Field Of Intelligent Transportation Research Articles

Marine Vision-Based Situational Awareness Using Discriminative Deep Learning: A Survey

The primary task of marine surveillance is to construct a perfect marine situational awareness (MSA) system that serves to safeguard national maritime rights and interests and to maintain blue homeland security. Progress in maritime wireless communication, developments in artificial intelligence, and automation of marine turbines together imply that intelligent shipping is inevitable in future global shipping. Computer vision-based situational awareness provides visual semantic information to human beings that approximates eyesight, which makes it likely to be widely used in the field of intelligent marine transportation. We describe how we combined the visual perception tasks required for marine surveillance with those required for intelligent ship navigation to form a marine computer vision-based situational awareness complex and investigated the key technologies they have in common. Deep learning was a prerequisite activity. We summarize the progress made in four aspects of current research: full scene parsing of an image, target vessel re-identification, target vessel tracking, and multimodal data fusion with data from visual sensors. The paper gives a summary of research to date to provide background for this work and presents brief analyses of existing problems, outlines some state-of-the-art approaches, reviews available mainstream datasets, and indicates the likely direction of future research and development. As far as we know, this paper is the first review of research into the use of deep learning in situational awareness of the ocean surface. It provides a firm foundation for further investigation by researchers in related fields.

A Novel Hierarchical Model in Ensemble Environment for Road Detection Application

As a popular research direction in the field of intelligent transportation, road detection has been extensively concerned by many researchers. However, there are still some key issues in specific applications that need to be further improved, such as the feature processing of road images, the optimal choice of information extraction and detection methods, and the inevitable limitations of detection schemes. In the existing research work, most of the image segmentation algorithms applied to road detection are sensitive to noise data and are prone to generate redundant information or over-segmentation, which makes the computation of segmentation process more complicated. In addition, the algorithm needs to overcome objective factors such as different road conditions and natural environments to ensure certain execution efficiency and segmentation accuracy. In order to improve these issues, we integrate the idea of shallow machine-learning model that clusters first and then classifies in this paper, and a hierarchical multifeature road image segmentation integration framework is proposed. The proposed model has been tested and evaluated on two sets of road datasets based on real scenes and compared with common detection methods, and its effectiveness and accuracy have been verified. Moreover, it demonstrates that the method opens up a new way to enhance the learning and detection capabilities of the model. Most importantly, it has certain potential for application in various practical fields such as intelligent transportation or assisted driving.

Short-Term Traffic Flow Prediction for Urban Road Sections Based on Time Series Analysis and LSTM_BILSTM Method

The real-time performance and accuracy of traffic flow prediction directly affect the efficiency of traffic flow guidance systems, and traffic flow prediction is a hotspot in the field of intelligent transportation. To further improve the accuracy of short-term traffic flow prediction, a short-term traffic flow prediction model based on traffic flow time series analysis, and an improved long short-term memory network (LSTM) is proposed. First, perform time series analysis on traffic flow data and perform smoothing and standardization processing to obtain a stable time series as model input data, which can improve the accuracy of model training and eliminate the impact of a wide range of feature values. Then, an improved LSTM model based on LSTM and bidirectional LSTM networks are established. Combining the advantages of sequential data and the long-term dependence of forwarding LSTM and reverse LSTM, the bidirectional long-term memory network (BILSTM) is integrated into the prediction model. The first layer of the LSTM network learns and predicts the input time series and further learns and trains through the bidirectional LSTM network to effectively overcome the large prediction errors. Finally, the performance of the proposed method is evaluated by comparing the predicted results with actual traffic data. The model that is proposed in this paper is compared with the long short-term memory network (LSTM) model and the bidirectional long-term memory network (BILSTM) model. The results demonstrate that the proposed method outperforms both compared methods in terms of accuracy and stability.

