LSTM-based lane change prediction using Waymo open motion dataset: The role of vehicle operating space

Lane change prediction is critical for crash avoidance but challenging as it requires the understanding of the instantaneous driving environment.With cutting-edge artificial intelligence and sensing technologies,autonomous vehicles(AVs)are expected to have exceptional perception systems to capture instantaneously their driving environments for predicting lane changes.By exploring the Waymo open motion dataset,this study proposes a framework to explore autonomous driving data and investigate lane change behaviors.In the framework,this study develops a Long Short-Term Memory(LSTM)model to predict lane changing behaviors.The concept of Vehicle Operating Space(VOS)is introduced to quantify a vehicle's instantaneous driving environment as an important indicator used to predict vehicle lane changes.To examine the robustness of the model,a series of sensitivity analysis are conducted by varying the feature selection,prediction horizon,and training data balancing ratios.The test results show that including VOS into modeling can speed up the loss decay in the training process and lead to higher accuracy and recall for predicting lane-change behaviors.This study offers an example along with a methodological framework for transportation researchers to use emerging autonomous driving data to investigate driving behaviors and traffic environments.

Influence of lane change on stability analysis for two-lane traffic flow

This paper deals mainly with the influence of lane changing behaviours on the stability of two-lane traffic flow under a periodic boundary condition. Following the description of an optimal velocity model for two vehicle groups and the derivation of their stability conditions, the feedback signals, which involve information about vehicles from both lanes acting on the two-lane traffic system, are introduced into the optimal velocity model. The control signals play a role in alleviating the traffic jam only if the traffic state is in congestion, and their role will vanish if the traffic state is in the steady state. The numerical simulations show that lane changing behaviours can break the steady state of two-lane traffic flow and aggravate the traffic disturbance, but the control method would successfully suppress the traffic jam eventually, which implies that the conclusions obtained here have certain theoretical and practical significance.

A Trajectory Planning Method of Automatic Lane Change Based on Dynamic Safety Domain

Traditional research on automatic lane change has primarily focused on high-speed scenarios and has not considered the dynamic state changes of surrounding vehicles.This paper addresses this problem by proposing a trajectory planning method to enable automatic lane change at medium and low speeds.The method is based on a dynamic safety domain model,which takes into account the actual state change of surrounding vehicles,as well as the upper boundary of the safety domain for collision avoidance and the lower boundary of comfort for vehicle stability.The proposed method involves the quantification of the safety and comfort boundaries through parametric modeling of the vehicle.A quintic polynomial trajectory planning method is proposed and evaluated through simulation and testing,resulting in improved safety and comfort for automatic lane change.

Exploring the stimulative effect on following drivers in a consecutive lane change using microscopic vehicle trajectory data

Improper lane-changing behaviours may result in breakdown of traffic flow and the occurrence of various types of collisions.This study investigates lane-changing behaviours of multiple vehicles and the stimulative effect on following drivers in a consecutive lanechanging scenario.The microscopic trajectory data from the HighD dataset are used for driving behaviour analysis.Two discretionary lane-changing vehicle groups constitute a consecutive lane-changing scenario,and not only distance-and speed-related factors but also driving behaviours are taken into account to examine the impacts on the utility of following lane-changing vehicles.A random parameters logit model is developed to capture the driver’s psychological heterogeneity in the consecutive lane-changing situation.Furthermore,a lane-changing utility prediction model is established based on three supervised learning algorithms to detect the improper lane-changing decision.Results indicate that 1)the consecutive lane-changing behaviours have a significant negative effect on the following lane-changing vehicles after lane change;2)the stimulative effect exists in a consecutive lane-change situation and its influence is heterogeneous due to different psychological activities of drivers;and 3)the utility prediction model can be used to detect an improper lane-changing decision.

