Lane-changing Model Research Articles

Capacity Drop at Freeway Ramp Merges with Its Replication in Macroscopic and Microscopic Traffic Simulations: A Tutorial Report

Capacity drop (CD) at overloaded bottlenecks is a puzzling traffic flow phenomenon with some internal and complicated mechanisms at the microscopic level. Capacity drop is not only important for traffic flow theory and modelling, but also significant for traffic control. A traffic model evaluating traffic control measures needs to be able to reproduce capacity drop in order to deliver reliable evaluation results. This paper delivers a comprehensive overview on the subject from the behavioral mechanism perspective, as well as from microscopic and macroscopic simulation points of view. The paper also conducts comparable studies to replicate capacity drop at freeway ramp merges from both macroscopic and microscopic perspectives. Firstly, the subject is studied using the macroscopic traffic flow model METANET with respect to ramp merging scenarios with and without ramp metering. Secondly, one major weakness of commercial microscopic traffic simulation tools in creating capacity drop at ramp merges is identified and a forced lane changing model for ramp-merging vehicles is studied and incorporated into the commercial traffic simulation tool AIMSUN. The extended AIMSUN carefully calibrated against real data is then examined for its capability of reproducing capacity drop in a complicated traffic scenario with merging bottlenecks. The obtained results demonstrate that reproducible capacity drop can be delivered for the targeted bottlenecks using both macroscopic and microscopic simulation tools.

Modelling autonomous vehicle interactions with bicycles in traffic simulation

With the undeniable advance of automated vehicles and their gradual integration in day-today urban traffic, many new technologies have been developed that offer great potential for this emerging field of research. However, testing automated vehicle technologies in real road traffic with vulnerable road users (VRUs) is still a complicated and time consuming procedure. The virtual development and evaluation of automated vehicles using simulation tools offers a good opportunity to test new functions efficiently. However, existing models prove to be insufficient in modeling the interaction between autonomous vehicles and vulnerable road users such as bicyclists and pedestrians. In this paper, an automated vehicle model for microscopic traffic simulation tools is developed with the open-source traffic simulation software Simulation of Urban Mobility (SUMO) and evaluated in common traffic interaction scenarios with bicyclists. A controller model is proposed using different path-finding algorithms from the field of robotic and automated vehicle research, which covers all important control levels of a self-driving vehicle. Finally, its performance is compared to existing car-following and lane-changing models. Results showcase that the autonomous vehicle model achieves either comparable results or has a much steadier and more realistic driving behavior when compared to existing driving models while interacting with bicyclists. The whole source code developed over the course of this work is freely accessible at: https://github.com/FlixFix/Kimarite.

The use of microscopic simulators for safety assessment in automated and partially automated scenarios: a comparison

In the field of safety assessments, the use of simulators to predict unknown scenarios is becoming crucial, due to the ability of predicting conflict points. The microscopic simulators, accountable for modeling single-vehicle interactions, differ for the type of car-following and lane-changing model, hence for the number of definable parameters. The possibility of modifying all the parameters inside the model equations makes them suitable for simulating autonomous vehicles too. To evaluate the road safety of a site, the output of traffic simulators can become the input of the SSAM (Surrogate Safety Assessment Model), which relies on surrogate safety measures (SSMs). Thus, the choice of the simulator is fundamental. The focus of this dissertation concerns the comparison of the car-following and lane-changing models at the basis of the most used commercial microscopic simulators (VISSIM, AIMSUN, PARAMICS) and the related SSAM outputs in the automated and partially automated traffic environment. The Gipps Model, valid for the car-following and the lane-changing model, aims at always achieving the safety distance, intervening on kinematic parameters like the acceleration, speed, and distance. It is possible to intervene on a limited number of parameters of this model. The Wiedemann model belongs to the family of psycho-physical one and it is governed by several parameters, which makes it useful to remodel complex situations, simulating the human mental workload in driving. The Gipps model and the Wiedemann one (implemented respectively in AIMSUN and VISSIM) have already been used for safety purposes, while the Fritszche one (implemented in PARAMICS) has been currently used only for environmental issues related to automation. It is possible to modify up to 16 parameters in the Gipps and Wiedemann model to simulate the automated and partially autonomous vehicles, even if these models belong to different car-following model family. Based on the site variables, each simulator showed potential strengths. Scenarios involving automated vehicles, which do not account for human behaviors, might be better represented by the Gipps model, which aims at reproducing only the safety distance and not the human mental behavior. Thus, this model could also be the most suitable for the general purpose of crash analysis above mentioned.

