Search

Monocular road segmentation is a fundamental component of autonomous driving. High accuracy has always been a focus of the community for this task. However, the exploration of high-speed methods is still lacking. This limits the prevalence of the monocular road segmentation method. In this paper, an efficient encoder-decoder network is proposed for segmenting the road from a single image. Specifically, to achieve high-speed inference, we leverage a shallow encoder for feature extraction and design a lightweight decoder for feature resolution recovery. To avoid the accuracy drop caused by network simplification, we introduce an efficient asymmetric dilated block to improve the ability of features to distinguish drivable/nondrivable roads to maintain the performance without excess computation. Experiments on three datasets verify the effectiveness of our method. On KITTI-Road, our approach ranks second among all monocular methods and even achieves a competitive performance compared to multimodal methods, with an inference speed over 135 FPS on a single TITAN Xp, which is at least 20 times faster than the state-of-the-art monocular method. Furthermore, our approach can be deployed on the embedded platform Jetson TX2 and achieves a speed of over 80 FPS. Codes will be available at https://github.com/gongshichina/FastRoadSeg. [ABSTRACT FROM AUTHOR]

Although high-accuracy networks have been applied to semantic segmentation at present, their inference speeds remain slow. A trade-off between accuracy and speed is demanded for real-time applications. To approach this problem, we propose Multi-Feature Fusion Network (MFNet) with real-time efficient prediction capacity. MFNet adopts three branches (attention, semantic and spatial information) to capture low-level and high-level features. Additionally, MFNet exerts asymmetric factorized (AF) blocks to extract local and long-range features. As a result, without any pre-training or post-processing, MFNet using only 1.34 M parameters, achieves 72.1% mean intersection over union (mIoU) on the Cityscapes test set at a speed of 116 frames per second (FPS), with $512\times 1024$ high resolution on a single Titan Xp graphics card. Our network’s performance stands out from other state-of-the-art networks on four datasets (Cityscapes, CamVid, KITTI, and Gatech). [ABSTRACT FROM AUTHOR]

In this work, we focus on real-time 3D reconstruction or localization and mapping for outdoor scene using an aerial vehicle called quadcopter. Quadcopter provides the advantages of high flexibility and wide view field in spatial movement. However, existing feature-based and direct methods (using dense or semi-dense approach) are not suitable for outdoor environment, in which multiple challenging scenarios arise such as lighting variance, jittering views, high-speed and non-smooth flight trajectory. The main reason is that the existing methods rely on the assumption of brightness constancy across multiple images and only raw pixel intensities are employed for direct image alignment. In order to tackle these scenarios, a novel method called Feature-based Direct Tracking and Mapping (FDTAM) is proposed, which i) incorporates an efficient binary feature descriptor into direct image alignment module to tackle the challenging scenarios, such as drifting issue under lighting variance problem; ii) applies semi-dense approach to obtain high reconstruction quality; iii) provides a framework with low computational complexity for real-time reconstruction. Compared to other state-of-the-art feature-based and direct methods, our proposed method is shown to tackle the challenging scenarios and improve the accuracy and robustness even in CPU (rather than GPU) platform. [ABSTRACT FROM AUTHOR]

The present work aims to improve the communication security of Internet of Vehicles (IoV) nodes in intelligent transportation through studying the safety of IoV in smart transportation based on Blockchain (BC). An IoV DTs model is built by combining big data with Digital Twins (DTs). Then, regarding the current IoV communication security issues, a secure communication architecture for the IoV system is proposed based on the immutable and trackable BC data. Besides, Wasserstein Distance Based Generative Adversarial Network (WaGAN) model constructs the IoV node risk forecast model. Because the WaGAN model calculates the loss function through Wasserstein distance, the learning rate of the model accelerates remarkably. After ten iterations, the loss rate of the WaGAN model is close to zero. Massive in-vehicle devices in IoV are connected simultaneously to the base station, causing network channel congestion. Therefore, a Group Authentication and Privacy-preserving (GAP) scheme is put forward. As users increase during authentication, the GAP scheme performs better than other authentication access schemes. In summary, the Intelligent Transportation System driven by DTs can promote intelligent transportation management. Besides, introducing BC into IoV can improve access control’s accuracy and response efficiency. The research reported here has significant value for improving the security of the information sharing of the IoV. [ABSTRACT FROM AUTHOR]

In this paper, we present a deep learning model for high-accuracy, high-speed detection of vehicle taillights in traffic. The model consists of three major modules: the lane detector, the car detector, and the taillight detector. Unlike most previously proposed algorithms where hand-coded schemes are used, we have adopted a data-driven approach. This data-driven scheme was implemented in both the car and taillight detection modules. First, we used an intricately designed lane detection module, then we adopted the Recurrent Rolling Convolution (RRC) architecture and tracking mechanism for detecting car boundaries. Subsequently, we used the same RRC architecture to extract the taillight regions of the detected cars. The lane detection and car detection modules improve both the speed and detection rate of the final taillight detection. The robustness of our model was verified using datasets from Sungkyunkwan University (SKKU) as well as the Karlsruhe Institute of Technology and Toyota Technological Institute (KITTI). Our model works well even in hostile conditions. It achieves detection rates as high as 99% in testing with the SKKU dataset. When using the KITTI 2D Object dataset, the model achieves a taillight detection rate of 86%. The model achieves 100% taillight detection rate on a certain, small subset of the KITTI Tracking dataset. [ABSTRACT FROM AUTHOR]

Video object detection is a tough task due to the severe appearance degradation caused by rapid motion, sudden occlusion or rare poses. The great challenge facing video object detection is the simultaneous requirements on both accuracy and speed because the pursuit of one aspect usually causes significant expense to the other. Most existing methods mainly focus on improving detection accuracy with little attention to computationally efficient solutions, and thus they are impractical for many real-world applications. This motivates us to develop a real-time and high-accuracy video object detection method. In this paper, we propose a novel video object detector, called FastVOD-Net, which can yield highly accurate detection results at real-time speed. Specifically, we first develop a temporally-cascaded deformable alignment (TCDA) module to model the object displacements induced by video motion. Then, we introduce another two modules, namely spatially-refined temporal aggregation (SRTA) and attention-guided semantic distillation (AGSD), to improve the appearance feature of the currently processed frame and enhance the semantic representation of non-keyframes, respectively. For keyframe scheduling, we design an adaptive keyframe selection scheduler (AKSS) to adjust the keyframe interval online, making the keyframe usage more rational. On one hand, the characteristics of our FastVOD-Net enable it to sparsely perform expensive feature extraction, which significantly reduces the computational cost and thus guarantees real-time speed. On the other hand, the collaboration of the above tightly-coupled modules and adaptive keyframe scheduler makes FastVOD-Net fully exploit inter-frame temporal dependencies and thus guarantees high accuracy. Experiments on the ImageNet VID dataset show that our FastVOD-Net achieves 79.3% mAP at 29.6 fps or 81.2% mAP at 23.0 fps on an Nvidia RTX 2080 Ti GPU, which is the state-of-the-art performance in real time. [ABSTRACT FROM AUTHOR]