The Architecture and Development of Intelligent Traffic Signal Control Platform

The project used intelligent video analysis, traffic big data information processing and analysis technology to accurately grasp the traffic demand of urban traffic. The platform developed by this project could be combined with the vehicle level to adjust and optimize the traffic signal timing scheme in real time, which could improve road utilization of cities. This platform could use the big data analysis technology to predict future traffic situation based on historical traffic conditions and develop a best timing plan in advance. Under the existing road network conditions, the basis of this project was to rationally organize and plan traffic flow, improve road traffic management facilities, and strengthen and enhance the modern traffic awareness of traffic participants. The goal of the platform was to provide order, smooth, safe and traffic management services, it also aims at rapid response, and decision-making in the field of intelligent transportation.

Vehicle Dispatching Technology Based on Geographic Grid Division

With the increasing popularity of online ride platforms, more and more historical data can be used in the field of intelligent transportation. However, the taxi behavior of users is difficult to predict and has a great relationship with time and geographical location, which makes the supply and demand of vehicles show an imbalance. Reasonable vehicle scheduling can not only increase driver’s income, improve service quality, but also make effective use of traffic resources. In this paper, a grid-based scheduling model is proposed. By gridding maps, the computational complexity of Donger is reduced to output vehicle scheduling actions quickly. The problem of vehicle supply and demand imbalance can be effectively solved by correcting model parameters based on return loss function. In this paper, the model is optimized from three aspects: loss function, learning rate and grid scale. The goal is to maximize the sum of orders. Compared with the traditional machine learning model, the model proposed in this paper can have deeper features and thus obtain higher prediction accuracy. The experimental results on a large number of real historical taxi order datasets show that the proposed model has high prediction accuracy.

Free‐Fixed Rotational Triboelectric Nanogenerator for Self‐Powered Real‐Time Wheel Monitoring

AbstractDeveloping an applicable triboelectric nanogenerator (TENG) for train wheel energy harvesting is a key step to meet the urgent need of wheel safety monitoring. Herein, an innovative design of free‐fixed TENG (FF‐TENG) is reported, without a serious negative impact on the wheel. The key of this design is the magnets fixed on the device and bogie, providing attractive force to immobilize the stator. With a rotational structure, FF‐TENG can provide a high short‐circuit current of 55 µA, an open‐circuit voltage of 500 V, and a charge of 235 nC at a rotation speed of 400 rpm. At an external load resistance of 10 MΩ, FF‐TENG delivers the maximum power of 15.68 mW. Furthermore, the superior robustness of FF‐TENG in vibration environment is proved. In addition, a power management circuit designed by LTC 3588 is tested for more efficient capacitor charging, leading to better performance to power electronics. Finally, a self‐powered real‐time wheel temperature and wheel speed monitoring system is developed with FF‐TENG as a safety alert demo for feasibility demonstration. Given the rational structure design and high performance, this work paves a practical way for TENGs in the field of intelligent transportation.

Research on Eco-driving of Signalized Intersection Fleet in Vehicle-Road Cooperative Environment

In recent years, traditional motor vehicles have suffered increasingly severe constraints on energy resources. Coupled with the increasingly serious hazards of pollutants emitted by transportation, an important strategic goal for the development of China’s transportation industry has been set for energy conservation and emission reduction. Vehicle-road collaboration technology is a cutting-edge technology in the field of intelligent transportation, which can guide drivers in ecological driving strictly in accordance with the actual traffic environment in which the vehicle is located. Therefore, researchers are paying more and more attention to how to reduce vehicle fuel consumption and emissions. Signalized intersection as an important node in urban roads needs to be studied. However, the focus of its research so far is on bicycles, and most researchers are indifferent to the situation in which vehicles are likely to run together within the intersection. In addition, when establishing an eco-driving model, most researchers will still focus on how to achieve energy conservation and emission reduction only from the perspective of reducing the number of parking times, and do not care about the essential attributes of fuel consumption. As a result, it is difficult for us to better control and quantify the overall implementation effect of the model. Based on this, it is very important to actively establish a corresponding fleet ecological driving model based on the fleet within the signalized intersection in the vehicle-road collaborative environment.