A Vehicle-to-Vehicle Communication-based Integrated Car-Following and Lane-Changing Model using A Deep Learning Approach

A Vehicle-to-Vehicle Communication-based Integrated Car-Following and Lane-Changing Model using A Deep Learning Approach

Lane-changing decision rule with the difference of traffic flow's variation in multi-lane highway for connected and autonomous vehicles

Abstract Drivers are not far-sighted when they execute lane-changing manipulation. To address this issue, this study proposes a rule to improve vehicles' lane-changing decisions with accurate information of surrounding vehicles (e.g. time headway). More specifically, connected and autonomous vehicles (CAVs) change lanes in advance if they find severer flow reducing in the lanes, while CAVs should maintain the car-following state if the variations of traffic flow in all lanes have a similar trend. To illustrate the idea, this study first calibrates two classic car-following models and a lane-changing model, and then conducts numerical simulations to illustrate the short-sighted decision of drivers. The study incorporates the idea into a lane-changing decision rule by changing the lane-changing model's parameter, and conducts numerical tests to evaluate the effectiveness of the lane-changing decision rule in a multi-lane highway with a bottleneck. The results of this study indicate that the new lane-changing decision rule can substantially improve the throughput of the traffic flow, especially when the inflow exceeds the remaining capacity of the road. The lane-changing rule and results can bring insights into the control of CAVs, as well as the driver assistance system in connected vehicles.

Strategy of lane-changing coupling process for connected and automated vehicles in mixed traffic environment

Heterogeneous traffic agents consisting of human-driven vehicle (HDV) and connected-automated vehicle (CAV) appear on the road and form a mixed traffic environment. When combining the characteristics of CAVs and HDVs, it's still an open issue of investigating how heterogeneous traffic agents will execute lane-changing. With respect to the coupling process in the lane-changing behaviour of vehicles, this paper derives the lane changing optimization method for CAVs under different penetration rates. This method captures not only the full spectrum of CAV penetration rates but also all types of vehicle platoon. Firstly, the different car-following and lane-changing models are used to capture the characteristics of different types of vehicles. Moreover, For CAV, the Cooperative Adaptive Cruise Control (CACC) vehicle would degrade into the Adaptive Cruise Control (ACC) vehicle due to communication failure. Secondly, the framework for the coupling analysis of CAV lane-changing behaviour and mixed traffic stability is derived. Then, this study proposes and examines optimal control strategies based on numerical simulation. These consist of a longitudinal phase, in which the proposed strategies improve the operational efficiency of the overall mixed traffic flow, and a lateral phase, in which they safely change lanes. The results show that they improve the comfort, efficiency, safety, and stability of the heterogeneous traffic agents compared to the maneuvers performed by the baseline. This method can serve as a useful and simple decision tool for future CAV lane management.

Attention-Based Lane Change and Crash Risk Prediction Model in Highways

Lane change and crash risk prediction are critical technologies for autonomous driving. An attention-based LSTM model is proposed in this paper for lane change behavior prediction in highways, considering both the surrounding vehicles’ information around the target vehicle at the current time and the historical trajectory data of the vehicle, and it shows higher accuracy and better interpretability than other models. There are two sections in this prediction model, including a pre-judgment model based on the C4.5 decision tree and bagging ensemble learning, and a multi-step lane change prediction model of LSTM with attention mechanism. In addition, the model is applied in the NGSIM datasets to test practicability and accuracy, and the precision of left lane change and right lane change is 98% and 94% respectively 1 second before the vehicle changes lane. Finally, to judge the safety of vehicles in the driving process, a crash risk prediction model based on Time-To-Collision is proposed, which further verifies the effectiveness of the method.

