Proceedings of the Second International Conference on Intelligent Transportation

Frontmatter

A New Feature Evaluation Algorithm and Its Application to Fault of High-Speed Railway

Abstract

Multi-criterion feature ranking algorithms can ease the difficulty on selecting appropriate ranking criterion caused by single-ranking algorithms, and improve the reliability of feature ranking results. However, the issue of conflict between different single-ranking algorithms is often overlooked. By treating this task as a search and optimization process, it is possible to use the D-S theory and evidence conflict to reduce conflicts between different single-criterions and improve the stability of feature evaluation. This work presents a new multi-criterion feature ranking algorithm based on D-S theory and evidence conflict theory combining different criteria improving classification performance of feature selection results. Comparison between the new algorithm and Borda Count, Fuzzy Entropy, Fisher’s Ratio and Representation Entropy methods are done on train fault dataset. The obtained results from the experiment demonstrate that the new algorithm has highest classification accuracy than the other four criterions on all cases considered.

Jing Du, Weidong Jin, Zhenzhen Cai, Fei Zhu, Zhidan Wu

Steering Angle Balance Control Method for Rider-Less Bicycle Based on ADAMS

Abstract

Bicycle is a typical unstable system. Although it is partly similar to the inverted pendulum systems, the complex mechanical characteristics cause that it is difficult to be analyzed and controlled. In this paper, Constrained Lagrangian method is used to building a rider-less bicycle dynamic model. Through the analysis of the dynamic model, the control system is designed to be divided into two parts: the forward control and the balance control. The rear wheel gets a stable speed by controlling the torque of the motor. Steering Angle Balance Control method is designed to keep the system balance. Use ADAMS to simulate a rider-less bicycle control system. The simulation shows that using Steering Angle Balance Control method can balance the bicycle to run forward and swerve stably.

Yuanyuan Feng, Runjia Du, Yuping Xu

The Management of Slight Traffic Accident Based on the Internet

Abstract

With the rapid development of social economy, the number of motor vehicles is increasing rapidly, especially the number of private cars, which leads to many urban traffic congestion. The rate of traffic accident is increasing, therein slight traffic accidents account for the most. Using mobile Internet technology, remote guidance without casualties and accidents by the two sides to negotiate and determine the responsibility, shorten the time of the accident waiting and processing, can greatly ease the traffic congestion problem, optimization and promotion based on the Internet traffic minor accident processing mode, is an effective measure to achieve intelligent traffic.

Desheng Yu, Laimei Fu, Gong Qin

Application of FBG Sensors in High-Pile Wharf Structure Monitoring System

Abstract

Structure monitoring technology is gradually applied in the field of hydraulic structure; However, the traditional electrical sensor technology has some limitations in the field of hydraulic structure monitoring system, especially for long-term and real-time structure monitoring. In order to achieve long-term and real-time monitoring of hydraulic structures, Fiber Bragg Grating (FBG) sensor technology is adopted in engineering practice in this paper, and first used in a high-pile wharf structure monitoring system of a supported project, the results of loading test and finite analysis verified that FBG sensors have great advantages in hydraulic structure monitoring.

Jiashuai Xu, Guangshuang Ge

The Rail Transportation and Bus Intelligent Connection Under the Comprehensive Transportation System

Abstract

In order to alleviate traffic pressure, most cities in the world using the development of rail transportation as backbone, buses as the main body of the urban transportation network structure. This article puts forward to building the integrated intelligent scheduling system based on Internet platform in order to improve the efficiency, unblocked and consistency of the transfer. This way can realize the goal of the rail transportation and public traffic intelligent connection. This system includes passengers and scheduling command two sub-systems, two subsystems contact with each other and cooperation.

Yong Hu, Min Huo, Gong Qin

Development of a Surrogate Conflict Indicator for Freeway Exits Using Trajectory Data

Abstract

Crash occurrences are random rare events which are difficult to detect and reconstruct. Evaluating traffic safety using crash surrogate measures instead of historical crash data is attracting more and more attention. The trajectory, a continuous function of a vehicle’s temporal-spatial kinestate, is capable of depicting the process of crash potentials by means of theoretical analysis or simulation. The trajectory provides easy access to accurate calculation of the commonly used safety surrogate indicators like TTC and PET, which are difficult to obtain in practice. TTC and PET are the time differences between two vehicles during a quasi accident process. Compared with the situation in the final conflicting point, two vehicles may encounter a smaller distance or a bigger speed variation while approaching the conflict point. TTC and PET are deficient in describing abreast driving, or distinguishing the severity levels for approximately equal TTC (PET) cases. To remedy these shortcomings, this paper proposes a surrogate indicator Kj, the ratio of conflicting distance divided by relative speed. And an exponential model is developed to predict the conflict probability. Kj and the conflict probability are both time frame based, illustrating the changing process at each time frame during a conflict phase. The indicator Kj and conflict probability make it easier to describe the conflict mechanism and distinguish the levels of conflict severity.

