Jinyuan Li¹, Hao Li¹, Guorong Cui¹, Yan Kang^{1, *}, Yang Hu¹, Yingnan Zhou²

CMC-Computers, Materials & Continua, Vol.64, No.2, pp. 925-940, 2020, DOI:10.32604/cmc.2020.09903

Abstract With continuous urbanization, cities are undergoing a sharp expansion within the regional space. Due to the high cost, the prediction of regional traffic flow is more difficult to extend to entire urban areas. To address this challenging problem, we present a new deep learning architecture for regional epitaxial traffic flow prediction called GACNet, which predicts traffic flow of surrounding areas based on inflow and outflow information in central area. The method is data-driven, and the spatial relationship of traffic flow is characterized by dynamically transforming traffic information into images through a two-dimensional matrix. We introduce adversarial training to improve performance… More >

Yonghua Zhou¹, Xun Yang¹, Chao Mi¹

CMES-Computer Modeling in Engineering & Sciences, Vol.89, No.6, pp. 497-512, 2012, DOI:10.3970/cmes.2012.089.497

Abstract The movements of vehicles equipped with various positioning systems such as global and wireless positioning ones have provided beneficial channels to acquire abundant traffic flow information for total road network. However, not all vehicles are mounted with positioning systems and not all equipped positioning facilities are always active. This paper will address how to estimate the number and the states of unequipped vehicles through a series of observations on equipped ones. The proposed estimation process initiates employing the non-analytical microscopic traffic model for particle filter to estimate the number, positions and speeds of unequipped vehicles between an equipped one and… More >

Yonghua Zhou¹, Chao Mi¹, Xun Yang¹

CMES-Computer Modeling in Engineering & Sciences, Vol.86, No.6, pp. 533-550, 2012, DOI:10.3970/cmes.2012.086.533

Abstract The cellular automaton (CA) model for traffic flow describes the restrictive vehicle movements using the distance headway (gap) between two adjacent vehicles. However, the autonomous and synergistic behaviors also exist in the vehicle movements. This paper makes an attempt to propose a microscopic traffic simulation model such that the feedback control behavior during the driving process is incorporated into the CA model. The acceleration, speed holding and deceleration are manipulated by the difference between the gap and the braking reference distance the driver perceives, which is generally observed in the realistic traffic. The braking reference distance is related to the… More >

Yonghua Zhou¹, Chao Mi¹

CMES-Computer Modeling in Engineering & Sciences, Vol.83, No.1, pp. 1-22, 2012, DOI:10.3970/cmes.2012.083.001

Abstract Cellular automata (CAs), model the dynamics of complex systems as the state update of cells restricted from their own neighbors. This paper regards the tempo-spatial constraints as dummy neighborhoods of cells for train movement, such as scheduled movement authority and speed restriction, equivalent to the maximum displacements during the future certain time steps and each time step, respectively. Under the framework of CA modeling, this paper attempts to propose an improved CA model for moving-block railway network which incorporates the tempo-spatial constraints to capture the restrictive, synergistic and autonomous dynamics. We divide the one-dimensional cell lattice into several segments, called… More >

Durgesh Vikram¹, Sanjay Mittal², Partha Chakroborty¹

CMES-Computer Modeling in Engineering & Sciences, Vol.79, No.3&4, pp. 237-260, 2011, DOI:10.3970/cmes.2011.079.237

Abstract A stabilized finite element formulation is presented to solve the governing equations for traffic flow. The flow is assumed to be one-dimensional. Both, PW-type (Payne-Whitham) 2-equation models and the LWR-type (Lighthill-Whitham-Richards) 1-equation models are considered. The SUPG (Streamline-Upwind/Petrov-Galerkin) and shock capturing stabilizations are utilized. These stabilizations are sufficient for the 1-equation models. However, an additional stabilization is necessary for the 2-equation models. For the first time, such a stabilization is proposed. It arises from the coupling between the two equations and is termed as IEPG (Inter-Equation/Petrov-Galerkin) stabilization. Two behavioral models are studied: Greenshields' (GS) and Greenberg's (GB) models. Numerical tests… More >