A cooperative energy efficient truck platoon lane-changing model preventing platoon decoupling in a mixed traffic environment

Truck platooning has gained increasing attention due to the benefits in energy and operation efficiency in freight transportation. One significant challenge for deploying truck platoons is the safe and efficient interaction with surrounding traffic, especially at freeway discontinuities where mandatory lane changes usually lead to the decoupling of truck platoons. This study proposes a cooperative truck platoon lane-changing model (CTPLC) to prevent the decoupling of truck platoons in a mixed traffic environment. Specifically, a two-step control strategy is presented, where vehicles in the target lane firstly cooperatively adjust speeds to create an appropriate gap for a truck platoon, and then trucks within the truck platoon conduct lane change sequentially. The cooperative speed profiles are generated by solving an optimization problem considering the lane-changing influence and energy consumption. Based on that, a two-dimensional nonlinear model predictive control (MPC) algorithm is employed to generate vehicular acceleration and steering angle for each truck. A series of numerical simulation experiments were conducted to validate the proposed strategy. As shown by the results, our proposed method truck platoon could conduct a lane change in a traffic-efficient and safe manner, and meanwhile, our method was more energy-efficient than a benchmark strategy.

Analysis of Lane-Changing Decision-Making Behavior of Autonomous Vehicles Based on Molecular Dynamics.

Along with the rapid development of autonomous driving technology, autonomous vehicles are showing a trend of practicality and popularity. Autonomous vehicles perceive environmental information through sensors to provide a basis for the decision making of vehicles. Based on this, this paper investigates the lane-changing decision-making behavior of autonomous vehicles. First, the similarity between autonomous vehicles and moving molecules is sought based on a system-similarity analysis. The microscopic lane-changing behavior of vehicles is analyzed by the molecular-dynamics theory. Based on the objective quantification of the lane-changing intention, the interaction potential is further introduced to establish the molecular-dynamics lane-changing model. Second, the relationship between the lane-changing initial time and lane-changing completed time, and the dynamic influencing factors of the lane changing, were systematically analyzed to explore the influence of the microscopic lane-changing behavior on the macroscopic traffic flow. Finally, the SL2015 lane-changing model was compared with the molecular-dynamics lane-changing model using the SUMO platform. SUMO is an open-source and multimodal traffic experimental platform that can realize and evaluate traffic research. The results show that the speed fluctuation of autonomous vehicles under the molecular-dynamics lane-changing model was reduced by 15.45%, and the number of passed vehicles was increased by 5.93%, on average, which means that it has better safety, stability, and efficiency. The molecular-dynamics lane-changing model of autonomous vehicles takes into account the dynamic factors in the traffic scene, and it reasonably shows the characteristics of the lane-changing behavior for autonomous vehicles.

Lane-Changing Model of Intelligent Connected Vehicle Considering the Factor of Turn Signal

Unsafe lane-changing behaviors can easily lead to traffic accidents. Drivers usually turn on their turn signals to signal surrounding vehicles before changing lanes. At present, there is a lack of consideration of the impact of turn signals on the lane-changing behavior of intelligent vehicles. Therefore, based on the cellular automata theory, this paper improves the lane-changing rules in the STNS model and proposes a vehicle safe lane-changing model. The model considers the priority scheduling problem of different vehicles’ driving behavior when changing lanes, the influence of driver’s subjective factors on the driving speed when changing lanes, and the relationship between vehicle speed and safe lane-changing distance. After discussion and analysis, the model can reduce the number of lane changes of vehicles, increase the average speed of vehicles, and increase the traffic flow. It provides theoretical support for the safe lane-changing behavior of intelligent networked vehicles in the new era.