The dynamic spatiotemporal characteristics of queues at urban intersections are crucial to traffic operation tasks such as signal performance measure and signal optimization. This paper addresses the high time-resolution estimation of queue profile at urban signalized intersection using Extended Kalman Filtering (EKF) with data of connected vehicles (CVs). The main features of this work are as follows: (i) a machine learning method was applied to construct a dynamic shockwave propagation model based on shockwave theory and historical data of CVs; (ii) a heuristic approach was proposed to measure the shockwave speed for use in EKF; (iii) an urban queue estimator was designed to combine the dynamic shockwave propagation model and real-time shockwave information via EKF to deliver second-by-second queue profile estimates. The queue estimator does not require any priori information about vehicle arrival patterns and the market penetration rate (MPR) of CVs. The performance and robustness of the queue estimator were evaluated using both simulation and real-world CV data. The results show that the method can provide satisfactory queue estimation results at various MPR levels of CVs, with the estimation error of 2.5 vehicles at the MPR of 5%, and of 0.5 vehicle at the MPR of 40%. [ABSTRACT FROM AUTHOR]

Modern complexities associated with an arterial traffic makes existing safety prediction methods insufficient to meet desired standards required by recent developmental needs. This paper proposes an enhanced active safety prediction method based on big-data approach and Stacked AutoEncoder-Gated Recurrent Unit. Firstly, the big-data technology is used to construct a dynamic identification model to recognize real-time operation state and risk state. Secondly, the Stacked AutoEncoder-Gated Recurrent Unit is used to predict a level of safety based on associated recognition results. This paper uses data from working days of Sunset Boulevard, California, from January $1^{\mathrm{st}}$ , 2020, to February $28^{\mathrm{th}}$ , 2020. The results of analysis show that the accuracy of the proposed dynamic recognition model reaches 98.92%, which is better than existing models such as random forest, K-nearest neighbor, and naïve Bayes models. In addition, it is found that the Stacked AutoEncoder-Gated Recurrent Unit can achieve a prediction accuracy of 95.157% and has significant advantages in terms of efficiency. The proposed methods will provide feasible solutions for actively monitoring safety levels. [ABSTRACT FROM AUTHOR]

Estimating when pedestrians cross the street is essential for intelligent transportation systems. Accurate, real-time prediction is critical to ensure the safety of the most vulnerable road users while improving passenger comfort. In the present work, we developed a model called Pedestrian Graph +, an improvement of our previous work, Pedestrian Graph, which predicts pedestrian crossing action in urban areas based on a Graph Convolution Network. We integrated two convolutional modules in the new model that provide additional context information (cropped images, cropped segmentation maps, ego-vehicle velocity data) to the main Graph Convolutional module, thus increasing accuracy. Our model is faster and smaller than other state-of-the-art models, achieving equivalent accuracy. Our model is faster than state-of-the-art models, with an inference time of 6 ms (on a GTX 1080) and low memory consumption (0.3 MB). We tested our model on two datasets, Joint Attention in Autonomous Driving (JAAD) and Pedestrian Intention Estimation (PIE), achieving 86% and 89% accuracy, respectively. Another contribution of our work is the ability to dynamically process almost any input size in the time domain without significant loss of accuracy. It is possible due to the fully convolutional property of ConvNets. Our models and results are available at https://github.com/RodrigoGantier/Pedestrian_graph_plus. [ABSTRACT FROM AUTHOR]

This study focuses on the dynamic route control (DRC) problem for connected and automated vehicles (CAVs). The problem is usually solved using two popular methods, i.e., dynamic shortest path (DSP) and dynamic system optimal assignment (DSO). However, the DSP algorithm is unproductive under congested conditions, while the DSO, although can achieve system optimal, possesses considerable algorithm complexity. Furthermore, the two methods are usually solved in a centralized system due to the need for global information of the entire network, which is both communicationally and computationally ineffective. With the emergence of mobile edge computing (MEC), a new distributed DRC method only relying on local information is more achievable and practical. In this context, this research developed a novel DRC algorithm based on the distributed Backpressure (BP) principle for the MEC-enabled CAVs. The BP herein is modified as a function of the real-time density, which avoids the unrealistic point queue assumption in the original BP algorithm. In addition, the real-time traffic state is pre-identified to determine whether re-routing is essential, thereby reducing the possibility of CAVs guided to unnecessarily long routes. Results from the case study simulated by the traffic simulation software Simulation of Urban MObility (SUMO) indicate that the proposed method outperforms the BP method without congestion identification in low-demand cases, and the performance is significantly better than the DSP while approximating the DSO in congested cases. More importantly, both communicational cost and algorithm execution time was greatly reduced when compared with the DSO. [ABSTRACT FROM AUTHOR]

The regular detection of pavement cracks is critical for life and property security. However, existing deep learning-based methods of crack detection face difficulties in terms of data acquisition and defect counting. An automatic intelligent detection and tracking system for pavement cracks is proposed. Our system is formed of a pavement crack generative adversarial network (PCGAN) and a crack detection and tracking network called YOLO-MF. First, PCGAN is used to generate realistic crack images, to address the problem of the small number of available images. Next, YOLO-MF is developed based on an improved YOLO v3 modified by an acceleration algorithm and median flow (MF) algorithm to count the number of cracks. In a counting loop, our improved YOLO v3 detects cracks and the MF algorithm tracks the cracks detected in a video. This improved algorithm achieves the best accuracy of 98.47% and F1 score of 0.958 among other algorithms, and the precision-recall curve was close to the top right. A tiny model was developed and an acceleration algorithm was applied, which improved the detection speed by factors of five and six, respectively. In on-site measurement, three cracks were detected and tracked, and the total count was correct. Finally, the system was embedded in an intelligent device consisting of a calculating module, an automated unmanned aerial vehicle, and other components. [ABSTRACT FROM AUTHOR]

Modulus is a critical parameter for evaluating the bearing capacity and remaining life of pavements. The prevalent modulus back-calculation method by using the pavement surface deflections captured from falling weight deflectometer (FWD) could characterize the overall bearing capacity of pavements. However, a non-uniqueness issue may occur when identifying the modulus of each layer from the perspective of mathematics. Alternatively, this paper presents a novel application to real-time modulus evaluation based on asphalt pavement health monitoring with built-in sensors. First, the sensor layout is optimized on the basis of the theoretical relationship between modulus and mechanical response inside the pavement. The proposed sensor layout, including input (random loads), output (mechanical responses), and environmental sensing modules, is illustrated in detail. Then, the procedure for real-time modulus evaluation is proposed, and the convergence and uniqueness of modulus evaluation results are investigated. Lastly, a case study of the realistic asphalt pavement health-monitoring system is conducted to demonstrate the effectiveness of the application to real-time modulus evaluation. The results indicate that the proposed modulus evaluation method has a potential to identify the modulus of each pavement structural layer in real time with a convergent and unique solution under a realistic traffic load. The evaluation process has the advantage of not interfering with traffic compared with the prevalent FWD-based modulus evaluation method. The pavement health monitoring with built-in sensors is recommended for newly built roads to facilitate long-term real-time performance assessment and maintenance decision making. [ABSTRACT FROM AUTHOR]

