Three-dimensional Path Planning for Unmanned Ground Vehicles in Complex Building Environments

In contrast to common ground environments, complex buildings possess a more intricate three-dimensional (3D) environment structure. Existing methods face difficulties in solving the path planning problem for unmanned ground vehicles (UGVs) when navigating between multiple floors. In this paper, we propose a 3D path planning framework for UGVs in complex building environments. Firstly, we adopt the Normal Distribution Transformation (NDT) method to efficiently model the complex building environment. Further, we propose an improved 3D heuristic search algorithm and design the stability cost value and building structure heuristic value to improve the path stability and search efficiency. Simulation experiments are conducted in typical complex building environments. The experimental results show the effectiveness of the proposed method, and the average pitch and roll angle in stair traversal scenario are 82.0 % and 38.7% of the traditional method, respectively.