Tangyi Guo, Xuejiao Jiang, Wei Fan

A More Practical Traffic Lights Cellular Automata Model for Traffic Flow Simulation

Abstract

Based on real collected data to analyze the effects to vehicle speed near the line of parking area under different remaining of green time with or without signal countdown device on urban roads. Then we can analysis of drivers’ characteristics and its spatio-temporal characteristics. Conclusion is drawn as follows: The influence of signal lights on vehicle speed, which mainly concentrate in the 70 m area near the stop line; During the green light period, at the same time point, the operational characteristics of vehicles on different positions have significant difference, which changes with the passing of green time; Compared with no signal countdown case, the vehicle speed fluctuation is larger and the average speed is obviously lower in signal countdown case; Under the traffic without blocking, most drivers want to increase speed through the stop-line. Eventually, by performing the analysis of pilot behaviors in with or without signal countdown device, an attempt is made to provide the urban road entrance lanes of more realistic CA model, which describes the operational characteristics of urban road entrance lanes with the signal controlling has obvious advantages through simulation results.

Xinlu Ma, Bei Li, Shijian Zhang, Xiaodong Guan

Robot Driver’s Motion Analysis and Simulation Based on Vehicle Speed Control

Abstract

In order to improve the control performance of the robot driver, the dynamic analysis of its key parts such as throttle and brake mechanical legs are carried out, furthermore, a co-simulation platform of the robot based on speed control is built, and the movement of the robot in the process is analyzed. The mechanical leg’s dynamic simulation model is built by ADAMS, and then the vehicle model is built by CarSim while the electromechanical co-simulation model “Robot Driver—Vehicle” is built based on the closed-loop speed control in MATLAB/Simulink. Simulation results show that the dynamic model of mechanical leg has good dynamic response; In addition, the electromechanical co-simulation model is able to complete the basic speed tracking simulation experiment, so as to provide a virtual simulation platform to improve the mechanical structure and control strategy.

Kunming Liu, Guoyan Xu, Guizhen Yu

Comparative Analysis the Rectifier Part of Electric Locomotive Auxiliary Converter

Abstract

Based on the development of the auxiliary converter of SS4B electric locomotive, the key research is focused on the front stage of the rectifier. Comparative analysis the two widely used in the auxiliary converter circuit topology (PWM rectifying circuit and phase-controlled rectifier), and finally choose the phase controlled rectifier circuit for auxiliary converter front stage the rectification scheme.

Quanzhu Zhang, Xiaohui Lu, Min Lei, Yonghong Deng

An Improved Multi-Kernel Estimation Method for Vehicle Localization

Abstract

Vehicle localization is one of the key functions in Intelligent Vehicles (IVs). Localization result is usually provided by the combination of GPS data and additional sensors, which are able to improve the localization precision. In this paper, a monocular camera and map database help vehicle localization, aiming to enhancing the localization performance. To this end, map-based road lane markings are constructed according to open source map. Then, vision-based markings and map-based markings are fused to obtain the improved vehicle fix, using an improved Multi-Kernel Estimation (MKE) method. The results using real data show that our method leads to an obvious improvement in vehicle localization accuracy.

Wenjie Lu, Liangwei Jiang, Min Wang, Qing Ma

The Calculation Method with Grubbs Test for Real-Time Saturation Flow Rate at Signalized Intersection

Abstract

In order to provide accurate and real time method for the calculation of saturation flow rate at signalized intersection, by the analysis of the motion law of traffic and the composition of the queue, this paper proposes that it should to eliminate the data which more probability contains outlier data and exclude the outliers in the rest of the measured headway. Finally, the saturated headway and saturation flow rate are calculated. The effectiveness of the proposed approach is verified through an example.

Weiming Luo, Yang Wu, Jianhua Yuan, Wenjie Lu

Traffic Optimization Design and Simulation Evaluation of Isolated Intersection

Abstract

Scientific optimization design of isolated intersection, which as the key point of city arterial road, is of great significance to improve traffic capability. The core intersection on the arterial road in Shenyang was selected as research object. Based on traffic parameter investigation and delay evaluation of intersection status, bus station settings, traffic organization and signal control were optimized and the optimization effect was simulated. The results show that intersection saturation and queue length decrease, while traffic flow average speed increase, further, vehicle average delay is reduced from 60 to 25 s, while intersection level of service is raised from E to C. The optimization effect is obvious and perfect. It suggests that the optimization design of intersection can actually not only reduce the traffic delay, improve the level of service, but also ensure city arterial road being safe and smooth.