Vehicle trajectory prediction is essential for enabling safety-critical intelligent transportation systems (ITS) applications used in management and operations. While there have been some promising advances in the field, there is a need for modern deep learning algorithms that allow real-time trajectory prediction on embedded IoT devices. This article presents DeepTrack, a novel deep learning algorithm customized for real-time vehicle trajectory prediction and monitoring applications in arterial management, freeway management, traffic incident management, and work zone management for high-speed incoming traffic. In contrast to previous methods, the vehicle dynamics are encoded using Temporal Convolutional Networks (TCNs) to provide more robust time prediction with less computation. DeepTrack also uses depthwise convolution, which reduces the complexity of models compared to existing approaches in terms of model size and operations. Overall, our experimental results demonstrate that DeepTrack achieves comparable accuracy to state-of-the-art trajectory prediction models but with smaller model sizes and lower computational complexity, making it more suitable for real-world deployment. [ABSTRACT FROM AUTHOR]

This paper proposes a 3D LiDAR aided global navigation satellite system (GNSS) non-line-of-sight (NLOS) mitigation method due to both static buildings and dynamic objects. A sliding window map describing the environment of the ego-vehicle is first generated, based on real-time 3D point clouds from a 3D LiDAR sensor. Subsequently, the NLOS receptions are detected based on the sliding window map using a proposed quick searching method which eliminates the reliance on the initial guessing of the position of the GNSS receiver. Instead of directly excluding the detected NLOS satellites from further estimating the position, this paper rectifies the pseudo-range measurement model by (1) correcting the pseudo-range measurements if the reflecting point of the NLOS signals is detected within the sliding window map, and (2) remodeling the uncertainty in the NLOS pseudo-range measurement using a novel weighting scheme. The performance of the proposed model was experimentally evaluated in several typical urban canyons in Hong Kong using an automobile-level GNSS receiver. Furthermore, the potential of the proposed NLOS mitigation method in GNSS and the integration of inertial navigation systems were evaluated via factor graph optimization. [ABSTRACT FROM AUTHOR]

This paper presents the design and analysis of an automated traffic signage inventory management system employing the radio-frequency identification (RFID) technology. Traffic signage inventory management faces the operational challenge of recognizing a large number of traffic signs spread in a large area without obvious identifications on them. Thanks to the remote sensing ability of RFID, the operational efficiency can be significantly improved by scanning traffic signs remotely in motion. The system consists of an in-vehicle RFID reader and a handheld RFID reader, both of which connect to mobile applications with user-friendly interfaces. Additionally, user-friendly control software and a database are developed. In this paper, the adequacy of the RFID technology for the task is assessed. On the theoretical side, this paper investigates the radio-wave propagation model and the antenna radiation range. Additionally, the influences of the vehicle speed and the sign-to-reader distance on the reading performance are studied and empirical tests of the tag positioning are conducted. Finally, a system protocol is validated by a field test. [ABSTRACT FROM AUTHOR]

Traffic congestion is an inescapable problem that frustrates drivers in megacities. Although there is hardly a way to eliminate the congestion, it is possible to mitigate the impact through predictive methods. This paper develops a data-driven optimization approach for the dynamic shortest path problems (DSPP), considering traffic safety for urban navigations. The dynamic risk scores and travel times at different times and locations are estimated by the Safe Route Mapping (SRM) methodology and Long Short-Term Memory (LSTM) with Autoencoder, respectively, where possible variations in the future are considered. The DSPP is formulated as a mixed-integer linear programming problem under risk constraints to minimize the total travel cost, defined as the weighted sum of distance and travel time. To improve the efficiency of the DSPP, we design an improved tabu search with alternative initial-solution algorithms to accommodate various problem scales. Moreover, subgraph and self-adaptive insertion techniques are adopted as acceleration strategies to enhance computational efficiency further. Numerical experiments investigate the computational performance and the solution quality of our algorithm. The result shows satisfactory solution quality and computational efficiency with the proposed acceleration strategies compared to the CPLEX solver, a label-setting algorithm, and a state-of-the-art algorithm. Our algorithm can also compete with Google Maps regarding the travel cost in a real network in Manhattan, NY, USA, which is promising for Urban Navigations. [ABSTRACT FROM AUTHOR]

Condition monitoring of road surfaces has acquired a lot of attention in the field of computer vision throughout the years. It is due to two main reasons; firstly, it produces safety and comfort to the community, and secondly, it causes less damage to the vehicles for an advanced driver assistance system (ADAS). To this extent, this article aims to present a real-time vision-based approach that automatically segments the road anomalies from the drivable area. An Intel RealSense D435 depth camera has been employed to capture RGB and depth (RGB-D) images of the road surface. An unsupervised learning method based on diffusion process has been employed to learn the affinity matrix of the RGB-D data and spectral clustering has been applied on the updated affinity matrix to cluster the road images. Image multiplex visibility graphs of the input sensor data are diffused by regularized diffusion process (RDP) to update the affinity matrix followed by generation of saliency map of the road surfaces. The prime motive to employ RDP is to use the graph Laplacian as a tool for similarity measurement for preserving the manifold structure. Qualitative and quantitative results reveal the efficacy of the proposed system with state-of-the-art methods on our RGB-D dataset. Benchmark datasets (KITTI and Cityscapes) are also used to validate the proposed method for segmentation of the drivable area for an intelligent transportation system. [ABSTRACT FROM AUTHOR]

In the conventional variable speed limit (VSL) strategy, the control area is fixed under confined conditions. With the facility of the roadside unit, the control area for the VSL will vary in real-time in the connected environment. This study proposed an extended model-based VSL controller to improve traffic efficiency in the connected environment. The controller was designed based on the scheme of model predictive control (MPC). In the controller, an extended cell transmission model (CTM) with variable-length cells was established. The cell length variation was introduced to describe the characteristics of the variable control area. The optimizer in MPC is an improved Genetic Algorithm. A numerical simulation was conducted to show how the method, with the cooperation of variable speed limit and variable control area, alleviates the shock waves. The performance of the method was compared with a conventional VSL without a variable control area. The results show that the extended model-based VSL controller reduces the total travel time by 14.57% compared with the conventional VSL controller. The compared results illustrate that the proposed VSL controller can effectively resolve shock waves produced by the incident. [ABSTRACT FROM AUTHOR]

This paper proposes a vehicle trip chain information-based hierarchical stochastic energy management system (Hi-SEMS) for a microgrid integrated with highly uncertain plug-in electric vehicles (PEVs) fleet, renewable energy sources, and loads. The two-layer Hi-SEMS scheme aims at optimal scheduling of PEV fleet to counteract the operational uncertainties encountered in real-time operation and minimizing the microgrid daily operating cost comprising the fuel cost and the emission cost. In the proposed work, the realistic behavior of PEVs fleet is modeled using a stochastic trip chain information to accurately forecast the charge/discharge demand taking into account the driver’s behavior as foundation. Realistic statistical trip data, multi-location charge/discharge of PEVs fleet to include coupled spatial-temporal dynamics of heterogeneous PEVs fleet is considered for intelligent charge/discharge scheduling. The effectiveness of the proposed strategy is verified on modified LV CIGRE and modified IEEE 33 bus radial distribution test networks and is compared with four different cases on each of the two test systems. The simulation results exhibit that the proposed Hi-SEMS outperforms the other four cases in terms of daily operating cost, the incentive to the PEVs owner, and robustness to multiple uncertainties. [ABSTRACT FROM AUTHOR]