AST-GCN: Attribute-Augmented Spatiotemporal Graph Convolutional Network for Traffic Forecasting

Traffic forecasting is a fundamental and challenging task in the field of intelligent transportation. Accurate forecasting not only depends on the historical traffic flow information but also needs to consider the influence of a variety of external factors, such as weather conditions and surrounding POI distribution. Recently, spatiotemporal models integrating graph convolutional networks and recurrent neural networks have become traffic forecasting research hotspots and have made significant progress. However, few works integrate external factors. Therefore, based on the assumption that introducing external factors can enhance the spatiotemporal accuracy in predicting traffic and improving interpretability, we propose an attribute-augmented spatiotemporal graph convolutional network (AST-GCN). We model the external factors as dynamic attributes and static attributes and design an attribute-augmented unit to encode and integrate those factors into the spatiotemporal graph convolution model. Experiments on real datasets show the effectiveness of considering external information on traffic speed forecasting tasks when compared with traditional traffic prediction methods. Moreover, under different attribute-augmented schemes and prediction horizon settings, the forecasting accuracy of the AST-GCN is higher than that of the baselines. The source code of the AST-GCN is available at https://github.com/lehaifeng/T-GCN/AST-GCN .

Improving V2X-6G network capacity using a new UAV-based approach in a Cloud/ICN architecture, case Study: VANET network

6G of mobile networks plays a crucial role in improving the capacity and enhancing the quality of services of Vehicle-to-Everything (V2X) based networks evolving in an intelligent environment. VANET is a promising project in the intelligent transportation field using V2X communications. The emergence of several 5G and 6G technologies has raised several challenges for scientists and researchers to allow vehicles and road users to enjoy several services while ensuring their safety on the road. Among these technologies, the unmanned aerial vehicle (UAV), which can perform different tasks for road users and vehicle drivers such as data caching, packet relaying and processing. In this article, we present a new approach based on 6G Unmanned Aerial Vehicles (UAV) technology on a vehicular cloud architecture while exploiting the exchange support of information-centric networking (ICN) for the improvement of network capacity.

Research on path selection system based on green transportation

The transportation sector already accounts for 14% of global greenhouse gas emissions. Therefore, controlling carbon emissions in the transportation sector has become a top priority for China and other countries around the world. In addition to the technological development of clean energy transportation, the most critical aspect of the globalization of new energy transportation industry and market is the optimization of access routes. This paper will be based on road travel time estimation and optimal path selection of energy efficient transportation research hotspots. Firstly, the accuracy of road travel time estimation is improved according to the basic law of traffic flow. At the same time, this paper defines the optimal route into two cases of shortest path and shortest time for solving. Finally, the actual solution process is given according to the actual problem, as well as the optimal route, aiming to promote the strong development in the field of intelligent navigation system and energy-efficient transportation.

Research on Target Tracking Algorithm in occlusion scene

Target tracking is widely used in video image, intelligent transportation, behavior analysis and other fields. At the same time, target tracking technology faces more and more challenges. In this paper, the target tracking technology in occlusion scene is studied, and the target tracking method based on spatiotemporal context and the target tracking method combined with depth network model are introduced respectively. The main ideas and principles of the two methods are analyzed, and the improvements based on these two methods are summarized. Finally, the problems existing in the future development of occluded target tracking algorithm are prospected.

