One of the challenges posed by the study of vehicular ad hoc networks (VANETs) is the transmission of data issued by valued traffic information services under incomplete link conditions. Many dissemination protocols have been developed by the community to solve the issue. In this survey, the authors explore the service discovery where data dissemination should serve as a foundation. Then, they propose an overview and taxonomy of a large range of data dissemination available for VANETs. Finally, they illustrate the simulation infrastructure by collaborating two independent simulators. The objective is to provide guidelines to easily understand and extend the capabilities of protocols according to the users’ needs.

In recent years, a number of vision-based classification methods have been proposed. However, a few of them were paid attention to vehicle-type classification in a real-world image, which is an important part of the intelligent transportation system. Owing to the large variances of the car appearance in images, it is critical to capture the discriminative object parts that can provide key information about the car pose. In the authors’ project, the traditional convolutional neural network (CNN) models are modified and experiments are followed as well. The model has two main contributions. First, the output shows a confidence score of how likely this box contains a car for each predicted box, which has some certain advantages compared with other models and is quite different from traditional approaches. Another contribution is the fine-grained classification of the makers and models of a car, which need to train the bounding box predictors as part of the network training. The experiment results show that data enhancement and pre-train of CNNs with original images can classify the vehicle makes and models with a high accuracy of nearly 80%. Cropping images by cascade methods can increase the precision to 86.6%.

Driver distraction is a leading factor in car crashes. With a goal to reduce traffic accidents and improve transportation safety, this study proposes a driver distraction detection system which identifies various types of distractions through a camera observing the driver. An assisted driving testbed is developed for the purpose of creating realistic driving experiences and validating the distraction detection algorithms. The authors collected a dataset which consists of images of the drivers in both normal and distracted driving postures. Four deep convolutional neural networks including VGG-16, AlexNet, GoogleNet, and residual network are implemented and evaluated on an embedded graphic processing unit platform. In addition, they developed a conversational warning system that alerts the driver in real-time when he/she does not focus on the driving task. Experimental results show that the proposed approach outperforms the baseline one which has only 256 neurons in the fully-connected layers. Furthermore, the results indicate that the GoogleNet is the best model out of the four for distraction detection in the driving simulator testbed.

This study provides a methodology to identify travellers’ transportation modes by tracking the mobile phone data, which aims to obtain the accurate mode split rate for providing decision support in urban traffic planning. First, the effective mobile phone singling data and GPS data are collected from the communication operators and a mobile phone app, respectively. Considering the differences in velocity and acceleration of different trip modes, a trip mode characteristic description model is built based on wave characteristics and moving average method. Compared with the wave characteristics, the moving average method shows a better accuracy of 90%. Then training samples are drawn by two data selection methods including probability proportional to size sampling and equal amount sampling. Furthermore, the classifier method for mode choice prediction is developed by support vector machines (SVMs) and back propagation neutral network. Finally, the results of the case study show that using a 30-point moving average training data set can improve the prediction accuracy largely, and the SVM method gets a better accuracy of 82%. The potential of using the mobile phone data to build a new mode choice prediction method in the field of transportation is shown.

The identification of a traffic shockwave traditionally is conducted offline, which inhibits its implementation for active traffic management strategies. This study aims to develop an online approach to detect traffic shockwaves on freeways, particularly the end-of-queue shockwaves, using spacing-based probe vehicles (SPVs) that can obtain the trajectories of its leading and/or following vehicles. The proposed framework consists of four stages: (i) local shockwave (LSW) position detection, (ii) LSW speed estimation, (iii) grouping of LSWs into a whole shockwave (WSW) and (iv) WSW speed estimation. In particular, two alternatives, namely the line connection-based method and the Lighthill–Whitham–Richards model-based method (LWRM), are proposed for stage 2, and other two alternatives, namely the simple averaging method and the hybrid method (HM), are proposed for stage 4. A set of next generation simulation data are utilised to evaluate the performance of the proposed method. The results demonstrate that the combination of LWRM + HM outperforms among the four combined methods. A series of the analysis indicate that the proposed method is computationally efficient, accurate and more importantly, it is applicable to sensor data from SPVs with real-world noise.