Muting Ma, Xiaodong Ma, Yichi Yang, Fengchun Han

Characteristics of Intelligent Transportation Development in Cities

Abstract

With rapid urbanization, environmental pressure on urban resources and ecological degradation, the government’s public management and public demand generated the “Smart City” concept. The Smart City: to log into the “intelligent transport cloud” platform with a mobile phone before going out, query bus arrival information, taxi location information, the location of parking and real-time traffic information and other comprehensive traffic information, to say goodbye to “traffic jam city” [1]. facilitated by intelligent transportation and satellite navigation systems integrated into the communication network, information communicated and exchanged among satellites, the ground receiving station, the terminal equipment and smart devices in people’s hands. Smart city’s application prospects bring opportunities for Beidou satellite navigation system which was independently developed by China. The Chief Engineer of the North Satellite Navigation System, researcher Sun Jiadong, is full of hope for the compass: “The Beidou Navigation Satellite System will not only play a role in the sky, but on the ground it will also apply more applications to benefit people.” Transportation security, traffic congestion and the reduction of environmental pollution will be greatly aided by Intelligent Transportation Technology (ITS). Intelligent Transportation Technology (ITS) in the Smart City optimizes the traffic infrastructure.

Qiu Rui

An Appointment Scheduling Method for China’s Driving License Exam Using Network Flow Modelling

Abstract

In this work, we consider the appointment scheduling problem for China’s driver licenses exam with session capacity constraints. It is modelled as a maximum flow problem solved by Edmonds-Karp algorithm. A more general case with appointment priorities is also considered as a minimum cost flow problem solved by Busacker-Gowan algorithm. Experimental comparison of proposed approaches with conventional used sorting and dispatch method is given, showing the improvement of resource utilization for proposed approaches.

Yongqiang Bao, Huiying Xu, Chenlu Qiu

Integration Design of Highway Traffic Safety Networked Control System

Abstract

Aimed at solving China’s main problems in highway traffic management and security services, we design and construct ministerial, provincial and prefectural highway traffic safety active prevention and control platform. Aggregation, analysis, exchanging and sharing of the cross regional highway traffic safety information and resources, and the real-time early warning on the effect of traffic emergencies are achieved. The construction of highway traffic safety active prevention and control platform will setup rapid and efficient traffic command system, establish new service mechanism, improve the capacity and performance level of traffic police, regulate road traffic order, and effectively prevent and reduce road traffic accidents.

Yongqiang Bao, Huiying Xu, Chenlu Qiu

Analysis of Vehicular Behavior at Bottlenecks Considering Lateral Separation

Abstract

The vehicular behaviors at bottlenecks are complex, and lateral separation makes traffic flow more unstable. Based on the lateral separation characteristic, this paper analyzes the vehicular behaviors at traffic bottlenecks and the movement is described. What’s more, video recognition technology is introduced to obtain the track of the vehicle to verify the decision-making process. The lane-changing behavior and gap-following behavior can be predicted well. And in further study, more factors should also be considered.

X. Shen, Z. C. He

Research on the Method of Driver’s Eye Location Based on MATLAB

Abstract

Accurate location of driver eyes has a positive meaning for eye state detection. This paper presents a new algorithm which is fusion of a mixed color model, integral projection and Prewitt algorithm. First, establishes a mixed color model to locate faces in color space, and finds face area through integral projection and formwork design based on MATLAB. Then eye area is detected through integral projection and a curve optimal. In the end, localizes eyes accurately using Prewitt algorithm and morphologic processing. The effectiveness of the algorithm is verified by the simulation analysis.

Derong Tan, Changmin Lv, Hongjia Zhang, Yang Yu

Catenary Poles Detection and Occlusion Reasoning for Electric Railway Infrastructure Visual Inspection

Abstract

Machine vision based automated inspection and monitoring for railway Infrastructures, such as pantograph overhead lines, is a promising technical trend to increase the efficiency and ease the manpower burdens. Vision data are naturally informative and comply with human sensing and cognition. However, automatically detecting and locate the infrastructure facilities from variance cluttered background in rail-lines inspection videos is still a challenging task due to the ill-pose essential. We propose a catenary poles and gantries segmentation framework by combining the appearance and motion patterns via a sequential Bayesian approach. The poles and supporting arms of power supply lines are detected firstly to yield the region of interest for detailed processing. Then the motion hypothesis of foreground and background are estimated from the edge flow extracted from local curve and line segments, so as to occlusion reasoning. After that, the pole model and background model are implemented to classify the candidates in hypothesis. Finally, promising experimental results demonstrate the potentials of the proposed poles segmentation method with respect to various insignificant patterns and cluttered backgrounds.

