The deep understanding of the behaviors of traffic participants is essential to guarantee the safety of automated vehicles (AV) in mixed traffic with vulnerable road users (VRUs). Precise trajectory prediction of traffic participants can provide reasonable solution space for motion planning of AV. Early works mainly focused on handcrafting the feature representation and designing complicated architectures in deep learning-based prediction models. However, these approaches overlooked the fact that different road users perceive the safety of the same interaction differently and also exhibit heterogeneous risk-taking styles. In this paper, we will develop a model for trajectory prediction based on risk-taking styles. The model accounts for the expected positions and occupancy of traffic participants in the surrounding environment. It consists of two sequential steps: risk-taking styles of multi-modal road users under interactive scenes are first clustered, and then reformulated in the heterogeneous graph model for trajectory prediction. The model is validated by the driving data collected on the urban road using a public dataset. Comparative experiments demonstrate that the proposed method can predict the trajectory of traffic participants much more accurately than the state-of-the-art methods.

@en

Modelling, predicting and analysing driver behaviours are essential to advanced driver assistance systems (ADAS) and the comprehensive understanding of complex driving scenarios. Recently, with the development of deep learning (DL), numerous driver behaviour learning (DBL) methods have been proposed and applied in connected vehicles (CV) and intelligent transportation systems (ITS). This study provides a review of DBL, which mainly focuses on typical applications in CV and ITS. First, a comprehensive review of the state-of-the-art DBL is presented. Next, Given the constantly changing nature of real driving scenarios, most existing learning-based models may suffer from the so-called “catastrophic forgetting,” which refers to their inability to perform well in previously learned scenarios after acquiring new ones. As a solution to the aforementioned issue, this paper presents a framework for continual driver behaviour learning (CDBL) by leveraging continual learning technology. The proposed CDBL framework is demonstrated to outperform existing methods in behaviour prediction through a case study. Finally, future works, potential challenges and emerging trends in this area are highlighted.

@en

Accurate recognition of driver behaviours is the basis for a reliable driver assistance system. This paper proposes a novel fusion framework for driver behaviour recognition that utilises the traffic scene and driver gaze information. The proposed framework is based on the graph neural network (GNN) and contains three modules, namely, the gaze analysing (GA) module, scene understanding (SU) module and the information fusion (IF) module. The GA module is used to obtain gaze images of drivers, and extract the gaze features from the images. The SU module provides trajectory predictions for surrounding vehicles, motorcycles, bicycles and other traffic participants. The GA and SU modules are parallel and the outputs of both modules are sent to the IF module that fuses the gaze and scene information using the attention mechanism and recognises the driving behaviours through a combined classifier. The proposed framework is verified on a naturalistic driving dataset. The comparative experiments with the state-of-the-art methods demonstrate that the proposed framework has superior performance for driving behaviour recognition in various situations.

@en

In urban environments, the complex and uncertain intersection scenarios are challenging for autonomous driving. To ensure safety, it is crucial to develop an adaptive decision making system that can handle the interaction with other vehicles. Manually designed model-based methods are reliable in common scenarios. But in uncertain environments, they are not reliable, so learning-based methods are proposed, especially reinforcement learning (RL) methods. However, current RL methods need retraining when the scenarios change. In other words, current RL methods cannot reuse accumulated knowledge. They forget learned knowledge when new scenarios are given. To solve this problem, we propose a hierarchical framework that can autonomously accumulate and reuse knowledge. The proposed method combines the idea of motion primitives (MPs) with hierarchical reinforcement learning (HRL). It decomposes complex problems into multiple basic subtasks to reduce the difficulty. The proposed method and other baseline methods are tested in a challenging intersection scenario based on the CARLA simulator. The intersection scenario contains three different subtasks that can reflect the complexity and uncertainty of real traffic flow. After offline learning and testing, the proposed method is proved to have the best performance among all methods.@en

Precisely modeling interactions and accurately predicting trajectories of surrounding vehicles are essential to the decision-making and path-planning of intelligent vehicles. This paper proposes a novel framework based on ensemble learning to improve the performance of trajectory predictions in interactive scenarios. The framework is termed Interactive Ensemble Trajectory Predictor (IETP). IETP assembles interaction-aware trajectory predictors as base learners to build an ensemble learner. Firstly, each base learner in IETP observes historical trajectories of vehicles in the scene. Then each base learner handles interactions between vehicles to predict trajectories. Finally, an ensemble learner is built to predict trajectories by applying two ensemble strategies on the predictions from all base learners. Predictions generated by the ensemble learner are final outputs of IETP. In this study, three experiments using different data are conducted based on the NGSIM dataset. Experimental results show that IETP improves the predicting accuracy and decreases the variance of errors compared to base learners. In addition, IETP exceeds baseline models with 50% of the training data, indicating that IETP is data-efficient. Moreover, the implementation of IETP is publicly available at https://github.com/BIT-Jack/IETP.