Development of a cyber-physical-system perspective based simulation platform for optimizing connected automated vehicles dedicated lanes

Recently, many studies have shown that during the transition period of the coexistence of autonomous vehicles (AVs)/connected automated vehicles (CAVs) and human-driven vehicles (HDVs), a dedicated lane (DL) for AVs/CAVs can improve traffic throughput and lower collision risk, which has become a hot topic in the field of intelligent transportation systems (ITS). This study presents an integrated simulation platform for DL optimization from cyber-physical-system (CPS) perspective. The developed simulation platform can be used to qualitatively and quantitatively evaluate the design parameters, operation modes, contingency plan, etc. of DL in a “What You See Is What You Get (Wysiwyg)” style. First, we introduce the architecture and features of the proposed platform, which comprises of SUMO, OMNeT++, and Unity3D. We implemented all experimental scripts for DL optimization in SUMO, including the configuration of the DLs, car-following models, lane-changing models, vehicle departure models, origin–destination demands, and 2D visualization of traffic flow simulation. We used OMNeT++ for vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication. Unity3D is a 3D visualization platform for simulations synchronized with SUMO. Connected with the driving simulator and paired with a virtual reality (VR) helmet, Unity3D can offer a high-fidelity immersive environment for all types of users. We developed a specific communication protocol to achieve data sharing and synchronization of all components of the proposed simulation platform. Second, this paper depicts the functions of the proposed simulation platform in detail, including the modeling of road networks, vehicles, traffic flow, optimal design, and evaluation indicators of DL. Finally, by using the developed simulation platform, we demonstrated and analyzed three cases: (1) without DL vs with DL, (2) inner lane vs outer lane, and (3) without V2I vs with V2I. The experimental results confirm the feasibility of the developed simulation platform, and yield some interesting findings, such as (1) market penetration rates (MPRs) are an important factor in deciding whether to deploy DLs, (2) the number of lane changes when the vehicle enters/leaves the designated lane is the key criterion for determining the placement of the DLs, and (3) the cooperative vehicle-infrastructure system can offer AVs/CAVs with better over-the-horizon perception capabilities, helping to improve the smoothness of traffic flow in the case of abnormal events. The developed simulation platform is expected to offer a new tool for optimal design and evaluation of DLs for AVs/CAvs.

Analysis of Lane-Changing Decision-Making Behavior and Molecular Interaction Potential Modeling for Connected and Automated Vehicles

With the technical support of an intelligent networking environment, autonomous driving technology is facing a new stage of development, and the decision-making behavior of autonomous vehicles is changing fundamentally, so it is urgent to explore the lane-changing decision-making behavior mechanism of autonomous driving. Firstly, through the analysis of system similarity, the similarity between autonomous vehicles and moving molecules is sought, and the attraction and repulsion between molecules are applied to the lane-changing process of vehicles to effectively recognize the traffic scene of lane-changing vehicles. Secondly, the molecular interaction potential is introduced to unify the attraction and repulsion, and explore the dynamic influencing factors of lane-changing behavior for vehicles. Moreover, we systematically analyze the interaction relationship in the lane-changing process of Connected and Automated Vehicles, and establish the molecular interaction potential lane-changing model to explore the lane-changing decision-making behavior mechanism. Furthermore, we study the impact of micro lane-changing behavior on macro traffic flow. Finally, the SL2015 lane-changing model and the molecular interaction potential lane-changing model are compared and analyzed by using the SUMO platform. The results show that the speed fluctuation of Connected and Automated Vehicles based on the molecular interaction potential lane-changing model is reduced by 15.5%, and the number of passed vehicles is increased by 3.26% on average, which has better safety, stability, and efficiency. The molecular interaction potential modeling of lane-changing decision-making behavior for Connected and Automated Vehicles comprehensively considers the interaction relationship of dynamic factors in the traffic environment, and scientifically shows the lane-changing decision-making mechanism of Connected and Automated Vehicles.

Lane change behavior with uncertainty and fuzziness for human driving vehicles and its simulation in mixed traffic

In the process of constructing the driver’s lane-changing model, in addition to the dynamic states of the surrounding vehicles affecting the driver’s lane-changing decision, various traffic environment factors and the driver’s own personality also affect the driver’s lane-changing probability. At the same time, these influencing factors also make the lane changing process ambiguous and uncertain. To address these issues, in this study the membership function of soft set is applied to quantify the impact of some typical ambiguous factors and relationships, such as weather conditions, driver personality, and vehicle type, on the driver’s driving process. The mapping relationships between these influencing factors and the driver’s expected speed and aggression are obtained through decision rules. Then a new lane changing model considering uncertain or ambiguous factors and relationships is proposed. On the basis of this model, simulation experiments of mixed traffic flow under the conditions of 0%, 20%, 40% and 60% penetration rate of autonomous vehicles were carried out. The experimental results reasonably present the impact of autonomous vehicle penetration and some ambiguity on the lance change behavior.