Dynamic Trajectory Planning of Autonomous Lane Change at Medium and Low Speeds Based on Elastic Soft Constraint of the Safety Domain

Most current research on the trajectory planning of the autonomous lane change focuses on high-speed scenarios and assumes that the states of the surrounding vehicles keep stable during the lane change.The methods based on geometric-curve are mostly used for trajectory planning.In this paper,considering the inevitable development of the autonomous driving,the surrounding vehicles are assumed to be driven by human drivers,while the ego vehicles are able to autonomously change lanes.Representative local lane-change scenarios are then designed and analyzed in detail aiming at medium-and low-speed lane-change conditions.Additionally,in contrast with most research,dynamic trajectory planning which considers the possible state variations of the surrounding vehicles and the driver characteristics is studied and described by a fifth-order polynomial function.The safety and comfort of the dynamic trajectory planning are validated through simulation.Moreover,the elastic soft constraint of the safety domain is designed,whereby the sensitivity of the studied dynamic trajectory planning system is reduced under the premise of ensuring safety.The effectiveness of the elastic soft constraint in terms of improving comfort during the lane change is verified through simulation.The availability of the dynamic trajectory planning system with the elastic soft constraint is demonstrated with the addition of trajectory tracking based on model predictive control,showing its potential in practical applications.

Stable trajectory planning and energyefficience control allocation of lane change maneuver for autonomous electric vehicle

Purpose–The purpose of this paper is to investigate problems in performing stable lane changes and tofind a solution to reduce energy consumption of autonomous electric vehicles.Design/methodology/approach–An optimization algorithm,model predictive control(MPC)and Karush–Kuhn–Tucker(KKT)conditions are adopted to resolve the problems of obtaining optimal lane time,tracking dynamic reference and energy-efficient allocation.In this paper,the dynamic constraints of vehicles during lane change arefirst established based on the longitudinal and lateral force coupling characteristics and the nominal reference trajectory.Then,by optimizing the lane change time,the yaw rate and lateral acceleration that connect with the lane change time are limed.Furthermore,to assure the dynamic properties of autonomous vehicles,the real system inputs under the restraints are obtained by using the MPC method.Based on the gained inputs and the efficient map of brushless direct-current in-wheel motors(BLDC IWMs),the nonlinear cost function which combines vehicle dynamic and energy consumption is given and the KKT-based method is adopted.Findings–The effectiveness of the proposed control system is verified by numerical simulations.Consequently,the proposed control system can successfully achieve stable trajectory planning,which means that the yaw rate and longitudinal and lateral acceleration of vehicle are within stability boundaries,which accomplishes accurate tracking control and decreases obvious energy consumption.Originality/value–This paper proposes a solution to simultaneously satisfy stable lane change maneuvering and reduction of energy consumption for autonomous electric vehicles.Different from previous path planning researches in which only the geometric constraints are involved,this paper considers vehicle dynamics,and stability boundaries are established in path planning to ensure the feasibility of the generated reference path.

A car-following model with the anticipation effect of potential lane changing

In this paper, a new car-following model is presented, taking into account the anticipation of potential lane changing by the leading vehicle. The stability condition of the model is obtained by using the linear stability theory. The modified Korteweg-de Vries （KdV） equation is constructed and solved, and three types of traffic flow in the headway-sensitivity space, namely stable, metastable and unstable ones, are classified. Both the analytical and simu- lation results show that anxiety about lane changing does indeed have an influence on driving behavior and that a consideration of lane changing probability in the car-following model could stabilize traffic flows. The quantitative relationship between stability improvement and lane changing probability is also investigated.

The effect of moving bottlenecks on a two-lane traffic flow

In this paper, we study the effect of moving bottlenecks on traffic flow. The full velocity difference （FVD） model is extended to the traffic flow on a two-lane highway, and new lane changing rule is proposed to reproduce the vehicular lane changing behavior. Using this model, we derive the fundamental current-density diagrams for the traffic flow with the effect of moving bottleneck. Moreover, typical time-space diagram for a two-lane highway shows the formation and dissipation of a moving bottleneck. Results demonstrate that the effect of moving bottleneck enlarges with the increase of traffic density, but the effect can be reduced by increasing the maximum velocity of heavy truck. The effects of multiple moving bottlenecks under different conditions are investigated. The effect becomes more remarkable when the coupling effect of multiple moving bottlenecks occurs.

