stable of our designed approach have been demonstrated through simulation experiments.