Parallel Hierarchical Control-Based Efficiency Enhancement for Large-Scale Virtual Reality Traffic Simulation

Abstract

Virtual reality (VR)-based traffic simulation plays an important role in the study of transportation systems, especially in the testing of connected and autonomous vehicles. However, a simulator that provides a large-scale traffic scene with high fidelity is still not available. To address this issue, the parallel hierarchical control method is proposed in this article, which presents a framework that enables the efficient simulation of a VR-based large-scale traffic scene. First, the original data of the basic traffic components are acquired through a modification and generation method. Then, based on the proposed method for spatial parallel slicing, the prepared data and simulation tasks are separated and distributed to subcontrollers considering the connected-vehicle environment. Meanwhile, a fidelity loss-based hierarchical control method is integrated to stratify the separate data into multiple levels. Finally, the experiments are carried out on a virtual driving platform that indicates that the proposed approach effectively ensures the fidelity of the visualized scenes and performs a better allocation of computational resources.