For deployment on an embedded processor for distracted driver classification, the model should satisfy the demand for both high accuracy, real-time inference, and limited storage resources. Conventional deep CNN models such as VGG, ResNet, DenseNet, often aim for high accuracy, making their model heavy for an embedded system with limited memory space and computing resources. In contrast, lightweight models are greatly compressed but at a significant sacrifice of accuracy. To bridge this gap, we propose an instance-specific multi-teacher knowledge distillation model (IsMt-KD) to learn more accurate, speedy, and lightweight CNNs for distracted driver posture classification. Specifically, in multi-teacher knowledge distillation, most of the current approaches either randomly select a teacher model and apply the prediction of such teacher model as the soft-label or allocate an equal weight to every teacher model and average all the predictions of the teachers as the soft label. In this paper, we observe that, when facing the same instance, the outputs of different teachers vary greatly, in which some teachers can predict it right whereas the others may give pretty high probabilities to the irrelevant classes. Thus, it is inappropriate to set fixed weights or the same weights for teachers. To this end, a simple yet effective instance-specific teacher grading module is designed to dynamically assign weights to teacher models based on individual instances. In this way, we can dynamically distill the knowledge from multiple teachers by considering both instance-specific high-level and instance-specific intermediate-level information. Our extensive experimental results on AUC and StateFarm datasets, and our implementation on edge hardware platforms including HUAWEI MediaPad c5 and Nvidia Jetson TX2, verify the effectiveness and feasibility of our approach. [ABSTRACT FROM AUTHOR]

The increasing popularity of on-demand ride-hailing services has introduced research problems, including those related to fleet management. Many existing fleet management approaches discretize a spatiotemporal distribution to reduce the problem complexity. Although spatial discretization greatly simplifies the problems, this approximation results in a suboptimal solution. To address this limitation, we present a novel spatial discretization-free, particle-based formulation of fleet management. Our model requires only temporal discretization. The nodes in our spatiotemporal network model represent the positions of the fleet and ride requests over the time horizon. Under this representation, a taxi dispatch problem can be solved by the minimum-cost maximum-flow (MCMF) algorithm. We show some structural properties that disclose a type of monotonicity by applying the MCMF algorithm to our problem. The obtained dispatch solution maximizes the number of served requests from these structural properties at least at the most current time step, regardless of the demand prediction quality. Moreover, these properties enable the decomposition of the MCMF algorithm without sacrificing optimality, and the decomposed multistage maximum-flow algorithm is significantly faster than the MCMF algorithm. Furthermore, we propose an efficient method that alleviates the dearth of input data for service providers who lack sufficient computing power to calculate substantial travel time information. We demonstrate the performance of our method by using data collected in New York City, USA. Our simulation studies indicate that a service provider can fulfill 74–85% of the ride requests within 2 minutes. [ABSTRACT FROM AUTHOR]

Under extreme conditions, vehicle states change rapidly between stable and unstable, resulting in dynamic requirements for the vehicle’s overall safety stability. Simultaneously, the coupled nonlinear characteristics of vehicle dynamics cannot be ignored in controller design. To address the above problems and improve vehicle longitudinal and lateral stability integrally, an envelope-based model predictive control (MPC) strategy with dynamic objectives is proposed for four-wheel independent motor-drive electric vehicles (4WIMD EVs). First, according to the current driving behavior and the collected road information, the envelope control regions concerning vehicle side-slip angle and yaw rate are obtained online, and divided into stable, critically stable, and instable regions with different safety requirements. Then, the safety dynamic requirements are constructed in the designed MPC-based control structure. A nonlinear vehicle dynamics model with a combined-slip tire model, which integrates the longitudinal and lateral dynamics, is utilized to predict vehicle states. The switching of requirements is reflected in the variation of weighting factors and constraint values. Finally, CarSim and Matlab/Simulink co-simulation, and hardware-in-the-loop simulation test results show better satisfactory performance in improving overall vehicle stability under extreme driving conditions. [ABSTRACT FROM AUTHOR]

Video object segmentation (VOS) is the fundamental problem of vision-based intelligent transportation, and many VOS algorithms relying on inference from reference masks have been proposed. Due to the inherent defects of the inference strategy and the complex changes of targets, VOS methods that perform well on public datasets are usually ineffective in airport scenarios. We propose a spatiotemporal alignment network (STA-Net) that makes use of Automatic Dependent Surveillance-Broadcast (ADS-B) data as prior information to guide the long-term segmentation of aircraft. ADS-B is an airport-specific signal, which indicates the location of aircraft in real time. Based on ADS-B, we continuously generate new reference masks instead of using previous masks for inference, which greatly reduces the accumulation of inference errors. To achieve this, previous masks of each aircraft are aligned on the temporal domain based on the position information in ADS-B. All temporally-aligned masks are compared, and the one most similar to the current instant is reserved. This mask is both temporally and spatially aligned; hence it is a better reference mask for inference. Aligned masks are updated every time new ADS-B data arrive, so that they can support long-term inference. With the selected mask as a reference, aircraft of interest are segmented within a unified encoder-decoder framework over the long term. Experiments on a benchmark dataset and in a real airport scenario verify the effectiveness of the presented method. [ABSTRACT FROM AUTHOR]

Obstacles in front of a train pose a significant threat to traffic safety, and many accidents happen under shunting mode when the speed of a train is below 45 km/h. The existing track object–detection algorithms encounter difficulty in balancing the detection precision and speed in shunting mode. Additionally, their accuracy is insufficient, particularly for small objects in complex environments. To address these problems, we propose a stable lightweight feature extraction and adaptive feature fusion network for real-time detection of obstacles in railway traffic scenarios to ensure driving safety. The proposed network consists of three modules. The stable bottom feature extraction module reduces the computational load and extracts more image information stably. The lightweight feature extraction module improves feature extraction using a simple and effective network. The enhanced adaptive feature fusion module fuses the image and original features, improving the multiscale detection accuracy under complex environments, particularly in the case of small objects. With a default input size of 416 $\times 416$ pixels (px), the proposed method achieves a detection speed of 81 FPS and a mean average precision of 94.75% for the railway traffic dataset as well as a detection speed of 78 FPS (26 FPS faster and 0.47% higher than those of YOLOv4, respectively) and a mean average precision of 42.5% for MS COCO. This indicates its potential for real-world railway object detection and other multi-target detection tasks. Additionally, the experimental results based on PASCAL VOC2007 and VOC2012 indicate that the proposed approach is considerably better than the state-of-the-art models. [ABSTRACT FROM AUTHOR]

