Path Planning Based on Traffic Flow Prediction for Vehicle Scheduling

The construction of road networks is becoming increasingly complex, and the problem of traffic congestion is becoming more and more serious. Since the traditional route planning algorithm cannot meet the real-time requirements in complex and dynamic road networks, we propose a vehicle route planning method based on traffic flow prediction and actual road conditions. Firstly, we use LSTM neural network to predict traffic flow, which can effectively improve the real-time response of route planning in complex situations. Then, based on the road network information, we model the real road conditions with a graph structure. Next, the traffic flow and waiting time of traffic lights are used to update the congestion coefficient matrix of the road network graph. We design a dynamic updating algorithm of the road network information to match the dynamic changes of the road conditions. On this basis, we propose a path-planning algorithm based on Dynamic Updating Road Network Information. In addition, the Bezier curve smoothes the output curve to obtain a smoother driving route. Finally, we conduct simulations based on the road network data in Nanjing City. The results show that the proposed method can provide a shorter travel time route and improve users’ travel efficiency.