The vicinity of traffic signals is one of the most special and critical areas in the whole road system. Considering that the driver can obtain traffic signals information through two approaches: vision and the vehicle-to-vehicle (V2X) communication equipment in the vehicle, this study proposes two vehicle cooperative driving models that apply to the vicinity of traffic signals: the intelligent driver model (IDM) in the vicinity of traffic signals (IDM-VT) and IDM in the vicinity of traffic signals under V2X environment (IDM-VT-V2X). These two models are both based on the intelligent driver model. In accordance with different situations of vehicles in the vicinity of traffic signals, such as distance from the traffic lights, whether there was another vehicle in front and different states of traffic lights, the pertinent analysis was conducted, and the cooperative driving strategy that satisfied the complicated situations of vehicle was proposed. These two models were verified and compared in the simulation experiment. The results of simulation show that when comparing with the IDM-VT, the IDM-VT-V2X can reduce the average travel time and the average stop delay time by 12.98 and 98.32%, respectively, and it can also reduce 22.53% fuel consumption.

This study is focused on the design of cooperative adaptive cruise control (CACC) to regulate the longitudinal motion of connected and automated vehicles (CAVs) in mixed traffic that is composed of human-driven vehicles and CAVs. Wireless vehicle-to-vehicle communication is exploited to monitor the motion of multiple broadcasting vehicles, and a strategy is designed to determine whether the received data of other vehicles are incorporated into CACC. A condition is derived for choosing control gains that ensure the internal stability of CAVs in the presence of time delays and switching connectivity topologies of information flow. Moreover, because the switching connectivity topologies may change the dynamics of the whole vehicle chain, the authors apply a data-driven approach for online optimisation of control gains such that CACC adapts to the variations of connectivity topologies. The proposed selective CACC is validated through numerical simulations. To enhance the fidelity of simulations, they use the data collected through on-road experiments to simulate the motion of human-driven vehicles and apply the physics-based vehicle dynamic model to simulate the motion of CAVs. Simulation results demonstrate the advantages of the proposed selective CACC in improving vehicle safety and in mitigating perturbations in mixed traffic.

This study analyses the effects of individual and trip characteristics on passenger travel choice behaviours when presented with all-stop, skip-stop, and transfer services. Bus stops are classified into two types: an ‘AB Station’ provides both all-stop and skip-stop services and an ‘A Station’ provides only all-stop service. Passenger travel choice behaviours ‘from AB to AB’ and ‘from AB to A’ are studied based on travel choice probabilities. A stated preference survey was conducted in Beijing to collect individual and trip characteristics for various travel circumstances and used to develop passenger choice probability models based on logit model. The results show that the probability of choosing skip-stop service increases with the increase in travel distance and the decrease in in-vehicle time; transfer service is not popular, even for a long trips; skip-stop and transfer services are more attractive to passengers taking a mandatory trip; there are differences in choice behaviours between male and female passengers; compared to high-income passengers, low-to-middle income passengers exhibit a lower probability of choosing transfer service because of the additional travel cost. This study contributes to predicting future demands for different bus services, the implementation and optimisation of skip-stop strategies, and bus schedule improvement.

In Shanghai, China, two transportation demand management (TDM) measures, auctions of Shanghai vehicle licence plates and a narrow-time-based travel restriction policy, have been implemented to control the vehicle ownership and the use of vehicles registered outside Shanghai (VROS). To investigate the impact of these two TDM measures on VROS, vehicle use behaviour is analysed with automatic licence plate recognition (ALPR) data. A two-step k-means clustering algorithm is proposed to classify VROS' use behaviour from ALPR raw data. Moreover, the spatiotemporal patterns of each type of VROS and the structure of total transportation demand are analysed. The results show that VROS in Shanghai expressway network can be classified into five types. Type 1 vehicles are used for commuting (COM), and type 2 vehicles are used with high intensity during both on workdays and non-workdays (HHI). COM and HHI are mainly used by local Shanghai residents who cannot obtain a local licence plate under the auction policy. These two types of VROS only account for 3.6% of total vehicles but generate 14.2% of total traffic demand. If the users of COM and HHI transferred to public transit, the traffic congestion of expressway network would be greatly alleviated.

