Two-Vehicle Dynamics of the Car-Following Models on Realistic Driving Condition

• Published in: Journal of transportation systems engineering and information technology Vol. 12; no. 2; pp. 67 - 75
• Main Author: GUNAWAN, Fergyanto E
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.04.2012
• Subjects: car-following model; Delay; Drivers; Dynamic characteristics; Dynamic tests; Dynamics; intelligent driver model; intelligent transportation; micro-simulation; optimal velocity model; Optimization; Traffic engineering; Vehicles; car-following model; intelligent transportation; micro-simulation; intelligent driver model; optimal velocity model
• ISSN: 1570-6672
• DOI: 10.1016/S1570-6672(11)60194-3

The paper discusses the traffic dynamics in microscopic level and analyzes the dynamics characteristics of the traditional Gazis-Herman-Rothery model, the optimal velocity model with delay, and the intelligent driver model. An essential feature differentiating those models is that the traditional Gazis-Herman-Rothery model only governs the vehicle dynamics in the car-following state, but the other two models encompass larger interaction state including the free-flow state and the acceleration from the vehicle initial state. From this study, it can be concluded: (i) the optimal velocity model and intelligent driver model are more complete than the traditional model; (ii) the existing optimal velocity model may produce an unrealistic vehicle interaction; (iii) the optimal velocity model with a realistic delay can produce a stable interaction, and (iv) the intelligent driver model still needs further development particularly to take into account the driver delay which is an important aspect in the traffic dynamics on the microscopic level, and finally, (v) those three models may produce similar dynamics characteristics. 本文分析了交通微观动态性和传统GHR模型、考虑延误的最优速度模型和智能驾驶员模型的动态特性。传统GHR模型区别于其它两个模型的根本特征在于其仅描述车辆跟驰状态的关系，而其余两个模型则可较好地反映具有很强相互作用时车辆间关系，如自由流和车辆从初始状态开始加速的阶段。结果表明：(1) 最优速度模型和智能驾驶员模型较传统GHR模型具有更好的完整性；(2) 现有的最优速度模型可能产生不符合实际的车辆相互作用；(3) 考虑实际延误的最优速度模型可产生稳定的车辆间相互作用；(4) 智能驾驶员模型还需进一步改进，尤其是应考虑驾驶员延误这一微观交通仿真中的重要因素；(5) 这三个模型可产生相似的动态特性。