Deep Reinforcement Learning Approach for Automated Vehicle Mandatory Lane Changing

This paper proposes a reinforcement learning-based framework for mandatory lane changing of automated vehicles in a non-cooperative environment. The objective is to create a reinforcement learning (RL) agent that is able to perform lane-changing maneuvers successfully and efficiently and with minimal impact on traffic flow in the target lane. For this purpose, this study utilizes the double deep Q-learning algorithm structure, which takes relevant traffic states as input and outputs the optimal actions (policy) for the automated vehicle. We put forward a realistic approach for dealing with this problem where, for instance, actions selected by the automated vehicle include steering angles and acceleration/deceleration values. We show that the RL agent is able to learn optimal policies for the different scenarios it encounters and performs the lane-changing task safely and efficiently. This work illustrates the potential of RL as a flexible framework for developing superior and more comprehensive lane-changing models that take into consideration multiple aspects of the road environment and seek to improve traffic flow as a whole.

A model of lane-changing intention induced by deceleration frequency in an automatic driving environment

Achieving high-level, anthropomorphic lane-changing control for autonomous vehicles is an important method for promoting the safe and efficient operation of autonomous vehicles. According to existing lane-changing models, the important phenomenon that frequent acceleration and deceleration cause a poor riding experience and thus induce the intention of lane-changing needs to be given more attention. In this study, a discretionary lane-changing control method for autonomous vehicles is designed based on anti-interference ability (AIA, where AIA is the number of decelerations per unit time of an autonomous vehicle) to extract the traffic flow variation characteristics of autonomous vehicles with AIA and to determine methods to improve the driving efficiency and stability of autonomous vehicles. AIA is introduced to the lane-changing decision-making process of autonomous vehicles to simulate the phenomenon of lane-changing intention caused by poor riding experience. A control model that can evaluate the lane-changing conditions is then proposed. NetLogo software was selected to construct a one-way, two-lane scenario for autonomous vehicle operation, and the impacts of the proposed decision-making method in each scenario were tested. The experimental results show that autonomous vehicles are more likely to produce frequent deceleration behaviors under large-traffic and small-space operating conditions. A traffic flow composed of autonomous vehicles that use AIA to induce lane-changing intentions has a more obvious phase transition process. An autonomous vehicle whose AIA induces lane-changing intention can improve the speed and stability for a certain traffic volume (a saturation of less than approximately 0.45). Although lane-changing behavior breaks the stable state of two-lane traffic flow, it will also successfully suppress the aggravation of traffic congestion. Lane-changing has an impact on the local oscillation of traffic flow, and a stable traffic flow operation state and reduction in the AIA clearing frequency are mutually reinforcing.

DEVELOPMENT OF LANE CHANGE MODELS THROUGH MICROSCOPIC SIMULATION UNDER MIXED TRAFFIC

The number of motor vehicles and traffic demand is growing in tandem with society's rapid economic development and the quickening process of urbanization. In India, traffic congestion has become the most serious issue. Vehicle's poor lane-changing behaviors have a significant impact on the speed of the traffic system. The lane change behavior was observed in the study with the traffic flow simulation model VISSIM. The purpose of this investigation is to estimate the number of lane changes by simulating them at different volume levels. The study also shows the capacity estimate and its relationship with the lane change. The maximum traffic flow and lane changes were calculated based on the proportions of each vehicle type in the standard car traffic flow. The maximum number of lane change models for known traffic compositions on 4-lane, 6-lane, and 8-lane divided highways has been developed. According to the results, the number of observed lane changes depends on the volume of traffic and the number of lanes provided for a particular direction of travel. The composition and vehicle types also have a significant influence on lane changes and highway capacity, which involves uncertainties in the analysis of traffic flow properties.

