Search Results

In order to ensure the safety of tunnel traffic and promote the digital construction of tunnel lighting facility maintenance, this paper studies a target detection and fault identification method based on intelligent video surveillance. This method uses the luminance characteristics of the illuminator in the image to preprocess the image. Then, the video image features of the lamp are extracted as well as the improved HOG feature and SVM classifier are used for training and recognition to realize the target detection and location of the lamp. Subsequently, the current target equipment is identified, and finally, combined with the equipment input signal, a fault identification model of the lighting lamp that can distinguish the type of equipment fault is established, so as to realize the real-time fault alarm. According to the results of the experiment, Compared with the traditional HOG feature, the improved HOG feature has higher recognition accuracy.

With the construction of large-scale transport infrastructure, the operations management of transport infrastructure has become the next development priority. In the era of big data, data communication transmission technology and data integration applications have been developed and transformed, and are effectively applied to transportation management. To explore the digital management of transport infrastructure, information technology and transport infrastructure management are combined. The status quo and process of digital management of transport infrastructure are analyzed. Three systems of traffic infrastructure digital management are explored: network-level management, full life cycle management and multi-source information fusion management. The new challenges brought by the digital management of transport infrastructure in the era of big data are analyzed, and some explorations for the digital road construction in China are made.

Urban intersection is the key element to determine the traffic operation of road network. Under the CAVs environment, the roadside control equipment of intersection can communicate with CAVs in real time, collect vehicle state data and optimize traffic control schemes. This paper presents a method for intersection traffic signal control based on deep learning of CAVs data. In addition, intelligent control agent of traffic signal (ICATS) is designed to simulate CAVs. ICATS can perceive real-time changes of traffic flow, model different conditions of intersection and generate the corresponding traffic signal scheme. ICATS used double Q-learning method combination with deep neural network, which is an effective model-independent deep learning algorithm. Moreover, the real traffic data is collected and tested in this paper for evaluating the experiment performance, including vehicle delay, number of passing vehicles, total stop times and passing time.

In recent years, with the advancement of science and technology, research on unmanned vehicle-related technologies has become a major focus of research at home and abroad, and related research results have gradually been applied to various industries. In order to achieve the purpose of autonomous navigation for unmanned vehicles, synchronous positioning and map construction and path planning have become problems to be solved urgently. This article is oriented to indoor services, based on the open-source robot operating system (ROS) as the development platform, which combines mapping and path planning algorithms to realizes autonomous navigation. Finally, a simulation experiment was carried out by using the 3D physics simulation software Gazebo in ROS to verify the feasibility of the algorithm.

As an important part of intelligent driving vehicles and intelligent networked transportation systems, environmental perception technology can provide important decision-making basis for the overall planning of intelligent driving vehicles and transportation systems. This paper reviews the current research on environment perception technology in the current intelligent networked transportation system, and analyzes four key research directions and related progress of environmental sensing technologies, including single sensor device, high-precision map, multi-sensor information fusion and vehicle-road collaboration. On the basis of analyzing and summarizing existing related research, this article elaborates the development trend and key directions of future environmental perception technology, including the integration of deep learning, vehicle-road integration, information security and multi-dimensional perception technology related development directions.

This paper proposed a short-term traffic flow prediction algorithm based on LSTM-SVR, which aims to use machine learning algorithms to solve the high real-time and accuracy requirements faced by short-term traffic flow prediction, and to implemented and compared multiple traffic flow prediction machine learning algorithm performance. In this method, the first 12 segments of 15-minute traffic flow data are reconstructed and normalized and input into the LSTM network. After denormalizing result from LSTM, we input it into the SVR model, and the prediction result is optimized by SVR. The final SVR model output result is the following A 15-minute traffic flow prediction value realizes real-time high-precision prediction of traffic flow.

