A car-following model considering asymmetric driving behavior based on long short-term memory neural networks

Huang, Xiuling; Sun, Jie; Sun, Jian

doi:10.1016/j.trc.2018.07.022

Cited by 185 publications

(120 citation statements)

References 53 publications

Supporting

Mentioning

117

Contrasting

Order By: Relevance

“…Previous studies suggested that the four factors (i.e., driver, vehicle, road, and environment) are considered as the main factors for estimating the minimal safety distance in the vehicle rear-end collision models [4,[12][13][14][15]. We followed the same rule when establishing our vehicle rear-end collision model.…”

Section: Analysis Of Influencing Factorsmentioning

confidence: 99%

“…x 11 , x 12 , and x 13 indicate the scores of drivers' age, fatigue levels, and driving behavior styles, respectively. x 21 , x 22 , and x 23 are the scores of the vehicle grade, service life ratio, and accident rate, respectively.…”

Section: Determination Of Correctionmentioning

confidence: 99%

“…However, the safety distance algorithm does not consider the driver's driving characteristics and actual road conditions; thus, the practical application of the warning system is low. Huang et al [12] trained the carfollowing model by using the LSTM (long short-term memory) neural network, but the network belongs to a deep neural network with multiple hidden layers. e structure of the network has too much target parameters and is complex.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Research of Vehicle Rear-End Collision Model considering Multiple Factors

Luo

Zang

Yuan

et al. 2020

Mathematical Problems in Engineering

View full text Add to dashboard Cite

e accuracy of the rear-end collision models is crucial for the early warning of potential traffic accident identification, and thus analyzes of the main factors influencing the rear-end collision relevant models is an active topic in the field. e previous studies have tried to determine the single factor influence on the rear-end collision model performance. Less attention was paid to exploit mutual influences on the model performance. To bridge the gap, we proposed an improved vehicle rear-end collision model by integrating varied factors which influence two parameters (i.e., response time and road adhesion coefficient). e two parameters were solved with the integrated weighting and neural network models, respectively. After that we analyzed the relationship between varied factors and the minimum car-following distance. e research findings support both the theoretical and practical guidance for transportation regulations to release more reasonable minimum headway distance to enhance the roadway traffic safety.On the basis of the traditional safety distance model, Kong et al. conducted information acquisition, distance judgment, braking, and other processes for active collision avoidance during vehicle travel. However, the judgment model used is relatively simple [5]. Jiang and Yu established a highway antitailing model based on radial basis function neural network by using traditional algorithms [6]. Nevertheless, the model is mainly applied to highways, and the influencing factors are considered to be single. Jing and Wang explored the influence of multiple vehicles on the driver and established a vehicle-following model considering the traffic state based on the full velocity difference (FVD) model [7]. Nonetheless, the model is too complicated to conduct to practical applications. Xia established a fuzzy inference system based on the characteristics (e.g., driving age, and gender) of the driver to determine the response time of different drivers and proposed an improved version of the safe model [8]. Peng and Sun derived the modified Korteweg-de Vries (mKdV) equation to describe the traffic

show abstract

Section: Analysis Of Influencing Factorsmentioning

confidence: 99%

Section: Determination Of Correctionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Research of Vehicle Rear-End Collision Model considering Multiple Factors

Luo

Zang

Yuan

et al. 2020

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…CF behaviour is a continuous behaviour, so a driver can make a corresponding decision based on the memory of the previous time period [13][14][15]. However, a large number of existing models do not fully consider the driver's memory effect and only consider the instantaneous interaction between the following and leading cars.…”

Section: Introductionmentioning

confidence: 99%

The Driver Time Memory Car-Following Model Simulating in Apollo Platform with GRU and Real Road Traffic Data

Fei

Hei

et al. 2020

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Car following is the most common phenomenon in single-lane traffic. The accuracy of acceleration prediction can be effectively improved by the driver’s memory in car-following behaviour. In addition, the Apollo autonomous driving platform launched by Baidu Inc. provides fast test vehicle following vehicle models. Therefore, this paper proposes a car-following model (CFDT) with driver time memory based on real-world traffic data. The CFDT model is firstly constructed by embedded gantry control unit storage capacity (GRU assisted) network. Secondly, the NGSIM dataset will be used to obtain the tracking data of small vehicles with similar driving behaviours from the common real road vehicle driving tracks for data preprocessing according to the response time of drivers. Then, the model is calibrated to obtain the driver’s driving memory and the optimal parameters of the model and structure. Finally, the Apollo simulation platform with high-speed automatic driving technology is used for 3D visualization interface verification. Comparative experiments on vehicle tracking characteristics show that the CFDT model is effective and robust, which improves the simulation accuracy. Meanwhile, the model is tested and validated using the Apollo simulation platform to ensure accuracy and utility of the model.

show abstract

“…Asymmetric driving behaviour possessing gap difference, reaction time (RT) difference, discrete driving (DD) difference, response intensity (RI) difference characteristic in accelerating and decelerating process under the car-following (CF) condition is reported by many researchers [1]- [3], which can affect the traffic flow significantly [4]. Thus, the performance of CF models in reproducing the asymmetric driving behaviour affects the numerical simulation results [5]. Many CF models [6], [7] based on simple assumptions cannot capture the realistic traffic flow characteristics as well as that considering asymmetric driving behaviour [3], [5], [8]- [11].…”

Section: Introductionmentioning

confidence: 99%

Asymmetric Driving Behaviour Analysis Using Field Trajectories

Huang

Sun

2019

WIT Transactions on the Built Environment

Self Cite

View full text Add to dashboard Cite

Asymmetric driving behaviour in acceleration and deceleration processes under car-following (CF) state can affect traffic flow significantly. Thus, to improve the performance of CF models, the formation mechanism of the asymmetric driving characteristics (gap difference, reaction time difference, response intensity difference, discrete driving difference) and the internal relations between them should be investigated. Thus the quantification methods are proposed to study the asymmetry for each driving characteristic. Then the mean values of the gap, reaction time, discrete intensity, and response intensity at each velocity fragment in acceleration and deceleration process are measured respectively using the NGSIM data with the proposed quantification methods. By comparing these mean values in acceleration and in deceleration processes, the asymmetry of the gap, reaction time, discrete intensity and response intensity are validated. Moreover, the correlation analysis between the four asymmetric driving characteristics are implemented while the results indicate they influence each other, and the formation mechanisms of the four asymmetric driving behaviours are obtained.

show abstract

A car-following model considering asymmetric driving behavior based on long short-term memory neural networks

Cited by 185 publications

References 53 publications

Research of Vehicle Rear-End Collision Model considering Multiple Factors

Research of Vehicle Rear-End Collision Model considering Multiple Factors

The Driver Time Memory Car-Following Model Simulating in Apollo Platform with GRU and Real Road Traffic Data

Asymmetric Driving Behaviour Analysis Using Field Trajectories

Contact Info

Product

Resources

About