Active Lane-Changing Control of Intelligent Vehicle on Curved Section of Expressway

In order to improve the intelligent vehicle lane-changing performance, an active lane-changing control algorithm is proposed considering the changes of road curvature and vehicle speed. Firstly, the vehicle dynamics model considering vehicle speed variation and lane-changing safety distance is established, and the expected lane-changing trajectory model under the curved road is designed simultaneously. Then, taking the yaw rate and longitudinal speed as the control objectives of lateral and longitudinal motions, respectively, the sliding-mode variable structure control method based on Lyapunov stability condition is adopted, and the trajectory tracking controller is designed by combining the inverted method to track the desired lane-changing trajectory. Finally, the lane-changing trajectory model and trajectory tracking controller are verified in simulation platform of CarSim/Simulink and hardware-in-the-loop (HIL) test bench. The results show that the proposed trajectory tracking control method can perform the lane-changing behavior well under different road curvatures and vehicle speeds while maintaining high trajectory tracking control accuracy.

A Novel Dynamic Lane-Changing Trajectory Planning Model for Automated Vehicles Based on Reinforcement Learning

Lane changing behavior has a significant impact on traffic efficiency and may lead to traffic delays or even accidents. It is important to plan a safe and efficient lane-changing trajectory that coordinates with the surrounding environment. Most conventional lane-changing models need to establish and solve constrained optimization models during the whole process, while reinforcement learning can just take the current state as input and directly output actions to vehicles. This study develops a lane-changing model using the deep deterministic policy gradient method, which can simultaneously control the lateral and longitudinal motions of the vehicle. To optimize its performance, a reward function is properly designed by combining safety, efficiency, gap, headway, and comfort features. To avoid collisions, a safety modification model is developed to check and correct acceleration at every time step. The driving trajectory data of 1169 lane-changing scenarios extracted from the Next Generation Simulation (NGSIM) dataset are used to train and test the model. The proposed model can quickly converge in training phase. Testing results show it can complete safe and efficient lane changing in different lane-changing scenarios with both shorter time headway and lane-changing duration than human drivers. Compared with the conventional dynamic lane-changing trajectory planning model, our model can reduce collision risk. It is also evaluated in automated and nonautomated mixed traffic in SUMO. Simulation results show that the proposed model also has a positive effect on the average speed of overall traffic flow.

Research on human-like lane-changing control of intelligent vehicles based on extension goodness evaluation

In order to improve the trajectory tracking accuracy and passenger comfort for intelligent vehicle lane-changing at medium and high speed, this paper proposes a new control algorithm based on extension evaluation system for intelligent vehicle lane-changing. In this research, firstly, a vehicle-road dynamics model has been developed. In addition, the actual driving trajectory data of skilled drivers are collected by driving simulator, and the lane-changing path planning model is established based on generalized regression neural network (GRNN). Last, in order to improve the tracking ability to the planning path, the trajectory tracking controller is established. A two-layer structure of lateral trajectory tracking control system is designed, in which the upper layer includes two algorithms: (1) PID feedback control based on preview deviation; (2) PID feedforward-feedback control based on the curvature of the path planning by skilled drivers. While the extension goodness evaluation is introduced in the lower layer which is used to evaluate two controllers in upper layer. According to the real-time state of vehicle-road system, the controller output with higher goodness is selected. Finally, the simulation results verify that comparing with the conventional single PID feedback control, the extension goodness evaluation control can effectively improve the passenger comfort while ensuring tracking accuracy.

Modeling pedestrian–vehicle mixed-flow in a complex evacuation scenario

Traffic simulation model has been widely used in urban traffic evacuation and pedestrian evacuation. However, few research efforts have been devoted to a complex evacuation scenario that includes vehicles and pedestrians. Even some of the research studies have focused on pedestrian–vehicle mixed-flow simulation models; they can hardly simulate the pedestrian–vehicle interaction dynamics in complex evacuation scenarios such as intersections, parking lots, and so on. This paper developed a pedestrian–vehicle mixed-flow model based on the Social Force Model (SFM), a car-following model, and a lane change model to solve this problem. Further, the vehicle–pedestrian force and Time to Conflict (TTC) model the interaction between people and vehicles. In the case study of parking lot evacuation, the model is proved to be effective in simulating the behavior of pedestrian–vehicle flow in a complex scenario.