This paper presents a data-driven simulation model to estimate the potential of replacing conventional taxis with electric autonomous vehicles (EAVs). Vehicle trajectory data collected by onboard global positioning system (GPS) units in Shanghai, China, are used to study taxi travel patterns, in terms of the distribution of taxi travel demand, idle time between two consecutive occupied trips, and the places where vehicle stock imbalances may occur. The operational performances and fleet size are quantified using a data-driven simulation model that stimulates EAV taxis’ charging, idling and relocating activities. It is found that EAV taxis can serve the same amount of travel demand using 69.4% of the current fleet size; while because of vehicle stock imbalances, relocating idling vehicles is inevitable. Simulation results also demonstrate that the deployment of EAVs can improve taxi operation efficiency in terms of increasing the proportion of occupied travel distance.

With the rapid development of science and technology, autonomous driving vehicle has become a research hot spot in the automotive industry, and it also has a great impact on the transportation system. The driving mode of autonomous driving vehicle is more convenient and efficient. The transition from manual driving vehicle to autonomous driving vehicle is changing the state of traditional traffic system. The addition of autonomous driving vehicle can alleviate traffic congestion and improve road capacity. Based on the cellular automata theory and multi-agent theory, in connected intelligent traffic system (ITS), this paper constructs a simulation model of mixed traffic flow of autonomous-manually driving vehicle agent by adopting different car-following rules for autonomous driving vehicles and manual driving vehicles, and analyzes the impact of mixing ratio and reaction time on mixed traffic flow in free-flow, steady-flow and blocked-flow respectively.

In recent years, research on connected and automated vehicles (CAVs) are gradually becoming a hot topic in traffic industry. Recent technological advances in wireless communication, artificial intelligence, and sensory systems lay a foundation for the development of CAV. Because the merging section is an important node in city road system, it is necessary to analyze the capacity of it. Researches on CAVs are mainly based on VISSIM and other traffic simulation software. In this paper, the capacity calculation formula of merging area based on the Cowan M3 distribution is established. The formula includes the number of mainline lanes, the critical merge gap and other parameters. Besides, the capacity of the merging area under two simulation environments is discussed in this paper, which is CAV in cooperative control method and automated vehicles (AVs) without cooperative merging control. The paper takes the three mainline lanes as an example.

With the development of autonomous driving technology, China, based on its national conditions and existing advantages, has clarified the technological development path of Vehicle-to-Infrastructure cooperation system to realize a series of high-end auton omous driving functions such as network-wide dispatching and beyond visual range perception. Compared with the design of the existing cloud control platform, this article clarifies that the expressway operation management unit is the service object. The functional architecture and technical architecture of the cloud control platform are derived through the combing of car-road collaborative application scenarios.

Connected and autonomous vehicles (CAVs) are regarded as the new generation vehicles. Nowadays, it is an ad-hoc worldwide research spot, especially the highly or fully CAVs. These high-level CAVs have not been put into the market yet. Since the innovation of CAV technologies keeps emerging, it is still a meaningful task to following the up-to-date public perceptions and expectations on these high-level CAVs. Using the Stated preference (SP) method, this study designed a questionnaire that included three kinds of questions, i.e., basic information, perceptions, and requirements of future potential users. The questionnaires were delivered by the social media, and finally 612 valid feedbacks were collected. Following a statistical analysis on the survey, a correlation analysis were applied to find the relationship between the individual backgrounds and their perceptions/expectations.

To address issues caused by urban traditional delivery, this study investigates the application of smart underground logistics system in central business district (CBD): builds the framework of CBD smart underground logistics system as “Consolidation Center +Smart Transport+ Smart Collection Center+ Operation Platform”; discusses the functions and planning key points of each component; compares the operational costs and benefits to traditional delivery mode; provides suggestions for further planning, design, and implementation of smart underground logistics system.

In order to reduce the blocking of traffic flow of the signalized intersection on the urban road, for individual vehicles can interact the information with the roadside facilities and the intersection control system under the connected vehicle environment, a speed control strategy in the signalized intersection is proposed. The method consists of two levels, i.e., optimal control range and onboard vehicle speed control. The paper calculates the optimal traffic velocity and vehicles’ arrival time to minimize the total travel time of all vehicles. The purpose of the vehicle passing through the intersection without stops under the guidance of speed guiding strategy could be achieved by analyzing the speed and position of the vehicle and judging whether the vehicle under different driving states can pass. The proposed speed control strategy was analyzed and evaluated in the established simulation environment based on the microscopic traffic simulator VISSIM and Python.