EAWNet: An Edge Attention‐Wise Objector for Real‐Time Visual Internet of Things

With the upgrading of the high‐performance image processing platform and visual internet of things sensors, VIOT is widely used in intelligent transportation, autopilot, military reconnaissance, public safety, and other fields. However, the outdoor visual internet of things system is very sensitive to the weather and unbalanced scale of latent object. The performance of supervised learning is often limited by the disturbance of abnormal data. It is difficult to collect all classes from limited historical instances. Therefore, in terms of the anomaly detection images, fast and accurate artificial intelligence‐based object detection technology has become a research hot spot in the field of intelligent vision internet of things. To this end, we propose an efficient and accurate deep learning framework for real‐time and dense object detection in VIOT named the Edge Attention‐wise Convolutional Neural Network (EAWNet) with three main features. First, it can identify remote aerial and daily scenery objects fast and accurately in terms of an unbalanced category. Second, edge prior and rotated anchor are adopted to enhance the efficiency of detection in edge computing internet. Third, our EAWNet network uses an edge sensing object structure, makes full use of an attention mechanism to dynamically screen different kinds of objects, and performs target recognition on multiple scales. The edge recovery effect and target detection performance for long‐distance aerial objects were significantly improved. We explore the efficiency of various architectures and fine tune the training process using various backbone and data enhancement strategies to increase the variety of the training data and overcome the size limitation of input images. Extensive experiments and comprehensive evaluation on COCO and large‐scale DOTA datasets proved the effectiveness of this framework that achieved the most advanced performance in real‐time VIOT object detection.

A Graph-Based Track-Before-Detect Algorithm for Automotive Radar Target Detection

Among various types of sensors, millimeter-wave radar is widely used, especially in the field of intelligent transportation, but often limited by low signal-to-noise ratio at the same time. As a result, sensor fusion is currently the most common way for target detection. In contrast to this, we focus on the problem of single radar target detection. Considering that automotive radar usually contains a plot generation module (which provides a limited number of candidate targets), the track-before-detect method can meet real-time requirements in ADAS. Successive track cancellation is one of the effective strategies for enhancing the performance of track-before-detect algorithms. We are inspired by this and derive a graph-based radar target detection algorithm. Since the data to be processed between frames have structural similarity, we can measure the correlation between plots better without a large amount of calculation by taking advantage of the dynamically maintained graph. Furthermore, we define a simple state transition process of each plot, which enhances robustness for temporary target signal loss. We verify the effectiveness of the proposed algorithm with both the simulation experiment and real data test in the last part of the paper.

Analysis of driver’s choice behavior based on the combined model of utility-regret

In the context of increasing urban traffic trajectory data, based on large-scale urban traffic trajectory data, in the field of intelligent transportation, it has become an urgent need to analyze travelers’ routing behaviors and establish effective routing models to provide travelers with efficient and reasonable driving route recommendations. It is often the case that urban taxi groups has better route choice behavior than other travel group under the cost-benefit constraints. Therefore, providing route recommendations for other travelers according to the route choice habits of urban taxi group is a feasible solution to the above need. This paper starts with the utility maximization model and the stochastic regret minimization model, for the problem that the traveler cannot satisfy the premise of the complete rationality of the utility maximization model when choosing routes and the random regret minimization model ignores the traveler’s sensitivity to the original value of the alternative attribute, and improve the calculation formula of the regret value in the random regret minimization model. A route choice model under the utility-regret combination rule is proposed based on the combination of the utility model and the improved regret model, and then, based on the historical trajectory data of taxis in Xi’an, the RP data required by the model is extracted, and the parameters of the model are estimated and verified. The experimental results show that the route choice probability calculated by the model is closer to the route choice probability in the real scene than the simple utility maximization model or the simple random regret minimization model.

Improved Multi-domain Convolutional Neural Networks Method for Vehicle Tracking

In the field of intelligent transportation, background complexity, lighting changes, occlusion, and scale transformation affect the tracking results of moving vehicles in the video. We propose an improved vehicle object tracking algorithm based on Multi-Domain Convolutional Neural Networks (MDNet), combining the instance segmentation method with the MDNet algorithm, adding two attention mechanisms to the algorithm. The module extracts better features, ensures that the vehicle object adapts to changes in appearance, and greatly improves tracking performance. Our improved algorithm has a tracking precision rate of 91.8% and a success rate of 67.8%. The Vehicle Tracking algorithm is evaluated on the Object Tracking Benchmark (OTB) data set. The tracking results are compared with eight mainstream object tracking algorithms, and the results show that our improved algorithm has excellent performance. The object tracking precision rate and tracking success rate of this algorithm have achieved excellent results in many cases.

