A GPU based trafficparallel simulation module of artificial transportation systems

Abstract

Traffic micro-simulation is an important tool in the Intelligent Transportation Systems (ITS) research. In the micro simulation, a bottom up system can be built up by the interactions of vehicle agents, road agents, traffic lights agents, etc. The Artificial societies, Computational experiments, and Parallel execution (ACP) approach suggests integrating other metropolitan systems such as logistic, infrastructure, legal and regulatory, weather and environmental systems to build an Artificial Transportation System (ATS) to help solve ITSproblems. This is reasonable as the transportation system is complex that is affected by many systems interacting with each other. However, there is a challenge that the computing burden can be very heavy as there can be many agents of different kinds interacting in parallel in ATS. In recent years, the Graphics Processing Units (GPUs) have been applied successfully in many areas for parallel computing. Compared with the traditional CPU cluster, GPU has an obvious advantage of low cost of hardware and electricity consumption. In this paper, we build a parallel traffic simulation module of ATS with GPU. The simulation results are reasonable and a maximum speedup factor of 105 is obtained compared with the CPU implementations.

GPU Based Genetic Algorithms for the Dynamic Sub-area Division Problem of the Transportation System

“…A speedup factor of around 110 is achieved. As reported in (Wang and Shen, 2012c), for a lattice road network as small as 5×5, the speedup factor is 1.85. Only when the network is as large as 40×40, the speedup factor is 105.13.…”

Parallelization Strategies for Spatial Agent-Based Models

“…Recently, the trend has been on hybrid [54,2], GPU [36,49,63,6] and heterogeneous [60,62] methods. While these approaches allow concrete gains in simulation performance on commodity hardware, they come with an increased cost in implementation time due to the substantial more complex programming models.…”

Masivo: Parallel Simulation Model Based on OpenCL for Massive Public Transportation Systems’ Routes

“…For GPU's parallel architecture, we found five implementations, where four were related to road traffic and one to public transit simulation. In 2012, Wang et al built a GPU based traffic parallel simulation module, which was able to simulate three hours (10,800 s) of road traffic of a 40 × 40 lattice road network, with 5000 vehicle agents in 109 s, achieving a speed-up of 105 and a real-time factor of 102 [99]. Later, in 2014, Xu et al implemented a mesoscopic road traffic simulation on CPU/GPU.…”