关键词: 深度学习；A*算法；路径规划；智能车

Abstract: The paper address the problem of intelligent vehicle path planning and propose a novel approach based on deep learning and the A* algorithm. By combining convolutional neural networks (CNN) and long short-term memory (LSTM), it achieves accurate prediction of the shortest path in complex environments. Through experiments conducted in three different terrain environments, the paper validate the effectiveness and performance advantages of the approach. The results demonstrate significant improvements in path planning time, search time, path length, and average velocity. This research contributes to the autonomous navigation and path planning of intelligent vehicles and provides new insights and methods for further exploration of intelligent vehicle path planning based on deep learning and the A* algorithm.

Key words: Deep learning; A* algorithm; Path planning; Intelligent vehicles