For autonomous vehicles (AVs), an intelligent awareness system is a fundamental task that provides crucial information on the driving environment. The main techniques for self-driving systems include solving tasks like self-localization, parsing the driving road, and understanding objects, enabling the system to reason and act. However, compared to other AVs, like cars, such robust awareness systems are not yet developed for inland autonomous vessels. Therefore, this paper proposes a system to accurately delimit safe navigation areas by simultaneously locating and mapping the environment. First, we construct the first open-source dataset; the InlandAutoDetect dataset comprises 3,377 images comprehensively labeled for object detection in a fluvial domain with almost 30,000 objects annotated. Second, we analyze and compare the results of nine deep learning perception models adapted to the inland environment in accuracy and run-time speed. The best one, namely Retinanet, is selected, and its different variants are tested by modifying the feature extractor block. Then, we propose the most suitable configuration for inland navigation. Finally, the selected model results are integrated into the complete system to delimit the safe sailing area. The achieved accuracy rate is above 88%, and the run-time speed rate is 0.8 seconds per image. Hence, the performance evaluations show the proposed system’s robustness and effectiveness to give accurate results and fulfill real-time operation requirements. [ABSTRACT FROM AUTHOR]

The availability of vehicle interaction data, which is obtained by an in-vehicle forward collision warning system, including spacing between the leading and the following vehicle and time-to-collision, provides a valuable opportunity to predict crash risks in real time. When this opportunity is combined with connected vehicle technologies including vehicle-to-vehicle wireless communications, it is expected that more effective crash prevention would be achievable by providing predictive warning information as a part of proactive traffic safety management (PTSM). The purpose of this study is to develop a more reliable in-vehicle warning information provision strategy based on the prediction of crash risks using vehicle interaction data. A crash risk prediction model based on a long short-term memory was able to predict the crash risk after 3 seconds with a mean absolute percentage error of 8% using the data for the past 5 seconds. The predicted crash risk data were applied to derive the optimal threshold for triggering in-vehicle warning information, which is the essence of the proposed warning provision strategy. This study defined three indicators to evaluate the reliability of warning information: correct detection rate (CDR), detection failure rate (DFR), and information provision rate (IPR). An exemplar analysis result showed that the optimal threshold to minimize IPR in a situation where CDR and DFR are 100% and 0%, respectively, was identified as 0.69. The proposed methodology that predicts crash risks in real time and provides V2V-based warning information in a more proactive manner is expected to mitigate the crash risk significantly. [ABSTRACT FROM AUTHOR]

This paper studies the joint problem of optimal pricing, charging scheduling and rebalancing for one-way car sharing systems. Firstly, a price driven customers travel mechanism is given which represents the customers ride demand as a linear function of the ride price. Then, a joint framework ensuring load balance and profit maximization of the system is introduced, in which the migration of vehicles and customers in the system is captured by fluid models, while the charging procedure of electric vehicles in the charging stations is described by multi-server queues. The well-posedness and equilibrium of the fluid model are analyzed, which yields the minimum fleet size needed. Next, a rebalancing method is derived to maximize the operating income while the stations operate at an equilibrium state, i.e., vehicles available and no waiting customers. To make the system better adapt to the changing customers demand in highly dynamic environments, a real-time rebalancing policy is developed by introducing a customer arrival predicting scheme based on exponential smoothing. Numerical experiments and case studies in Chengdu show that the proposed method can significantly improve the quality of service and rebalancing performance. [ABSTRACT FROM AUTHOR]

Single-image depth estimation represents a longstanding challenge in computer vision and although it is an ill-posed problem, deep learning enabled astonishing results leveraging both supervised and self-supervised training paradigms. State-of-the-art solutions achieve remarkably accurate depth estimation from a single image deploying huge deep architectures, requiring powerful dedicated hardware to run in a reasonable amount of time. This overly demanding complexity makes them unsuited for a broad category of applications requiring devices with constrained resources or memory consumption. To tackle this issue, in this paper a family of compact, yet effective CNNs for monocular depth estimation is proposed, by leveraging self-supervision from a binocular stereo rig. Our lightweight architectures, namely PyD-Net and PyD-Net2, compared to complex state-of-the-art trade a small drop in accuracy to drastically reduce the runtime and memory requirements by a factor ranging from $2\times $ to $100\times $. Moreover, our networks can run real-time monocular depth estimation on a broad set of embedded or consumer devices, even not equipped with a GPU, by early stopping the inference with negligible (or no) loss in accuracy, making it ideally suited for real applications with strict constraints on hardware resources or power consumption. [ABSTRACT FROM AUTHOR]

The evolution of headways (time intervals between consecutive vehicles) is an important performance indicator of public transportation systems. In this paper, we develop a data processing system that processes the expected arrival time data provided by the corresponding APIs of Madrid and London bus public transport systems with the aim of modeling and supervising their evolution along the time to detect their operational anomalies. By employing only estimates of the time headways between buses, we construct joint statistical modeling along the time of these headways within a stochastic process framework. Then, this model is employed to design a deviation detection procedure that provides a simple and adjustable anomaly detection scheme, as well as a complementary index for assessing the overall Quality of Service (QoS) provided by the operator. The proposed adjustable system can be easily tuned by managers or operators for online anomaly detection, which is an important asset to increase the efficiency of bus transportation services. [ABSTRACT FROM AUTHOR]

This paper aims at the problem of intersection traffic oversaturation and the spillover of queuing vehicles caused by random fluctuations of traffic flow. A self-organizing control model with road capacity constraints and taking minimum expected traffic flow delay as the goal is established to realize real-time adaptive control of traffic signal phase duration. Taking the local intersection and its adjacent intersections as the basic unit of traffic signal control, defining the constraints of road space in queuing state and the equations of traffic signal control basic unit are established. According to the equations, an online method for solving the optimal phase duration of traffic signal is established to efficiently allocate intersection space resources, by which a rule-based real-time traffic signal control model is established. Through the interaction between adjacent intersections, the traffic inflow rate of oversaturated phase is automatically restricted to alleviate the queuing spillover problem. Computer simulation shows that the control effect is significantly better than that of SCATS under the same boundary inflow conditions. Compared with the traditional centralized coordinated control, the rule-based real-time traffic signal control model can better adapt to the fluctuation of traffic flow and effectively relief the problem of vehicle queuing spillover caused by fluctuations of traffic flow. [ABSTRACT FROM AUTHOR]

Along with the battery technology advancements and government policy support, the penetration level of electric buses (EBs) in the urban public transportation system has been increasing in recent years. Considering the potential influence of the increasing EB charging demand on power systems, estimating the real-time energy consumption of EBs has become a principal issue. In this work, a data-driven approach for EB energy consumption estimation is proposed. In particular, a detailed physical model of EB is constructed to model its energy consumption considering the randomness in EB operation, including speed, acceleration, and passenger count. In order to improve the estimation accuracy, the conventional Kalman filter (KF) is modified involving EB mass estimation considering stochastic real-time passenger count, motion data dimension deduction based on EB operation route. To estimate the EB acceleration accurately and reduce the noise caused by the unimportant features, we extended the feature discarding algorithm of decision trees to the regression trees. In the case study, an Android application is developed to collect the EB motion data so that any general Android smartphone can be used for data collection. The performance of the proposed approach is evaluated based on real-world EB operation data collected from St. Albert Transit, AB, Canada. According to the results, our APP can track the real-time EB trace, and the proposed modified KF can filter most of the noises caused by the GPS data collection process and stochastic passenger count. Also, with the extended random forest algorithm, the unimportant features can be discarded and the real-time EB acceleration is estimated efficiently with a small sum of square error (SSE). Compared with the existing approaches, the proposed approach achieves a more accurate real-time energy consumption estimation of EBs, which in turn, provides a better characterization of power system loading and voltage variation. [ABSTRACT FROM AUTHOR]