Improved STCA Model for Multi-Lane Using Driving Guidance under CVIS

In a multi-lane area,the increasing randomness of lane changes contributes to traffic insecurity and local traffic flow instability.A study on safe lane shifting activity that focuses on threat assessment under real-time knowledge is necessary to enhance smooth vehicle flow.This paper proposed amore comprehensive lane changing guidance rule to investigate the status of surrounding vehicles to accommodate future vehicle-on-road collaborative environments based on these parameters 1)lane change demand and 2)treat assessment function.The collaborative relationships between vehicles are analyzed using a cellular automata model based on their location,velocity,and acceleration.We analyze and examine the relationship between the number of lanes and traffic flow when the road capacity is heavily mined via intelligent lane changing.Our analysis can further provide theoretical guidance for the selection of road expansion mode.Our proposed STCA-L is compared based on the average speed,average flow,lane changing frequency,spatial and temporal pattern of STCA,STCA-I,and STCA-S,and STCA-M under different vehicle densities.The numerical simulation results show that our proposed STCA-L provides themost flexible lane changing guidance in the multi-lanes road.Moreover,the simulated results show that the exponential growth of physical space cannot provide the corresponding increase in the average flow of vehicles.

Reference Modelof Desired Yaw Anglefor Automated Lane Changing Behavior of Vehicle

In this paper,it studies the problem of trajectory planning and tracking for lane changing behavior of vehicle in automatic highway systems. Based on the model of yaw angle acceleration with positive and negative trapezoid constraint,by analyzing the variation laws of yaw motion of vehicle during a lane changing maneuver,the reference model of desired yaw angle and yaw rate for lane changing is generated. According to the yaw angle model,the vertical and horizontal coordinates of trajectory for vehicle lane change are calculated. Assuming that the road curvature is a constant,the difference and associations between two scenarios are analyzed,the lane changing maneuvers occurred on curve road and straight road,respectively. On this basis,it deduces the calculation method of desired yaw angle for lane changing on circular road. Simulation result shows that,it is different from traditional lateral acceleration planning method with the trapezoid constraint,by applying the trapezoidal yaw acceleration reference model proposed in this paper, the resulting expected yaw angular acceleration is continuous,and the step tracking for steering angle is not needed to implement. Due to the desired yaw model is direct designed based on the variation laws of raw movement of vehicle during a lane changing maneuver, rather than indirectly calculated from the trajectory model for lane changing, the calculation steps are simplified.

Quantification of Ride Comfort Using Musculoskeletal Mathematical Model Considering Vehicle Behavior

This research aims to quantify driver ride comfort due to changes in damper characteristics between comfort mode and sport mode,considering the vehicle’s inertial behavior.The comfort of riding in an automobile has been evaluated in recent years on the basis of a subjective sensory evaluation given by the driver.However,reflecting driving sensations in design work to improve ride comfort is abstract in nature and difficult to express theoretically.Therefore,we evaluated the human body’s effects while driving scientifically by quantifying the driver’s behavior while operating the steering wheel and the behavior of the automobile while in motion using physical quantities.To this end,we collected driver and vehicle data using amotion capture system and vehicle CAN and IMU sensors.We also constructed a three-dimensional musculoskeletal mathematical model to simulate driver movements and calculate the power and amount of energy per unit of time used for driving the joints and muscles of the human body.Here,we used comfort mode and sport mode to compare damper characteristics in terms of hardness.In comfort mode,damper characteristics are soft and steering stability is mild,but vibration from the road is not easily transmitted to the driver making for a lighter load on the driver.In sport mode,on the other hand,damper characteristics are hard and steering stability is comparatively better.Still,vibration from the road is easily transmitted to the driver,whichmakes it easy for a load to be placed on the driver.As a result of this comparison,it was found that a load was most likely to be applied to the driver’s neck.This result in relation to the neck joint can therefore be treated as an objective measure for quantifying ride comfort.

Analysis of the Effect of Trailer Tire Size on the Articulated Vehicle’s Stability

This research paper aims to identify the effect of tire size on the handling characteristics of a trailer attached to a vehicle. In various stability tests, different models with different tires from the market were tested. A successful outcome of this research would generate an efficient tire selection process and improve the handling of a trailer attached to a vehicle while maximizing fuel efficiency. In this study, different accurate tire models using the magic formula were developed in vehicle dynamics modelling and simulation software. These models were then simulated on on-road conditions to predict vehicle and trailer behaviour under different conditions within the software. Two distinct tests were conducted, the J-Turn test and the Double Lane change test. The results of these tests were used to evaluate the handling characteristics and decide on a better tire size for the trailer attached to the vehicle.

