scholarly journals Modeling Car-Following Behaviors and Driving Styles with Generative Adversarial Imitation Learning

Sensors ◽  
2020 ◽  
Vol 20 (18) ◽  
pp. 5034
Author(s):  
Yang Zhou ◽  
Rui Fu ◽  
Chang Wang ◽  
Ruibin Zhang
Keyword(s):  
Autonomous Driving ◽  
Imitation Learning ◽  
Driver Model ◽  
Area Network ◽  
Car Following ◽  
Time Headway ◽  
Reward Function ◽  
Proposed Model ◽  
Wave Radar ◽  
Car Following Model

Building a human-like car-following model that can accurately simulate drivers’ car-following behaviors is helpful to the development of driving assistance systems and autonomous driving. Recent studies have shown the advantages of applying reinforcement learning methods in car-following modeling. However, a problem has remained where it is difficult to manually determine the reward function. This paper proposes a novel car-following model based on generative adversarial imitation learning. The proposed model can learn the strategy from drivers’ demonstrations without specifying the reward. Gated recurrent units was incorporated in the actor-critic network to enable the model to use historical information. Drivers’ car-following data collected by a test vehicle equipped with a millimeter-wave radar and controller area network acquisition card was used. The participants were divided into two driving styles by K-means with time-headway and time-headway when braking used as input features. Adopting five-fold cross-validation for model evaluation, the results show that the proposed model can reproduce drivers’ car-following trajectories and driving styles more accurately than the intelligent driver model and the recurrent neural network-based model, with the lowest average spacing error (19.40%) and speed validation error (5.57%), as well as the lowest Kullback-Leibler divergences of the two indicators used for driving style clustering.

scholarly journals The Research of Car-Following Model Based on Real-Time Maximum Deceleration

2015 ◽  
Vol 2015 ◽  
pp. 1-9 ◽  
Author(s):  
Longhai Yang ◽  
Xiqiao Zhang ◽  
Jiekun Gong ◽  
Juntao Liu
Keyword(s):  
Real Time ◽  
Vehicle Dynamics ◽  
Driver Model ◽  
Cruise Control ◽  
Maximum Acceleration ◽  
Car Following ◽  
Time Headway ◽  
Asphalt Road ◽  
Car Following Model ◽  
Maximum Deceleration

This paper is concerned with the effect of real-time maximum deceleration in car-following. The real-time maximum acceleration is estimated with vehicle dynamics. It is known that an intelligent driver model (IDM) can control adaptive cruise control (ACC) well. The disadvantages of IDM at high and constant speed are analyzed. A new car-following model which is applied to ACC is established accordingly to modify the desired minimum gap and structure of the IDM. We simulated the new car-following model and IDM under two different kinds of road conditions. In the first, the vehicles drive on a single road, taking dry asphalt road as the example in this paper. In the second, vehicles drive onto a different road, and this paper analyzed the situation in which vehicles drive from a dry asphalt road onto an icy road. From the simulation, we found that the new car-following model can not only ensure driving security and comfort but also control the steady driving of the vehicle with a smaller time headway than IDM.

Influences of vehicles’ fuel consumption and exhaust emissions on the trip cost without late arrival under car-following model

2016 ◽  
Vol 27 (01) ◽  
pp. 1650011 ◽  
Author(s):  
Tie-Qiao Tang ◽  
Qiang Yu
Keyword(s):  
Fuel Consumption ◽  
Exhaust Emissions ◽  
Optimal Time ◽  
Fuel Cost ◽  
Total Cost ◽  
Car Following ◽  
Time Headway ◽  
Late Arrival ◽  
Car Following Model ◽  
Open Road

In this paper, we use car-following model to explore the influences of the vehicle’s fuel consumption and exhaust emissions on each commuter’s trip cost without late arrival on one open road. Our results illustrate that considering the vehicle’s fuel cost and emission cost only enhances each commuter’s trip cost and the system’s total cost, but has no prominent impacts on his optimal time headway at the origin of each open road under the minimum total cost.

scholarly journals Investigating the Impact of Connected and Automated Vehicles on Signalized and Unsignalized Intersections Safety in Mixed Traffic

