Abstract: Rapidly-exploring random tree (RRT) algorithm is a common algorithm for path planning of intelligent vehicle. But traditional RRT and RRT^* algorithms have disadvantages of large path jitter, easy to fall into local region and low calculation efficiency. In view of these problems, an improved RRT^* algorithm for the path planning of intelligent vehicle based on safety field and real vehicle driving data is proposed in this paper. Firstly, a safety field based on safety distance model is established, and the key parameters of the model are extracted through driving data acquisition test. On this basis, an improved RRT^* algorithm with safety field guidance and angle constraint strategies is proposed. Finally, the algorithm is verified by simulation. The results show that the path planning method proposed can calculate the effective path meeting the curvature constraint of vehicle trajectory with faster search speed and higher success rate

Key words: intelligent vehicle, path planning, RRT^* algorithm, safety field, driving data