In the literature, the intensive research effort has been made on the trajectory planning for autonomous vehicles, while the integration of the trajectory planner with the trajectory controller is less focused. This study proposes the spatiotemporal-based trajectory planner and controller by a two-level dynamically integrated structure. In the upper level, the best trajectory is selected among a group of candidate time-parameterised trajectories, while the target vehicle ending position and velocity can be satisfied. Then the planned trajectory is evaluated by checking the feasibility when the actual vehicle dynamic motion constraints are considered. After that, the lower level trajectory controller based on the vehicle dynamics model will control the vehicle to follow the desired trajectory. Numerical simulations are used to validate the effectiveness of the proposed approach, where the scenario of an intersection and the scenario of overtaking are applied to show that the proposed trajectory controller can successfully achieve the control targets. In addition, compared with the potential field method, the proposed method based on the four-wheel independent steering and four-wheel independent driving electric vehicle shows great advantages in guaranteeing the vehicle handling and stability.

An Adaptive Local Texture Feature Background Model (ALTF-BM) is proposed to resolve the deficiency in current background models, which are easily contaminated by sudden and gradual illumination changes in complex urban traffic scenes. Based on Weber's law, the authors first develop Adaptive Local Texture Feature (ALTF), calculated over a predefined local region around the pixel employing an adaptive distance threshold, and then the background is modelled on the base of sample consensus scheme using the calculated features. Furthermore, to label the foreground pixels, the difference between the background model and input video frames is then directly compared by ALTF encoding. Finally, the model is updated using the random update policy to adapt to the changing illumination and the dynamic background. The experimental results on real-world urban traffic videos and the public Change Detection benchmark of 2014 (CDnet2014) show that the proposed ALTF-BM offers the best performance compared to the other state-of-the-art texture-based methods, and the average F-measures and similarity results of the proposed ALTF-BM are 0.547 and 0.393 higher than benchmarks on the night traffic-light sequence, respectively. The encouraging experimental results demonstrate the efficiency of the proposed ALTF-BM in handling sudden and gradual illumination changes in urban traffic scenes.

On-demand one-way carsharing systems are increasingly gaining popularity nowadays and the market is growing unprecedentedly. However, operating techniques such as vehicle surveillance and fleet management fall behind the industrial development. Practice on EVCARD – an on-demand one-way electric vehicle carsharing system operating in Shanghai – raises an issue on dynamically predicting the user destinations, in order to support decisions on dynamic fleet management. This study presents a global positioning system (GPS)-data-driven method to solve the problem. The historical vehicle GPS data is enabled to match the user current trajectories and infer their possible destinations. Based on the GPS trajectory similarity measurement, the study presents a four-step procedure, including (i) similarity calculation, (ii) most-similar track detection, (iii) adjustment, and (iv) sorting. The method also takes the station correlations and the user historical destinations into account. A case study is given to demonstrate the dynamic prediction procedure of this method. Experiment on 96,821 valid test tracks shows that the positive prediction rate can be above 92% if the test trip has been completed over 70%. Factors that may influence the prediction result are additionally discussed, which include the existence of round-trips, the coverage of samples, and the alternatives of destinations.

Electric vehicles have been the focus of the automotive industry in recent years. However, relatively small driving range of electric vehicles makes it not be broadly adopted in the market. Regenerative braking is one of the most effective ways to extend the endurance of electric vehicles. To sufficiently utilise the regenerative braking of electric vehicles and explore the potential of the electric motor plugging braking capability to simplify the braking system structure and reduce the cost, a new braking strategy based on the driver's braking intention and motor working characteristics is proposed. Driver's braking intention is classified as the emergency braking and the normal braking. In the case of normal braking, model predictive control (MPC) is used to express driver's braking intention. By adjusting the weight of the MPC cost function, different braking intentions can be achieved. This strategy is able to achieve as much as possible braking energy recovery without violating the driver's braking intention. In the case of the emergency braking, the sliding mode based optimal slip ratio control is adopted and it is able to obtain the shortest braking distance. In order to validate the effectiveness of the proposed approach, numerical simulations on a quarter-vehicle braking model are tested.

The variety of data collecting and communication methods used in intelligent transportation systems such as sensors, cameras, and communication networks bring about huge volumes of data that are available for numerous transportation applications and related research on smart cities. However, it is still a challenge to integrate these heterogeneous data sources into a singular data schema in practice. Compared to a single data source, higher data accuracy can be obtained through integration of the multiple data sources if the data quality from each source has been known. In this study, a data fusion method based on Bayesian fusion rules is proposed to merge traffic speed from different data sources according to their prior probability that can be inferred from a high-order multivariable Markov model by considering the relations of multiple traffic factors in a systemic perspective. Case studies based on freeway data, such as loop data, INRIX data, and data from the National Performance Management and Research Data Set, are performed to validate the effectiveness of proposed speed fusion method.