@en

All drivers have individual ways of driving. Still, there are groups of drivers with more or less similar characteristics. In this research, 28 drivers from Chengdu city (P.R. China) participated in an experiment where car following behaviour was measured with GPS devices. In every measured trip there was a leading and following vehicle both equipped with a GPS device. Drivers are classified based on a Driver Behaviour Questionnaire and observed acceleration and deceleration behaviour. The result shows four distinct classes of drivers: macho drivers, careful/inexperienced drivers, smooth going/ professional drivers, and experienced/fast drivers. Drivers in the different classes give different emission of air pollution and fuel consumption. Saturation flows are determined from the trajectories and vary between different driver types. The measured trajectories have been analysed in detail to determine some parameters for the Wiedemann 74 model. Most default parameters in the VISSIM program appear to be unsuited for the simulation of driving behaviour measured in the experiment. The emissions and fuel consumption calculated by a simulation model with default parameters are not consistent with the empirical data. The calibration done for different driver types shows that several model parameters are significantly different for the different driver classes.@en

The recently proposed b-modulation method for nonlinear Fourier transform-based fiber-optic transmission offers explicit control over the duration of the generated pulses and therewith solves a longstanding practical problem. The currently used b-modulation however suffers from a fundamental energy barrier. There is a limit to the energy of the pulses, in normalized units, that can be generated. In this paper, we discuss how the energy barrier can be shifted by proper design of the carrier waveform and the modulation alphabet. In an experiment, it is found that the improved b-modulator achieves both a higher Q-factor and a further reach than a comparable conventional b-modulator. Furthermore, it performs significantly better than conventional approaches that modulate the reflection coefficient.

@en

The importance of travel time reliability in traffic management, control, and network design has received a lot of attention in the past decade. In this paper, a network travel time distribution model based on the Johnson curve system is proposed. The model is applied to field travel time data collected by Automated Number Plate Recognition (ANPR) cameras. We further investigate the network-level travel time reliability by connecting the network reliability measures such as the weighted standard deviation of travel time rate and the weighted skewness of travel time rate distributions with network traffic characteristics (e.g., the network density). The weighting is done with respect to the number of signalized intersections on a trip. A clear linear relation between the weighted average travel time rate and the weighted standard deviation of travel time rate can be observed for different time periods with time-varying demand. Furthermore, both the weighted average travel time rate and the weighted standard deviation of travel time rate increase monotonically with network density. The empirical findings of the relation between network travel time reliability and network traffic characteristics can be possibly applied to assess traffic management and control measures to improve network travel time reliability.

@en

Fuel consumption and atmospheric pollution emissions of vehicles depend on driving conditions, the characteristics of the driver and the car. The influence of driving style on the environmental aspects of a car journey has been investigated. Driver characteristics were determined by a Driver Behaviour Questionnaire and observed acceleration and deceleration behaviour. That results in four types of drivers with similar characteristics within a type group. We measured 56 trajectories of 28 drivers using GPS devices. The measurements were done on a route of 8.4 km in an urban environment in Chengdu (PR China). From the trajectories, the emissions and fuel consumption were determined with the Comprehensive Modal Emissions Model. The results were related to the traffic control along the journey resulting in fuel consumption and emissions per stop and per second idling. There are significant differences in saturation flow, emissions and fuel consumption between different driver types. Cautious, novice drivers have the lowest emission and fuel consumption and give the lowest saturation flow and have the lowest cruise speed; experienced smooth driving drivers give a high saturation flow while keeping fuel consumption and emissions also low. Aggressive experienced drivers have a high saturation flow and fuel consumption / emissions. Therefore, microscopic traffic models that simulate emissions and fuel consumption should take the differences between driver types into account.

@en