Detection of railway traffic objects is an important task during train driving and is implemented to ensure safe driving. Although object detection has been investigated for years, many challenges exist in precisely detecting railway objects under complex railway scenes. These challenges mainly include adverse weather states, various railway backgrounds, diverse railway objects, and low-quality images. To address these issues, we introduce a novel deep learning method, called a multi-mode feature enhanced convolutional neural network (MMFE-Net), for accurate railway object detection. The network mainly consists of three modules. 1) An improved cross-stage partial connection darknet53 (CSPDarknet53), called adaptive dilated cspdarknet53, is used as our backbone to reduce image information loss. 2) A spatial feature extraction module is used to improve the feature extraction ability of the model for blurred objects and objects in a complicated background. 3) We introduce an attention fusion enhance module to strengthen the context information between adjacent feature maps to accurately detect multiscale and small objects. The proposed method achieves 0.9439 mAP and 79 FPS with an input size of $640\times 640$ pixels on the railway traffic dataset, and its performance is better than that of YOLOv4. Moreover, it is feasible to apply MMFE-Net into practical applications of railway object detection. [ABSTRACT FROM AUTHOR]

Detecting and preventing intrusions on in-vehicle buses is a topic of great importance which may have an even greater significance in the context of commercial vehicles that are liable for the security of the demanding tasks they carry, passengers or goods not least. In this respect, the SAE J1939 protocol, which is a CAN based higher-layer protocol for commercial vehicles, requires special attention due to the existence of both specific procedures in the standard, e.g., address claims and multi-frame transmissions, as well as due to sharp specifications regarding the content of messages which may facilitate the deployment of a more targeted intrusion detection system. Needless to say, most of the research works on CAN intrusion detection are treating in-vehicle traffic as black-box with no concerns over the actual meaning of the frames content. In this context, we pursue the development of a targeted solution for J1939 buses. We collect real-world traffic from a commercial vehicle bus, compliant to the J1939 standard, and make a comprehensive analysis of its structure and content. This allows us to design an effective intrusion prevention system that detects and eliminates in real-time all frames that were manipulated by an adversary by overriding them with error flags. To prove the correctness of our approach, we present results with a proof-of-concept implementation on high-end automotive-grade controllers. [ABSTRACT FROM AUTHOR]

With the development of intelligent transportation technology, more and more traffic information can be obtained to enhance arterial coordination efficiency. A time-varying arterial coordination control program is proposed in this paper. By extracting Origin-Destination (OD) information from Automatic number plate recognition (ANPR) data, an OD-based Multipath-Band method (MP-BAND) is developed to achieve the function of the program. The proposed method breaks vehicle routing at intersections, and treats one link as an analysis unit. The method can automatically select link paths for coordination, and multiple paths can be synchronized simultaneously according to the actual traffic condition. The overlap bandwidth is introduced to capture the connectivity between paths of adjacent intersections. The PM-BAND is formulated as a mixed-integer linear program, which can be solved by the standard branch-and-bound technique. Numerical tests are conducted to evaluate the performance of MP-BAND under different traffic conditions. The results have demonstrated that MP-BAND outperformed MULTIBAND and AM-BAND in various aspects. MP-BAND can improve network performance significantly almost in all scenarios. Moreover, the intersection efficiency was improved significantly, especially the efficiency of left-turning vehicles. Also, MP-BAND can well recognize major routes and improve traffic efficiency. Compared with Synchro, MP-BAND also had superiority when flow rate is not too high. Thus, MP-BAND has a much wider applicability, and can be treated as an optimization core to achieve time-varying arterial coordination. [ABSTRACT FROM AUTHOR]

Real-time semantic segmentation, which aims to achieve high segmentation accuracy at real-time inference speed, has received substantial attention over the past few years. However, many state-of-the-art real-time semantic segmentation methods tend to sacrifice some spatial details or contextual information for fast inference, thus leading to degradation in segmentation quality. In this paper, we propose a novel Deep Multi-branch Aggregation Network (called DMA-Net) based on the encoder-decoder structure to perform real-time semantic segmentation in street scenes. Specifically, we first adopt ResNet-18 as the encoder to efficiently generate various levels of feature maps from different stages of convolutions. Then, we develop a Multi-branch Aggregation Network (MAN) as the decoder to effectively aggregate different levels of feature maps and capture the multi-scale information. In MAN, a lattice enhanced residual block is designed to enhance feature representations of the network by taking advantage of the lattice structure. Meanwhile, a feature transformation block is introduced to explicitly transform the feature map from the neighboring branch before feature aggregation. Moreover, a global context block is used to exploit the global contextual information. These key components are tightly combined and jointly optimized in a unified network. Extensive experimental results on the challenging Cityscapes and CamVid datasets demonstrate that our proposed DMA-Net respectively obtains 77.0% and 73.6% mean Intersection over Union (mIoU) at the inference speed of 46.7 FPS and 119.8 FPS by only using a single NVIDIA GTX 1080Ti GPU. This shows that DMA-Net provides a good tradeoff between segmentation quality and speed for semantic segmentation in street scenes. [ABSTRACT FROM AUTHOR]

In the Vehicular Ad hoc Networks (VANET) environment, recognizing traffic accident events in the driving videos captured by vehicle-mounted cameras is an essential task. Generally, traffic accidents have a short duration in driving videos, and the backgrounds of driving videos are dynamic and complex. These make traffic accident detection quite challenging. To effectively and efficiently detect accidents from the driving videos, we propose an accident detection approach based on spatio–temporal feature encoding with a multilayer neural network. Specifically, the multilayer neural network is used to encode the temporal features of video for clustering the video frames. From the obtained frame clusters, we detect the border frames as the potential accident frames. Then, we capture and encode the spatial relationships of the objects detected from these potential accident frames to confirm whether these frames are accident frames. The extensive experiments demonstrate that the proposed approach achieves promising detection accuracy and efficiency for traffic accident detection, and meets the real-time detection requirement in the VANET environment. [ABSTRACT FROM AUTHOR]

Aiming at the shortcomings of the current YOLOv3 model, such as large size, slow response speed, and difficulty in deploying to real devices, this paper reconstructs the target detection model YOLOv3, and proposes a new lightweight target detection network YOLOv3-promote: Firstly, the G-Module combined with the Depth-Wise convolution is used to construct the backbone network of the entire model, and the attention mechanism is introduced and added to perform weighting operations on each channel to get more key features and remove redundant features, thereby strengthening the identification ability of feature network model’s to distinguish target objects among background; Secondly, in order to delete some less important channels to achieve the effect of compressing the model size and improving the calculation speed, the size of the scaling factor gamma in the batch normalization layer is used; Finally, based on NVIDIA’s TensorRT framework model conversion and half-precision acceleration were carried out, and the accelerated model was successfully deployed on the embedded platform Jetson Nano. The performed KITTI experimental results show that the inference speed of our proposed method is about 5 times that of the original model, the parameter volume is reduced to one tenth, the mAP is increased from 86.1% of the original model to 93.1%, and the FPS reaches 25.5fps, realizing the requirements of real-time detection with high precision. [ABSTRACT FROM AUTHOR]