Driver's fatigue detection, based on electroencephalography (EEG) signals, is a worthy field of research to study evidence regarding how to exactly pre-warn and avoid casualties nowadays. In this study, an EEG-based system of perfect performance and good stability for evaluating driver's fatigue with only one electrode by ensemble learning method is proposed. Given that EEG signals are unstable and non-linear that using several common entropy measurements to analyse EEG signals is more appropriate including spectral entropy, approximate entropy, sample entropy and fuzzy entropy. In this study, unlike other methods using a single classifier, three ensemble approaches (bagging, random forest and boosting) based on three base classifiers were employed and compared. A driving simulator in this study was used for 12 healthy and adult subjects to perform a continuous simulated driving experiment for 1–2 h. The experimental results show that the proposed method can make use of only one electrode (T6) by gradient boosted DT for driver's fatigue detection, while the average classification accuracy is >94%. The findings of this study indicated that a single EEG channel with optimal ensemble classifier may be a good candidate for usage in the portable system for driver's fatigue detection.

With the advancement of vehicle-to-vehicle and vehicle-to-infrastructure technologies, more and more real-time information regarding traffic and transportation system will be available to vehicles. This paper presents the development of a novel algorithm that uses available velocity bounds and powertrain information to generate an optimal velocity trajectory over a prediction horizon. When utilised by a vehicle, this optimal velocity trajectory reduces fuel consumption. The objective of this optimisation problem is to reduce dynamic losses, required tractive force, and completing trip distance with a given travel time. Sequential quadratic programming method is employed for this nonlinearly constrained optimisation problem. When applied to a GM Volt-2, the generated velocity trajectory saves fuel compared to a real-world drive cycle. The simulation results confirm the fuel consumption reduction with the rule-based mode selection and the energy management strategy of a GM Volt 2 model in Autonomie.

The real time and adaptability of emergency lane-changing trajectory planning method is critical to the safe operation of the intelligent vehicle. This study presents the design and implementation of the trajectory planning and optimisation method of intelligent vehicle lane changing emergently based on hp-adaptive pseudospectral method. The method divides the emergency lane-changing process into the initial stage and tracking stage for trajectory planning based on road steering experiment and sigmoid functions. An hp-adaptive pseudospectral method is introduced to connect and optimise the lane-changing trajectory. PreScan and Matlab are used to simulate the planning of emergency lane-changing trajectory under six conditions, and the simulation results verify the effectiveness and real time of the planning and optimisation method. In comparison with trajectory planning methods based on polynomial functions, this method is characterised by shorter response time and safety distance and has better adaptability under different conditions.

The primary function of a collision risk index is to determine the time when ships take action to avoid a collision. In this study, based on the complex non-linear relationship between the collision risk degree and its influencing factors, classification and regression trees (CARTs) are applied to construct a prediction model for ship collision risk. The fuzzy comprehensive evaluation method is used to evaluate the risk of ship encounter samples to build a collision risk identification library containing expert collision avoidance experience information. The authors’ proposed CART regression model is trained using the samples in this identification library to develop a collision risk prediction model based on the CART. Their experimental results show that their proposed CART prediction model is better that the existing ship collision risk prediction model in terms of prediction accuracy and prediction speed when the feature dimension is low and the sample size is small.

A companion of moving objects is an object group that move together in a period of time. Platoon companions are a generalised companion pattern, which describes a group of objects that move together for time segments, each with some minimum consecutive duration of time. This study proposes a method that can instantly discover platoon companions from a special kind of streaming traffic data, called automatic number plate recognition data. Compared to related approaches, the authors transform the companion discovery into a frequent sequence mining problem. The authors propose a data structure, platoon tree (PTree), to record discovered platoon companions. To reduce the cost of tree traversal during mining platoon companions, they utilise the last two together-moving objects of a group to update PTree. Finally, a lot of experiments have been carried out to show the efficiency and effectiveness of the proposed approach.

