Prediction and Evaluation of Heterogeneous Traffic Flow Based on Spatiotemporal Slices in Cooperative Vehicle Infrastructure System 2020-01-5238

With the development of vehicle-road coordination technology, driving modes of the vehicle are in the process of development from manual driving, assisted driving, autonomous driving, mixed driving between people and vehicles to advanced unmanned driving. Heterogeneous traffic flows are essential for the development of vehicle-road coordination systems. However, in real life, it is necessary to intelligently monitor heterogeneous traffic flow because they involve many types of vehicles, complex scenarios, complex hidden factors in traffic conditions, and different operating characteristics of vehicles at different times and places. To evaluate and predict the traffic efficiency, this paper first builds a heterogeneous traffic efficiency prediction and evaluation system, uses the SUMO (Simulation of Urban Mobility) simulation platform to build a simulation environment and collects data, analyzes and evaluates efficiency indicators for a single scenario, then evaluates and predicts efficiency for multiple scenarios. Taking into account the complexity and randomness of the road traffic system, a road traffic efficiency evaluation model including an input layer, a hidden layer and an output layer is further established based on BP (Back Propagation) neural network technology. The simulation is performed according to the spatiotemporal slice characteristics of the scenario. The index evaluates the system traffic efficiency and predicts the overall efficiency evaluation result of the road network based on its individual scenario characteristic value. The results show that the evaluation results obtained by the constructed road traffic efficiency evaluation model are consistent with the simulation results, and the road traffic efficiency evaluation can be effectively carried out. The method of predicting the overall performance evaluation result of the road network based on its single scene feature value is effective and feasible.