Internet of Vehicle Empowered Mobile Media Scenarios: In-Vehicle Infotainment Solutions for the Mobility as a Service (MaaS)

Recently, the commercial potentials and practical values of Mobility as a Service (MaaS) have been witnessed for sustainable development of smart cities in the field of intelligent transportation. Driven by the Internet of Vehicles (IoV), vehicles gradually become mobile living spaces to satisfy diverse demands of occupants, where the media functionalities, scenario scopes, and user engagements can be all expended dramatically. In the era of 5G/B5G, the intelligent connected vehicles are considered as super media because of the improved In-Vehicle Infotainment (IVI). In this paper, relying on the advanced communication and intelligent techniques, we investigate the IoV-based vehicle-mounted media scenarios and construct IVI solutions for the MaaS. Firstly, we propose the characteristics of vehicle-mounted media scenarios and present the IVI evolution from the aspects of hearing, vision and multi-sensory media services. Secondly, we point out the main features and limitations of IoV-based IVI in 3G/4G. Finally, exploring the potential of super media, both the linkage display and IVI collaborative modes in vehicle-centric media scenarios are introduced as IVI solutions for the MaaS in 5G-V2X and beyond, which intend to further improve the audio-visual experiences of occupants and observers in urban environments.

CANsec: A Practical in-Vehicle Controller Area Network Security Evaluation Tool.

The Internet of Things (IoT) is an industry-recognized next intelligent life solution that increases the level of comfort, efficiency, and automation for citizens through numerous sensors, smart devices, and cloud stations connected physically. As an important application scenario of IoT, the Internet of Vehicles (IoV) plays an extremely critical role in the intelligent transportation field. In fact, the In-Vehicle Network of smart vehicles that are recognized as the core roles in intelligent transportation is currently the Controller Area Network (CAN). However, the In-Vehicle CAN bus protocol has several vulnerabilities without any encryption, authentication, or integrity checking, which severely threatens the safety of drivers and passengers. Once malicious attackers hack the vehicular gateway and obtain the access right of the CAN, they may control the vehicle based on the vulnerabilities of the CAN bus protocol. Given the severe security risk of CAN, we proposed the CANsec, a practical In-Vehicle CAN security evaluation tool that simulates malicious attacks according to major attack models to evaluate the security risk of the In-Vehicle CAN. We also show a usage case of the CANsec without knowing any information from the vehicle manufacturer.

Vehicle Counting System using Deep Learning and Multi-Object Tracking Methods

Using deep learning technology and multi-object tracking method to count vehicles accurately in different traffic conditions is a hot research topic in the field of intelligent transportation. In this paper, first, a vehicle dataset from the perspective of highway surveillance cameras is constructed, and the vehicle detection model is obtained by training using the You Only Look Once (YOLO) version 3 network. Second, an improved multi-scale and multi-feature tracking algorithm based on a kernel correlation filter (KCF) algorithm is proposed to avoid the KCF extracting single features and single-scale defects. Combining the intersection over union (IoU) similarity measure and the row-column optimal association criterion proposed in this paper, matching strategy is used to process the vehicles that are not detected and wrongly detected, thereby obtaining complete vehicle trajectories. Finally, according to the trajectory of the vehicle, the traveling direction of the vehicle is automatically determined, and the setting position of the detecting line is automatically updated to obtain the vehicle count result accurately. Experiments were conducted in a variety of traffic scenes and compared with published data. The experimental results show that the proposed method achieves high accuracy of vehicle detection while maintaining accuracy and precision in tracking multiple objects, and obtains accurate vehicle counting results which can meet real-time processing requirements. The algorithm presented in this paper has practical application for vehicle counting in complex highway scenes.

