An Improved Probabilistic Roadmap Algorithm with Potential Field Function for Path Planning of Quadrotor
Jinbao Chen, Yimin Zhou, Jin Liang Gong, Yu Deng · 2019
Probabilistic Roadmap Method (PRM) is a path planning method based on random sampling strategy, which can solve the problem that effective paths are difficult to construct with most algorithms in high-dimensional space. In order to achieve the obtain all the possibilities, the PRM algorithm requires a large amount of sampling data, which leads to double increase of the computing time and low efficiency in the narrow passage planning. In this paper, an improved PRM algorithm is proposed with the introduction of potential field in the whole planning space, called P-PRM. The sampled potential field strength is used to select valuable nodes to avoid collision detection for the sampling points. Cost function is set to avoid the algorithm falling into local optimum in narrow channels or target areas. Experiments are performed for quadrotor unmanned vehicle of path planning in 2D environment, and it is proved that the proposed P-PRM algorithm is superior to the traditional PRM algorithm as the planning efficiency of narrow channel is higher and calculation time or path length is lower. Moreover, the feasibility of the proposed improved algorithm has been verified by simulation in 3D space.