The ever-growing air traffic demand arouses an urgent need for improved airport ground movement efficiency. New operational concepts are emerging which use time-based taxi trajectories to reduce uncertainty and make more efficient use of the airport resource. In this study, an improved approach is proposed for time-based taxi trajectory planning, which is formulated as the shortest path problem with time windows and the maximum traversal time constraint. With the introduction of the taxi time in the cost and the maximum traversal time constraint to limit the waiting time, more efficient and fluent ground movement of aircraft can be realised. An A*-based solution algorithm is developed for the investigated problem, which utilises the arrival time interval and dominance-based comparison to search for the best solution. Experimental results on real-world problem instances demonstrate the effectiveness of the proposed approach as well as its advantages over the existing approach.

Recent studies on train timetable are paying more and more attention to the dynamic characteristics. Among the dynamic characteristics, stability is a most important one, which determines the capacity of the train timetable to tolerate the disturbance in the train operation process. In this study, the authors build a complex network model to describe the train timetable, making it possible to utilise the complex network theory to study the train timetable optimisation problem. Then, they design the solving algorithm to solve the problem. Finally, they present a computing case to prove the approach to improve the train timetable stability is practical. The approach proposed in this study can generate referential advice for the railway operators design the train timetable.

This study presents a closed-loop hierarchical control strategy for a group of connected and autonomous hybrid electric vehicles (HEVs) with the purpose of optimising fuel economy while guaranteeing traffic mobility and vehicle safety. In the hierarchical control architecture, a decentralised stochastic model predictive control (SMPC)-based higher level controller incorporating signal phase and timing information is formulated to optimise the velocity profile of each vehicle, and an adaptive equivalent consumption minimisation strategy based lower level controller is employed for the energy management control of the HEVs. Creatively, random errors of the control variable are considered in the SMPC framework. The errors are discretised and modelled as a Markov process and the state transition matrix of the errors is generated randomly to capture the error transition dynamics. To solve the SMPC problem more efficiently, a scenario-based SMPC is employed. Moreover, the propulsion and recuperation efficiencies of the lower level controller are calculated at each time step with measurable variables, and fed back to the higher level controller for velocity optimisation in next time step. Simulation results validate the control effectiveness and advantages of the proposed control architecture.

The track-to-track fusion (T2TF) algorithm is currently an attractive fusion methodology in the industry because the algorithm can reflect the reliability of sensor tracks. However, the T2TF algorithm cannot be applied when the probability information of the sensor is unknown. The aim of this study is to exploit the T2TF algorithm even in the absence of the probability information of the sensor. The covariance is estimated using the recursive equations of the Kalman filter. In addition, a novel track-association approach using the total similarity is developed to improve association performance. The total similarity complements the defects of the track disposition and the estimated track history. Finally, by fusing the associated tracks using the estimated covariance, the T2TF algorithm is successfully applied to sensors with an unknown covariance. The fusion results are then evaluated using the correct association rate and the optimal subpattern assignment metric. The simulation results obtained show the superiority of the proposed algorithm under three scenarios.

Traffic signs play a crucial role in regulating traffic and facilitating cautious driving. Automatic traffic sign recognition is one of the key tasks in autonomous driving. Accuracy in the classification of traffic signs is therefore very important for the navigation of a vehicle. Here, a reliable and robust convolutional neural network (CNN) is presented for classifying these signs. The proposed classifier is a weighted multi-CNN trained with a novel methodology. It achieves a near state-of-the-art recognition rate of 99.59% when tested on the German traffic sign recognition benchmark dataset. Compared to the existing classifiers, the proposed one is a low-complexity network that recognises a test image in 10 ms when running on an NVIDIA 980 Ti GPU system. The results demonstrate its suitability and reliability in high-speed driving scenarios.

Real-time vehicle detection and counting of multiple types is a difficult problem. To solve this problem, this study presents an efficient method based on single shot detection (SSD) to construct a vehicle detection and counting system. The proposed method named Fast-SSD first combines the Slim ResNet-34 with Single Shot MultiBox Detector. Then the authors limit the location prediction at each cell in the feature map and modify the detection network. When the input size of the picture is 300 × 300, Fast-SSD achieves the accuracy of 76.7 mAP on the PASCAL visual object classes 2007 test set. The network can be implemented at the speed of 20.8 FPS based on the GTX650Ti. Furthermore, they obtain the centre point of each type of vehicle which is detected by the Fast-SSD model in the image and set the virtual loop detectors to specify the detection range. The number of vehicles is calculated when the centre of the vehicle passes the virtual loop detector. Results show that the vehicle detection accuracy achieves 99.3% and the classification accuracy is 98.9%.