Cooperative Lane-Change Maneuver for Multiple Automated Vehicles on a Highway

With the development of vehicle-to-vehicle(V2V)communication,it is possible to share information among multiple vehicles.However,the existing research on automated lane changes concentrates only on the single-vehicle lane change with self-detective information.Cooperative lane changes are still a new area with more complicated scenarios and can improve safety and lane-change efficiency.Therefore,a multi-vehicle cooperative automated lane-change maneuver based on V2V communication for scenarios of eight vehicles on three lanes was proposed.In these scenarios,same-direction and intersectant-direction cooperative lane changes were defined.The vehicle that made the cooperative decision obtained the information of surrounding vehicles that were used to cooperatively plan the trajectories,which was called cooperative trajectory planning.The cooperative safety spacing model was proposed to guarantee and improve the safety of all vehicles,and it essentially developed constraints for the trajectory-planning task.Trajectory planning was treated as an optimization problem with the objective of maximizing safety,comfort,and lane-change efficiency under the constraints of vehicle dynamics and the aforementioned safety spacing model.Trajectory tracking based on a model predictive control method was designed to minimize tracking errors and control increments.Finally,to verify the validity of the proposed maneuver,an integrated simulation platform combining MATLAB/Simulink with CarSim was established.Moreover,a hardware-in-the-loop test bench was performed for further verification.The results indicated that the proposed multi-vehicle cooperative automated lane-change maneuver can achieve lane changes of multiple vehicles and increase lane-change efficiency while guaranteeing safety and comfort.

Traffic flow velocity disturbance characteristics and control strategy at the bottleneck of expressway

In the three-phase traffic flow studies, the traffic flow characteristic at the bottleneck section is a hot spot in the academic field. The controversy about the characteristics of the synchronized flow at bottleneck is also the main contradiction between the three-phase traffic flow theory and the traditional traffic flow theory. Under the framework of three-phase traffic flow theory, this paper takes the on-ramp as an example to discuss the traffic flow characteristics at the bottleneck section.In particular, this paper mainly conducts the micro-analysis to the effect of lane change under the two lane conditions, as well as the effect of the on-ramp on the main line traffic flow. It is found that when the main road flow is low, the greater the on-ramp inflow rate, the higher the average speed of the whole road section. As the probability of vehicles entering from the on-ramp increases, the flow and the average speed of the main road are gradually stabilized, and then the on-ramp inflow vehicles no longer have a significant impact on the traffic flow. In addition, this paper focuses on the velocity disturbance generated at the on-ramp, and proposes the corresponding on-ramp control strategy based on it, and the simulation verified that the control strategy can reasonably control the traffic flow by the on-ramp, which can meet the control strategy requirements to some extent.

Evaluation of Automatic Lane-Change Model Based on Vehicle Cluster Generalized Dynamic System

The lane-change transportation research usually focuses on the efficiency and stability of the macro traffic flow while ignoring the driving comfort of individual vehicles.And many studies of lane-change models are often limited to the performance of a single vehicle,which leads to a lack of macroscopic evaluation.To solve the above limitations,an automatic lane-change generalized dynamic model is adopted.In this model,the lane-change behavior of an individual vehicle is considered as the generalized excitation and the restraining force between vehicles is described with the car-following model.Macro and micro evaluation indexes are also adopted to evaluate the automatic lane-change behavior in traffic flow.Furthermore,this paper proposes a modified intelligent driver model(IDM)to describe the state change process during lane change.The hyperbolic tangent transition function is used to eliminate the vehicle state mutation.The simulation results show that the proposed automatic lane-change generalized dynamic model can reflect the macro and micro parameters of the traffic flow.And compared with the traditional IDM model,the proposed HC-IDM model achieves higher comfort performance and lower fluctuation of traffic flow.

Lane changing and congestion are mutually reinforcing?

This study presents a comprehensive analysis of the relationship between congestion and lane changing,using vehicle trajectory data from the M1 motorway in Sydney.We establish a connection between the distribution of travel time and lane changing frequency and employ a Poisson process to describe the intensity of lane changing occurrences in different travel time ranges.From an individual perspective,lane changing does not bring significant speed benefits in most cases,except when the speed range is between 45 and 50 km/h.From a system perspective,the relationship between lane change rate and speed depends on the purpose of the lane changes.In merging,diverging,and lane restriction areas,for instance,mandatory lane changes dominate.In most sections of the motorway,discretionary lane changes are motivated by the expectation of improving speed and/or safety.Additionally,we demonstrate a mutual causality relationship between lane changing and congestion through the Granger causality test.This relationship is more pronounced in general areas during peak periods and contributes to the deterioration of the driving environment.