In recent years, with the development of autonomous driving technology, the unsafe, unreliable and unstable of autonomous driving systems have gradually been exposed. It is necessary to build test scenarios through a large amount of data in the real driving scene database to conduct comprehensive and strict tests and evaluation. There is a lack of an autonomous vehicle test scenario library data system that provides unified standard data. This paper designs and implements a test scenario library data system for autonomous vehicles. It explains the build steps of data system and analyses the methods of the implementation. It explains the build steps of data system and analyses the methods of the implementation. The data system contains real driving scene management module and test scenario management module. The system is based on B/S structure and decomposed into front-end and back-end.

In order to study the problem of short-term traffic prediction more effectively, experts and scholars have put forward various car-following models. Among data-driven algorithms, k-nearest neighbor algorithm is the most widely used due to its simplicity, flexibility and high accuracy. In this paper, the three-dimensional model based on improved knn algorithm is constructed. It transforms the input vector of basic knn algorithm into a three-row matrix. Data of three dimensions are considered, including only the previous moment data, short period of history data and long period of history data. Three-dimensional matrixes are constructed for prediction and similarity is measured by sum of 2-norm of 3 row vectors. Besides, weighted distance calculation method based on temporal distance is introduced to differentiate impacts data of different dimensions have on the results.

With the rapid development of autonomous driving technologies, the proportion of autonomous vehicles (AVs) will increase and influence road capacity. In this study, the simulation of mixed traffic flow was studied using an improved cellular automata model. Safety inter-vehicle spacing, the length of vehicles and reaction time are introduced into the cellular automata model. We delete the acceleration, deceleration and randomization rule for ideal conditions. Numerical simulations are utilized to analyze road capacity with different proportions of AVs. Road capacity is about 2200 pcu/h/lane for pure manual vehicle (MV) traffic flow and about 3600 pcu/h/lane for pure AV traffic flow. The capacity increases by 19.2% when there are 50% AVs in the traffic flow. And the capacity increases by around 63.6% due to the pure AV traffic flow.

The recent research on location approaches of the intelligent vehicle based on Light Detection and Ranging (LiDAR) is analyzed in this paper. According to the features of these approaches, it can be divided into three categories: simultaneous localization and mapping (SLAM), offline mapping and online localization (OMOL) and fusion localization (FL). Past research and applications of the main algorithms and critical research scenarios in each localization approaches are reviewed. Three aspects of the current trend in location approaches of the intelligent vehicle based on LiDAR are discussed. Based on object detection, object recognition and object analysis algorithms in the field of deep learning, semantic SLAM and real-time three-dimensional reconstruction are important research trends for SLAM. The performance of robustness and real-time performance of localization algorithm of intelligent vehicles based on LiDAR need to be improved.

Vehicle localization is an important part for the sensing function of automated driving. Previous studies can perceive the localization of vehicle based on the on-board lidar and camera with transformations of coordinated system. However, these vision-based and lidar-based localization methods have the complicated transformation calculations of coordinated system and on-board localization devices easily cause the errors arising from the vibration during the driving process. To address such a problem, a new solution of roadside localization-based method can be proposed with a core idea of these sensing devices implemented into the roadside. Firstly, the working mechanism of two localization system including vision-based and lidar-based is described, and the implementation steps can be further proposed for improve the performance of vehicle localization. A deployable model of roadside sensing devices is formulated to solve the deployment problems of multiple localization system.

With the development of vehicle-road coordination technology, driving modes of the vehicle are in the process of development from manual driving, assisted driving, autonomous driving, mixed driving between people and vehicles to advanced unmanned driving. Heterogeneous traffic flows are essential for the development of vehicle-road coordination systems. However, in real life, it is necessary to intelligently monitor heterogeneous traffic flow because they involve many types of vehicles, complex scenarios, complex hidden factors in traffic conditions, and different operating characteristics of vehicles at different times and places.