With the help of the new generation information technology, the Internet of Vehicle (IoV) networks have become widespread. IoV can improve the automatic driving ability, and provide users with intelligence, comfort, safety, energy saving and efficiency traffic services. However, the IoV networks face serious challenges due to the complex wireless environment. The vehicles cannot obtain the real-time traffic condition and early warning information, which leads to the decrease of link quality and the failure of information transmission. To evaluate the communication quality of IoV networks, the outage probability (OP) is commonly employed as a metric. This paper considers mobile IoV networks, and investigates communication quality prediction. Novel OP expressions are derived, which can analyze the OP performance. Then, to predict OP in real time, an intelligent OP prediction approach with an Elman model is proposed. This is evaluated with data generated using the OP expressions. In terms of computational complexity and prediction accuracy, the results obtained show that the Elman-based approach provides better forecasting effect than other methods. For prediction accuracy, the proposed Elman approach is increased by 84.6%. For computational complexity, the execution time is reduced by 79.9%. [ABSTRACT FROM AUTHOR]

A new trend of using deep reinforcement learning for traffic signal control has become a spotlight in the Intelligent Transportation System (ITS). However, the traditional intelligent traffic signal control system always collects and transmits vehicle information (e.g., vehicle location, speed, etc.) in the form of plaintext, which would result in the leakage of commuters’ privacy and thus bring unnecessary troubles to users. In this paper, we propose a privacy-preserving traffic signal control for an intelligent transportation system (PrivacySignal). It relies on the existing road facilities to achieve the privacy of commuters, which guarantees the practicality of the system. Real-time decision-making and confidentiality of the system can be achieved simultaneously via the design of a series of secure and efficient interactive protocols, that are based on additive secret sharing, to perform the deep $Q$ -network (DQN). Moreover, the security of PrivacySignal is testified, meanwhile, the system effectiveness, and the overall efficiency of PrivacySignal is demonstrated through theoretical analysis and simulation experiments. Compared with the existing privacy-preserving schemes of the intelligent traffic signal, PrivacySignal provides a general DQN based privacy-preserving traffic signal control strategy architecture with high efficiency and low-performance loss. [ABSTRACT FROM AUTHOR]

In recent years, pedestrian and vehicle detection at night has become an important subject of computer vision applications. Because the environment light is weak at night, the common traditional camera-input algorithms often performs poorly. Pedestrians and vehicles in infrared images are usually brighter than the surrounding environment and have salient features. In this paper, a fast saliency map of pedestrian and vehicle targets in infrared images at night is used to achieve rapid acquisition of areas of interest for pedestrians and vehicles. Based on this, we propose a method to refine and separate the target area to obtain accurate pedestrian and vehicle candidate bounding boxes. Specifically, this paper proposed a multi-feature fusion algorithm of pedestrian and vehicle feature extraction combined with support vector machine (SVM) to determine whether the extracted target area really includes pedestrians and vehicles. Our experimental results show that the proposed method can achieve the expected effect of the classification and detection for pedestrians and vehicles at night, and can meet the real-time requirements of actual road scenarios. [ABSTRACT FROM AUTHOR]

Positioning and navigation are critical functions of automated driving functions, which help autonomous vehicles determine their absolute and relative positions in the environment that they operate. Integrity Monitoring (IM) systems, which are intended to assess the reliability and trustworthiness of the information provided by the navigation systems, are crucial for ensuring the safety of automated driving functions. This paper provides a comprehensive review of the existing IM frameworks for safety-critical navigation applications and expands on the state-of-the-art of the most recent development of such systems for connected automated vehicles. We mainly focus on IM methods for Global Navigation Satellite Systems (GNSS) and Inertial Navigation Systems (INS). However, we also cover IM for map assisted and wireless signal augmented navigation systems, which are promising for high-performance navigation applications, such as automated driving functions. For each main category of solutions, key aspects such as the characteristics of measurement errors and faults related to various data sources are discussed to provide deeper insights into designing of reliable IM systems. Also, some of the major open research challenges to the best knowledge of the authors have been identified and discussed. [ABSTRACT FROM AUTHOR]

Providing transport users and operators with accurate forecasts on travel times is challenging due to a highly stochastic traffic environment. Public transport users are particularly sensitive to unexpected waiting times, which negatively affect their perception on the system’s reliability. In this paper we develop a robust model for real-time bus travel time prediction that departs from Gaussian assumptions by using Student-t errors. The proposed approach uses spatiotemporal characteristics from the route and previous bus trips to model short-term effects, and date/time variables and Gaussian processes for long-run forecasts. The model allows for flexible modeling of mean, variance and kurtosis spaces. We propose algorithms for Bayesian inference and for computing probabilistic forecast distributions. Experiments are performed using data from high-frequency buses in Stockholm, Sweden. Results show that Student-t models outperform Gaussian ones in terms of log-posterior predictive power to forecast bus delays at specific stops, which reveals the importance of accounting for predictive uncertainty in model selection. Estimated Student-t regressions capture typical temporal variability between within-day hours and different weekdays. Strong spatiotemporal effects are detected for incoming buses from immediately previous stops, which is in line with many recently developed models. We finally show how Bayesian inference naturally allows for predictive uncertainty quantification, e.g. by returning the predictive probability that the delay of an incoming bus exceeds a given threshold. [ABSTRACT FROM AUTHOR]

Semantic segmentation and depth estimation lie at the heart of scene understanding and play crucial roles especially for autonomous driving. In particular, it is desirable for an intelligent self-driving agent to discern unexpected obstacles on the road ahead reliably in real-time. While existing semantic segmentation studies for small road hazard detection have incorporated fusion of multiple modalities, they require additional sensor inputs and are often limited by a heavyweight network for real-time processing. In this light, we propose an end-to-end Real-time Obstacle Detection via Simultaneous refinement, coined RODSNet (https://github.com/SAMMiCA/ RODSNet) which jointly learns semantic segmentation and disparity maps from a stereo RGB pair and refines them simultaneously in a single module. RODSNet exploits two efficient single-task network architectures and a simple refinement module in a multi-task learning scheme to recognize unexpected small obstacles on the road. We validate our method by fusing Cityscapes and Lost and Found datasets and show that our method outperforms previous approaches on the obstacle detection task, even recognizing the unannotated obstacles at 14.5 FPS on our fused dataset ($2048\times 1024$ resolution) using RODSNet- $2\times $ . In addition, extensive ablation studies demonstrate that our simultaneous refinement effectively facilitates contextual learning between semantic and depth information. [ABSTRACT FROM AUTHOR]