In this study, a novel prediction model for the number of vacant parking spaces after a specific period of time is proposed based on support vector regression (SVR) with fruit fly optimisation algorithm (FOA). In the proposed model, the SVR parameters are initialised as the fruit fly population, and FOA is utilised to search the optimal parameters for SVR. Sufficient experiments within various scenarios, i.e. predicting the vacant parking space availability in parking lots with various capacities after various periods of time, have been conducted to verify the effectiveness of the proposed FOA-SVR prediction model. Three other commonly used prediction models, i.e. backpropagation neural network (NN), extreme learning machine and wavelet NN, are used as the comparison models. The experimental results show that the proposed FOA-SVR method has higher accuracy and stability in all the prediction scenarios.

This study presents results from an investigation into the effect of positive incentives on cycling behaviour among 1802 commuters in the Twente region of the Netherlands. The authors used an on-line survey, which included mock-up apps with incentives to commute to work by bicycle. They tested five reward schemes, namely social rewards (such as badges), in-kind gifts, money, competition, and cooperation. They used the survey data in a multinomial logit model to estimate to what extent travellers will use the app and increase their cycling frequency and which incentives they prefer. The model results show that respondents who sometimes cycle to work are more positive about incentive schemes than respondents who never cycle and that offering an app with in-kind gifts is probably most effective. Interestingly, non-cyclists are more likely to change their behaviour for a reward if they care about travel costs, while occasional cyclists are more likely to cycle more often in response to incentives if they care about attributes that are related to the cycling itself. This also depends on attitudes towards cycling and on socio-demographic variables.

Here, a hidden Markov model (HMM) and driver path preference (DPP)-based algorithm was proposed for floating car trajectory map matching. The algorithm focused on two improvements over existing HMM-based map matching algorithm: (i) the use of distance difference feature and average speed difference feature for transition probability calculation, which reasonably describe the context information between the two adjacent sampling points. It results in a more accurate matching capability; (ii) the DPP overcomes the shortcoming of feature attenuation in calculating the transition probability at low floating car sampling rates. It assures the matching accuracy of the algorithm at low sampling rates. The algorithm was evaluated using ground truth data and the results of the experiment show that the new transition probability significantly improves the matching capability. The proposed DPP can significantly help to maintain the matching accuracy under the condition of low sampling rates.

In this study, an analytical model for roadside unit (RSU) location is proposed. The objective of the model is to promote information propagation on two parallel roadways in vehicular ad hoc networks. The information propagation is achieved through multi-hop vehicle-to-vehicle and vehicle-to-infrastructure communication. The relationship between the information travel time from one RSU to another and the distance between the two RSUs are derived. The RSU location problem is formulated as a shortest path problem and then as an integer linear programming. The proposed model is applicable for traffic flow with a general vehicle headway distribution. A detailed comparison between the proposed model and an existing RSU location model is conducted through Monto Carlo simulation experiments under different vehicle headway distributions, traffic densities, budget sizes, and road separation distances. The results indicate that the proposed model outperforms the existing model in all cases. Moreover, it is shown that it is necessary to consider the general vehicle headway distribution and road separation in RSU locations that are focused on information propagation promotion.

Motion control problems remain to be fully solved for automated vehicles in dynamic traffic circumstances. Existing approaches usually first make a driving behaviour decision, then design a reference trajectory that may not match the vehicle dynamic constraints explicitly and finally adopt a local feedback control method to track the reference. Important commands may be lost or not well translated in the process of information exchange and transmission. Moreover, multiple methods specifically designed for different tasks may not cooperate well in one system. In this study, the authors propose a comprehensive predictive control method which can directly generate the control commands from the traffic circumstance and the vehicle dynamics, without involving any driving decision-making modules and any predefined reference trajectories. Virtual potential fields are introduced to model the traffic circumstance including the road boundaries, lane markings and moving obstacle vehicles. A model predictive control problem is formulated with the overall potential function and constraints including the vehicle dynamics and the safety distances between the ego vehicle and other vehicles. Lane keeping, lane changing, car following and overtaking driving behaviours are simulated in different scenarios. Results show that this method is capable of controlling the automated vehicle in different traffic circumstances.