2022 ◽  
Vol 2 (1) ◽  
pp. 24-40
Author(s):  
Amirhosein Karbasi ◽  
Steve O’Hern
Keyword(s):  
Case Studies ◽  
Human Error ◽  
Road Traffic ◽  
Driver Model ◽  
Cruise Control ◽  
Driver Assistance Systems ◽  
Road Traffic Crashes ◽  
Car Following ◽  
Car Following Model ◽  
The Impact

Road traffic crashes are a major safety problem, with one of the leading factors in crashes being human error. Automated and connected vehicles (CAVs) that are equipped with Advanced Driver Assistance Systems (ADAS) are expected to reduce human error. In this paper, the Simulation of Urban MObility (SUMO) traffic simulator is used to investigate how CAVs impact road safety. In order to define the longitudinal behavior of Human Drive Vehicles (HDVs) and CAVs, car-following models, including the Krauss, the Intelligent Driver Model (IDM), and Cooperative Adaptive Cruise Control (CACC) car-following models were used to simulate CAVs. Surrogate safety measures were utilized to analyze CAVs’ safety impact using time-to-collision. Two case studies were evaluated: a signalized grid network that included nine intersections, and a second network consisting of an unsignalized intersection. The results demonstrate that CAVs could potentially reduce the number of conflicts based on each of the car following model simulations and the two case studies. A secondary finding of the research identified additional safety benefits of vehicles equipped with collision avoidance control, through the reduction in rear-end conflicts observed for the CACC car-following model.

Conformity and stability analysis of a modified spring–mass–damper system dynamics-based car-following model

2019 ◽  
Vol 33 (06) ◽  
pp. 1950025 ◽  
Author(s):  
Caleb Ronald Munigety
Keyword(s):  
Stability Analysis ◽  
System Dynamics ◽  
Traffic System ◽  
Car Following ◽  
Proposed Model ◽  
Mass Damper ◽  
Traffic Stream ◽  
Behavioral Parameters ◽  
Car Following Model ◽  
The Stability

Modeling the dynamics of a traffic system involves using the principles of both physical and social sciences since it is composed of vehicles as well as drivers. A novel car-following model is proposed in this paper by incorporating the socio-psychological aspects of drivers into the dynamics of a purely physics-based spring–mass–damper mechanical system to represent the driver–vehicle longitudinal movements in a traffic stream. The crux of this model is that a traffic system can be viewed as various masses interacting with each other by means of springs and dampers attached between them. While the spring and damping constants represent the driver behavioral parameters, the mass component represents the vehicle characteristics. The proposed model when tested for its ability to capture the traffic system dynamics both at micro, driver, and macro, stream, levels behaved pragmatically. The stability analysis carried out using perturbation method also revealed that the proposed model is both locally and asymptotically stable.

A REVISED STOCHASTIC OPTIMAL VELOCITY MODEL CONSIDERING THE VELOCITY GAP WITH A PRECEDING VEHICLE

2011 ◽  
Vol 22 (09) ◽  
pp. 1005-1014 ◽  
Author(s):  
KEIZO SHIGAKI ◽  
JUN TANIMOTO ◽  
AYA HAGISHIMA
Keyword(s):  
Velocity Model ◽  
Model Parameters ◽  
Fundamental Diagram ◽  
Cellular Automata Model ◽  
Optimal Velocity ◽  
Car Following ◽  
Proposed Model ◽  
Preceding Vehicle ◽  
Long Time ◽  
Car Following Model

The stochastic optimal velocity (SOV) model, which is a cellular automata model, has been widely used because of its good reproducibility of the fundamental diagram, despite its simplicity. However, it has a drawback: in SOV, a vehicle that is temporarily stopped takes a long time to restart. This study proposes a revised SOV model that suppresses this particular defect; the basic concept of this model is derived from the car-following model, which considers the velocity gap between a particular vehicle and the preceding vehicle. A series of simulations identifies the model parameters and clarifies that the proposed model can reproduce the three traffic phases: free, jam, and even synchronized phases, which cannot be achieved by the conventional SOV model.

