I. Introduction
With the acceleration of urbanization and the growth of traffic demand, traffic congestion has become more and more prominent, which has seriously affected people’ s production and life. In order to alleviate traffic pressure and improve the efficiency of road network, path planning has become an important research field [1]. Traditional path planning algorithms are mainly based on mathematical models and optimization theory. However, these algorithms often do not fully consider the actual traffic flow, so they cannot provide accurate path suggestions for drivers [2]. In recent years, DRL has made breakthroughs in many fields, especially in the fields of games, robots and autonomous driving. This article aims to apply DRL to the field of path planning and design an optimized path planning algorithm that can fully consider the actual traffic flow.