关键词: 智能车/
换道模型/
下匝道路段/
自适应巡航控制
English Abstract
Hybrid traffic flow model for intelligent vehicles exiting to off-ramp
Dong Chang-Yin1,2,Wang Hao1,2,
Wang Wei1,2,
Li Ye1,2,
Hua Xue-Dong1,2,3
1.Jiangsu Key Laboratory of Urban Intelligent Traffic System, School of Transportation, Southeast University, Nanjing 210096, China;
2.Jiangsu Province Collaborative Innovation Center of Modern Urban Traffic Technologies, Nanjing 210096, China;
3.School of Architecture, Southeast University, Nanjing 210096, China
Fund Project:Project supported by the National Natural Science Foundation of China (Grant Nos. 51478113, 51508122), the Jiangsu Planned Projects for Postdoctoral Research Funds, China (Grant No. 1701082B), and the Scientific Research Foundation of Graduate School of Southeast University, China (Grant No. YBJJ1734).Received Date:27 December 2017
Accepted Date:14 May 2018
Published Online:20 July 2019
Abstract:With the rapid development of vehicular technology, hi-tech manufacturing facilities are equipped in intelligent vehicles to improve road capacity and traffic safety. However, freeway diverge segment has significant influence on current traffic flow, and could affect the heterogeneous traffic flow consisting of manual and intelligent vehicles. The primary objective of this study is to evaluate how intelligent vehicles affect traffic flow at an off-ramp bottleneck.In order to depict the car-following dynamics of manual vehicles, the modified comfortable model, one of the most classic cellular automata models, is employed to distinguish intelligent vehicles. In this paper, intelligent vehicles consist of adaptive cruise control (ACC) vehicles cooperative adaptive cruise control (CACC) vehicles. The ACC and CACC model are proposed by partners for advanced transportation technology (PATH), which are validated by real experimental data. Besides, vehicles equipped with CACC will degrade ACC vehicle if the leading vehicle is driven manually. From the perspective of vehicle's lateral movement, two novel lane-changing models, including the discretionary lane-change (DLC) model and mandatory lane-change (MLC) model, are developed to model the future behaviors of intelligent vehicles. A risk factor λ is introduced into the DLC model to distinguish vehicles from conventional ones. Based on environment perception technology, a five-step MLC decision-making model is designed specifically for intelligent vehicles exiting to off-ramp. It is comprised of environment perception, safe gap computation, measured gap ranking, measured gap classification and lane-changing gap selection. Based on the proposed hybrid traffic flow model, numerical simulations are conducted to study the influences of intelligent vehicles on the traffic flow near an off-ramp. Apart from the market penetration of intelligent vehicles, parameters considered in this paper include the demands of mainlines and off-ramp, range of environment perception, length of lane-changing area, and level of lane-changing risk.Analytical studies and simulation results are as follows. 1) The integration of car-following model and lane-changing model for the off-ramp system enables vehicles to have reasonable dynamic characteristics. 2) The capacity ascends to the peak after an initial decrease as CACC vehicle penetration increases. The maximum capacity obtained in 100% CACC vehicle scenario is improved by over 50%, compared with that in 50% CACC penetration scenario. 3) Enlarging the ranges of environment perception and lane-changing areas, and enhancing the lane-changing risk can significantly dissipate congestion upstream of the off-ramp and improve the efficiency of mainlines. However, they have little influence on traffic flow at off-ramp. 4) The worst performance of the system occurs in the scenario of 50% CACC penetration, where deterioration caused by degraded ACC vehicles suggests that enough patience and public confidence should be paid for the development of intelligent vehicles.
Keywords: intelligent vehicle/
lane-changing model/
freeway diverge segment/
adaptive cruise control