EEG-Based Driver Drowsiness Estimation Using an Online Multi-View and Transfer TSK Fuzzy System

In the field of intelligent transportation, transfer learning (TL) is often used to recognize EEG-based drowsy driving for a new subject with few subject-specific calibration data. However, most of existing TL-based models are offline, non-transparent, and in which features are only represented from one view (usually only one algorithm is used to extract features). In this paper, we consider an online multi-view regression model with high interpretability. By taking the 1-order TSK fuzzy system as the basic regression component and injecting the nature of the multi-view settings into the existing transfer learning framework and enforcing the consistencies across different views, we propose an online multi-view & transfer TSK fuzzy system for driver drowsiness estimation. In this novel model, features in both the source domain and the target domain are represented from multi-view perspectives such that more pattern information can be utilized during model training. Also, comparing with offline training, the proposed online fuzzy system meets the practical requirements more competently. An experiment on a driving dataset demonstrates that the proposed fuzzy system has smaller drowsiness estimation errors and higher interpretability than introduced benchmarking models.

Preface

2019 International Conference on Intelligent Transportation and Vehicle Engineering (ICITVE 2019), hosted by School of Automotive and Transportation Engineering, Heilongjiang Institute of Technology, was successfully held in Chongqing, China during December 6-8, 2019. ICITVE 2019 aims to provide a platform for experts, scholars, and practitioners in the field of Intelligent Transportation and Vehicle Engineering to share the latest research results, discuss existing problems and challenges, explore cutting-edge technologies, and enhance their cooperation.We are honored to have Prof. Baixue Fu, Dean of School of Automotive and Transportation Engineering, Heilongjiang Institute of Technology, China, and Prof. Traian Mazilu from University Politehnica of Bucharest, Romania to jointly chair this conference. And we invited over 30 excellent scholars form east and west to constitute the Academic Committee of ICITVE 2019. During the conference, five remarkable experts gave excellent keynote speeches related to the field of Intelligent Transportation and Vehicle Engineering on site. Prof. Devinder Yadav from University of Nottingham Ningbo, China discussed emission control at airports and advances in alternate propulsion in civil aviation industry. Assoc.Prof. Seyed Mohammadreza Ghadiri from Malaysia University of Science & Technology, shared his research on Speed Limit Obedience Factor (SLOF) A New Countermeasure to Improve the Efficiency of The Advisory Intelligent Speed Adaptation. Prof. Said Easa from Ryerson University Toronto, Canada held a speech on Smart Mobility: Challenges and Solutions. Prof. Fangwu Ma from Jilin University, China discussed the Outlook of Ecosystem of New Mobility. And Senior Engineer Chengyong Niu, Project Supervisor from Chongqing Vehicle Test & Research Institute Co., Ltd, China had a speech on Test and application challenges for autonomous vehicle at level 3 and below. Thanks to these wonderful speeches given by the distinguished experts, the conference has formed a lively and heated academic vibe on site.We are happy to present the proceedings of ICITVE 2019 featured the new advances and research results in the fields of transportation and vehicle engineering research. All papers have been through rigid review process to comply with the requirements of international publication.We would to like to express our sincere gratitude to all the participants attending the conference, who made this conference a great success. We want to thank our honorific chairmen for hosting this conference, the distinguished keynote speakers for giving insightful speeches, and all authors for submitting their study paper. We hope this conference and the proceedings of it could serve as a good reference for those who work in the field of Intelligent Transportation and Vehicle Engineering. We aim to hold this conference every year to make it an convenient platform for people to share views and experiences in intelligent transportation and vehicle engineering and related areas. We would like to see you next year. Thank you.Committee of ICITVE 2019List of Committee members are available in this pdf.