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

2020 ◽  
Vol 2020 ◽  
pp. 1-18
Author(s):  
Rong Fei ◽  
Shasha Li ◽  
Xinhong Hei ◽  
Qingzheng Xu ◽  
Fang Liu ◽  
...  
Keyword(s):  
High Speed ◽  
Road Traffic ◽  
3D Visualization ◽  
Control Unit ◽  
Autonomous Driving ◽  
Simulation Platform ◽  
Traffic Data ◽  
Simulation Accuracy ◽  
Car Following ◽  
Car Following Model

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.

scholarly journals Stability Analysis of Train Movement with Uncertain Factors

2015 ◽  
Vol 2015 ◽  
pp. 1-7 ◽  
Author(s):  
JingJing Ye ◽  
KePing Li ◽  
XueDong Jiang
Keyword(s):  
Relaxation Time ◽  
Stability Analysis ◽  
Fuzzy Theory ◽  
Traffic Model ◽  
Optimal Velocity ◽  
Car Following ◽  
Proposed Model ◽  
Car Following Model ◽  
Simulation Results ◽  
The Stability

We propose a new traffic model which is based on the traditional OV (optimal velocity) car-following model. Here, some realistic factors are regarded as uncertain quantity, such as the headway distance. Our aim is to analyze and discuss the stability of car-following model under the constraint of uncertain factors. Then, according to the principle of expected value in fuzzy theory, an improved OV traffic model is constructed. Simulation results show that our proposed model can avoid collisions effectively under uncertain environment, and its stability can also be improved. Moreover, we discuss its stability as some parameters change, such as the relaxation time.

AN EXTENDED MACROSCOPIC MODEL FOR TRAFFIC FLOW ON A HIGHWAY WITH SLOPES

2013 ◽  
Vol 24 (09) ◽  
pp. 1350061 ◽  
Author(s):  
JIANZHONG CHEN ◽  
ZHONGKE SHI ◽  
YANMEI HU ◽  
LEI YU ◽  
YUAN FANG
Keyword(s):  
Traffic Flow ◽  
Small Perturbation ◽  
Gravitational Force ◽  
Slope Angle ◽  
Macroscopic Model ◽  
Car Following ◽  
Proposed Model ◽  
Car Following Model ◽  
Simulation Results ◽  
The Relationship

In this paper, we present an extended car-following model with consideration of the gravitational force. A new macroscopic model taking into account the slope effects is developed using the relationship between the microscopic and macroscopic variables. The proposed model is applied to reflect the effect of the slope on uniform flow, traffic waves and small perturbation. The simulation results demonstrate that both the angle and the length of the slope have important impacts on traffic flow. The effect of the slope becomes more significant with the increase of the slope angle.

Incorporating Instantaneous Reaction Delay in Car-Following Models: A Hybrid Approach

2021 ◽  
pp. 036119812110152
Author(s):  
Mehdi Rafati Fard ◽  
Saeed Rahmani ◽  
Afshin Shariat Mohaymany
Keyword(s):  
Reaction Time ◽  
Road Safety ◽  
Statistical Tests ◽  
Hybrid Approach ◽  
Car Following ◽  
Driving Behaviors ◽  
Proposed Model ◽  
Traffic Performance ◽  
Fixed Parameter ◽  
Car Following Model

Car-following is considered as one of the most prevalent fundamental driving behaviors that substantially influences traffic performance as well as road safety and capacity. Drivers’ car-following behavior is affected by numerous factors. However, in practice, very few of these factors have been scrutinized, because of their latent essence and unavailability of appropriate data. Owing to its importance, drivers’ reaction time has attracted the attention of many researchers; nevertheless, it is considered as a fixed parameter in car-following models, which is far from reality. To take the variability of drivers’ reaction time into account, a flexible hybrid approach has been suggested in the present study. In the proposed structure, in the first step, the desirable acceleration of the driver is estimated by applying an equation-based car-following model. In the next step, the driver’s reaction delay in applying the calculated acceleration is estimated by an artificial neural network. The corresponding parameters are jointly estimated by applying an estimated distribution algorithm. Statistical tests indicate better performance of the hybrid model, which considers the variations of the driver’s reaction time, compared with a traditional model with fixed reaction time. Furthermore, the cross-validation results indicate better generalizability and transferability of the proposed model in action.

Sign in / Sign up

Export Citation Format

Share Document