Lane changing assistance strategy based on an improved probabilistic model of dynamic occupancy grids

Lane changing assistance in autonomous vehicles is a popular research topic. Scene modeling of the driving area is a prerequisite for lane changing decision problems. A road environment representation method based on a dynamic occupancy grid is proposed in this study. The model encapsulates the data such as vehicle speed, obstacles, lane lines, and traffic rules into a form of spatial drivability probability. This information is compiled into a hash table, and the grid map is mapped into a hash map by means of hash function. A vehicle behavior decision cost equation is established with the model to help drivers make accurate vehicle lane changing decisions based on the principle of least cost, while considering influencing factors such as vehicle drivability, safety, and power. The feasibility of the lane changing assistance strategy is verified through vehicle tests, and the results show that the lane changing assistance system based on a probabilistic model of dynamic occupancy grids can provide lane changing assistance to drivers taking into consideration the dynamics and safety.

Spatial-Temporal ConvLSTM for Vehicle Driving Intention Prediction

Driving intention prediction from a bird’s-eye view has always been an active research area. However,existing research, on one hand, has only focused on predicting lane change intention in highway scenarios and, on the other hand, has not modeled the influence and spatiotemporal relationship of surrounding vehicles. This study extends the application scenarios to urban road scenarios. A spatial-temporal convolutional long short-term memory(ConvLSTM) model is proposed to predict the vehicle’s lateral and longitudinal driving intentions simultaneously. This network includes two modules: the first module mines the information of the target vehicle using the long short-term memory(LSTM) network and the second module uses ConvLSTM to capture the spatial interactions and temporal evolution of surrounding vehicles simultaneously when modeling the influence of surrounding vehicles. The model is trained and verified on a real road dataset, and the results show that the spatial-temporal ConvLSTM model is superior to the traditional LSTM in terms of accuracy, precision, and recall, which helps improve the prediction accuracy at different time horizons.

Integrated control of highway traffic flow

Several efforts have been made during recent years on the control of traffic flow using ramp metering,variable speed limit and lane change control.Most of these techniques reported significant traffic mobility improvements in macroscopic simulations,which cannot be confirmed in some microscopic simulation cases.We demonstrate that the lack of travel time improvement is due to the forced lane changes at the bottleneck which introduce a capacity drop.In this paper,a lane change controller is proposed which provides lane change recommendations to avoid the capacity drop.Combined with the lane change control,a feedback linearisation variable speed limit controller which maintains the bottleneck flow at the maximum level and coordinates with ramp metering controllers is proposed.The integrated VSL,RM and LC controller can improve system performance,manage the ramp queues and maintain the fairness between mainline and on-ramp flows.Microscopic simulations show consistent improvements under different scenarios.

Accelerated testing for automated vehicles safety evaluation in cut-in scenarios based on importance sampling,genetic algorithm and simulation applications

Purpose–It would take billions of miles’field road testing to demonstrate that the safety of automated vehicle is statistically significantly higher than the safety of human driving because that the accident of vehicle is rare event.Design/methodology/approach–This paper proposes an accelerated testing method for automated vehicles safety evaluation based on improved importance sampling(IS)techniques.Taking the typical cut-in scenario as example,the proposed method extracts the critical variables of the scenario.Then,the distributions of critical variables are statistically fitted.The genetic algorithm is used to calculate the optimal IS parameters by solving an optimization problem.Considering the error of distribution fitting,the result is modified so that it can accurately reveal the safety benefits of automated vehicles in the real world.Findings–Based on the naturalistic driving data in Shanghai,the proposed method is validated by simulation.The result shows that compared with the existing methods,the proposed method improves the test efficiency by 35 per cent,and the accuracy of accelerated test result is increased by 23 per cent.Originality/value–This paper has three contributions.First,the genetic algorithm is used to calculate IS parameters,which improves the efficiency of test.Second,the result of test is modified by the error correction parameter,which improves the accuracy of test result.Third,typical high-risk cut-in scenarios in China are analyzed,and the proposed method is validated by simulation.