Comprehensive and accurate perception of the real 3D world is the basis of autonomous driving. However, many perceptual methods focus on a single task or object type, and the accuracy of existing multi-task or multi-object methods is difficult to balance against their real-time performance. This paper presents a unified framework for concurrent dynamic multi-object joint perception, which introduces a real-time monocular joint perception network termed MJPNet. In MJPNet relative weightings are automatically learned by a series of developed network branches. By training an end-to-end deep convolutional neural network on a shared feature encoder and many proposed decoding sub-branches, the information of the 2D category and 3D position/pose/size of an object are reconstructed both simultaneously and accurately. Moreover, the effective information among subtasks is transferred by multi-stream learning, guaranteeing the accuracy of each task. Compared to various state-of-the-arts, comprehensive evaluations on the benchmark of challenging image sequences demonstrate the superior performance of our 2D detection and 3D reconstruction of depth, lateral distance, orientation, and heading angle. Moreover, on the KITTI test set, the real-time runtime (up to 15 fps) of MJPNet significantly outran the public state-of-the-art visual detection methods. Accompanying video: https://youtu.be/Z-goToOlI94. [ABSTRACT FROM AUTHOR]

This research aims to model system dynamics for mixed traffic flow consisting of Connected and Automated Vehicles (CAVs) and Human-driven Vehicles (HVs). It quantifies the impact of CAVs’ speed change on the overall traffic state on a real-time basis. The model describes the impedance of CAVs’ speed reduction on traffic flow and considers the impact of potential additional lane change induced by the speed reduction. To validate the effectiveness of the proposed model, a VISSIM based microscopic simulation evaluation is performed. The results confirm that the accuracy of the proposed model is generally over 80% with the CAVs’ speed reduction constrained within 20 km/h. Sensitivity analysis is conducted in terms of various CAV penetration rates and congestion levels. The proposed model demonstrates consistently good performance across all CAV penetration rates and congestion levels. A showcase is presented to show the effect of the system dynamics in active traffic management. The proposed model could serve as the foundation of CAV based traffic management applications, such as variable speed limit and speed harmonization. [ABSTRACT FROM AUTHOR]

As one of the fundamental visual tasks in the unmanned driving area, lane detection attracts increasing attention. In practical applications, lane points are very difficult to localize because they usually appear to be sparse and incomplete due to the influence of illumination and environment. Conventional lane detection methods rely on coarse features and carefully designed postprocessing to detect the lane lines. However, these methods are usually slow, and the stability and generalization ability are unsatisfactory. In this paper, we propose a novel lane detection network – Fast-HBNet(Fast-Hybrid Branch Network), which exploits both global semantic information and spatial contexts. To enlarge receptive fields and encode more detailed information, the compound transformation is employed and the proposed hybrid branch network extracts four diverse feature maps with different receptive fields and spatial contexts. Besides, we design a Hierarchical Feature Learning (HFL) module to learn lane features from the scale, channel, and spatial levels to enhance the generalization ability of our detector. These features are further selectively coalesced to generate unified lane feature maps with large receptive fields and rich detailed information. In other words, our network can encode the global semantic information from the high-resolution feature maps and the fine-grained details in the low-resolution feature maps. Experimental results conducted on the TuSimple (2017) and CULane [Pan et al. (2018)] datasets demonstrate that the proposed Fast-HBNet outperforms numerous state-of-the-art lane detectors in both speed and accuracy. Particularly, Fast-HBNet achieves an accuracy of 96.88% on the TuSimple dataset at a speed of 76 FPS. [ABSTRACT FROM AUTHOR]

Predictions of transit delays are crucial to passengers and operators. Passengers utilize the predictions to decide on departure time, route choice, and mode choice, whereas operators decide on schedules, timetables, rolling stock allocation, and control actions. We introduce the concept of predictability of transit travel times as the study of the reduction of the predicted variability with the temporal approaching of a predicted event. We evaluate predictability on a real-life test case in Zurich, Switzerland, spanning multiple transit lines over one year of operations. The concept is shown based on predictions obtained by a state-of-the-art Bayesian network approach, where we show how predictability (in general) can be modeled as an exponential decay phenomenon. The study of predictability of transit operations leads to additional insights for control actions and system analysis compared to other complementary concepts such as punctuality or regularity, for instance, concerning bunching, identification of bottlenecks, and passenger routing. [ABSTRACT FROM AUTHOR]

We consider a signalized intersection under a partially connected automated vehicle (CAV) environment. There is a control center for the control zone that needs to predict the trajectories of human driving vehicles (HDVs) and control the trajectories of CAVs. By adopting model predictive control, we propose a real-time eco-driving strategy for the control center. In the proposed strategy, the Gipps’ car-following model is selected to update the acceleration of HDV and an optimal control problem (OCP) is proposed to optimize the trajectory of each CAV based on real-time travel information. The control objective is to minimize the total fuel consumption of each CAV during the current control period. Pontryagin’s minimum principle is employed to derive necessary optimality conditions under different scenarios. With the necessary optimality conditions, a numerical method is developed to solve the proposed OCP. Finally, numerical examples are provided to illustrate the performance of the proposed eco-driving strategy. [ABSTRACT FROM AUTHOR]

Proactive traffic safety management systems can reduce crashes by identifying crash precursors, evaluating real-time crash risks, and implementing suitable interventions. The basic prerequisite for developing such a system is to propose a reliable crash risk evaluation model that takes real-time traffic flow data as input. Previous studies have primarily focused on real-time crash prediction using some statistical or machine-learning methods. However, further quantitative evaluation and classification of crash risks have been ignored. In this study, we conduct a systematic crash risk evaluation workflow, including crash risk prediction, crash risk quantification, and crash risk classification. Specifically, the crash risk prediction using an extended logit model is proposed, from which CAS, CSD, UAS, DAS, DTV are identified to be contributing factors of crash risks. Then a crash risk quantification model based on the parameter evaluation of the extended logit model is developed. The crash risks of urban expressways and their spatial-temporal evolution trends are quantified. Finally, the crash risks are classified into high crash risk level, moderate crash risk level, and low crash risk level by the k-means cluster algorithm. Then the threshold boundaries of different crash risk levels are determined. The research results provide a proactive guidance for traffic safety management of urban expressways. [ABSTRACT FROM AUTHOR]

When a train leaves a platform, knowing the carriage load (the number of passengers in each carriage) of this train will support train managers to guide passengers at the next platform to choose carriages to avoid congestion. This capacity has become critical since the onset of the pandemic. However, with the dynamicity of passengers and the speed of trains improved (about 3 minutes travel between stations) as well as the station stop period reduced (60–90 second per station), the real-time prediction is more challenging. This paper presents an intelligent system, which is developed in collaboration with Sydney Trains, for real-time predicting carriage load across a city passenger train network. The system comprises three innovations. First, a fuzzy time-matching method significantly improves prediction accuracy in the uncertain situations and allows noisy historical data to be used for training. Second, the LightGBM model is extended with an incremental learning scheme to make forecasting in real-time possible. Third, a new multi-stream learning strategy that merges data streams with similar concept drift patterns is pioneered to increase the amount of suitable training data while reducing generalization errors. A comprehensive suite of practical tests on real-world datasets demonstrates the merit of these solutions. [ABSTRACT FROM AUTHOR]