Peng Tang, Weidong Jin

Detection and Tracking of Vehicles Based on Video and 2D Radar Information

Abstract

For practical application, over-speed behavior of a vehicle is generally detected by the combination of the radar speedometer and the camera capture. Little information, just a single image of the vehicle and the velocity is recorded as the evidence. If other vehicles are in the sight of the camera, this method cannot identify which one the measured velocity belongs to. The velocity is even inaccurate due to the influence. Focusing on this problem, we introduce a detection and tracking system of vehicles based on the combination of the video and 2D radar information. Multiple vehicles in the sight of the camera are detected and tracked, whose radar information is calibrated and embedded into the image coordinate system. The trajectory and velocity of each vehicle is recorded, and the over-speed judgment of multi-target can be solved. The system is evaluated in the experiments and the practical application. The results show that the proposed fusion algorithm of the video and radar information has a good performance on the detection, tracking and velocity measurement of vehicles.

Min Wang, Liangwei Jiang, Wenjie Lu, Qing Ma

Research on Route-Choice Behavior of Unexpected-Destination Trip Under Random Dynamic Conditions

Abstract

The study of route-choice behavior has been an important part of the study of urban-road transport system. In previous studies, the emergencies of the road network have been studied mostly from different angles under random dynamic conditions. But the condition of route-choice behavior of unexpected-destination trip was rarely mentioned. So in this paper, “unexpected radiation field” is put forward, at the same time, the characteristic of it is also studied. On the basis of this, route-choice behavior of unexpected-destination trip under random dynamic conditions is researched with a combination of “radiation field method” and discrete dynamic Bayesian network. Studies display that the location and important of the unexpected destination affect the route choice behavior closely. The simulation results show that, the model had a positive meaning, and enabled to simulate and reproduce the decision-making process more appropriately.

Xiao-yuan Wang, Jing-lei Zhang, Hai-bo Wang, Chao Yin, Cui-cui Yu

OD Matrix Estimation Based on Mobile Navigation Technology

Abstract

In traditional OD survey, too many manpower and material resources are consumed, but the life cycle of the data is shortening and survey cost is increasing. In tackling these questions, the paper put forward the theory of calculating flow of road segments by mobile navigation software, get all the traffic link flow in current city after fuzzy process, finally obtain estimate OD data. Analysis Nanchang main city zone, adopt STOCH algorithm, stochastic user equilibrium assignment and system optimum realize OD matrix estimation. The results of the example show that OD matrix estimation based on mobile navigation data has feasibility, which also large scale reduce the OD investigation cost at the same time improve the effectiveness of the OD data.

Tangjian Wei, Xiaosong Wang, Yuping Xv, Yuanyuan Feng

Weight-Dependent Equilibrium Solution for Weighted-Sum Multiobjective Optimization

Abstract

As a popular approach to solve Multiobjective Optimization Problem (MOP), weighted-sum (WS) method obtains a series of weight-dependent Pareto Optimalities (i.e. multi-objective global optimums) forming Pateto Front. Each priori (preset) combination of single-objective (SO) weights stands for a certain way to compromise all of SO, e.g. a popular opinion is “Balanced weights lead to the equilibrium solution”. To verify this notion, this paper proposes a method to obtain adaptive posteriori weights derived from heuristic search rather than human-judged priori weights, so as to generate an unique Equilibrium Pateto Optimality (Equi-PO) out of the Pareto Front of multiobjective-function (MOFunc), where mutual interest of every single-objective- function (SOFunc) is achieved to a certain “equal” extent. The numerical example reveal that an unique Equi-PO is obtainable with adaptive weights converging towards an unique end, and: (1) For and only for the WS-MOP whose Pareto Front is symmetric to the Equiangular Utopia Ray, “balanced weights” results in “equilibrium solution”; (2) For other conditions, “balanced weights” can’t.

Yang Wu, Zhiyong Zhang, Jianhua Yuan, Qing Ma

Research on Urban Spatial Structure of Nanchang City Based on Mobile Communication Data

Abstract

With the advent of the age of social mobile and big data, communication devices not only become the necessity of modern life, but also permeate almost every facet of people’s lives. The arrival and development of the information era has brought a series of revolution to people’s special activities, thus causing the change in the cities’ special structure. In recent years, the studies on urban internal spatial structure have always been the hot field of architecture, economics, geography and etc. This paper first explains the internal space structure of the city. According to the situation of limitations of traditional researching methods, the article based on the big data of mobile location-based service that mesh divides the cities and defines the regions where the density of mobile usage is higher than the average as mobile hotspot. Then through the observation of grid mobile hotspot, we can acquire the recognition of urban public system and internal functional divisions under different space and time law. Taking Nanchang City for example, through mobile hotspot method, we can come to the conclusion that the CBD business area in Honggutan region, industrial area in Xiaolan region, and most living area in the eastern part of River Fu qualifies the current city planning of Nanchang, which could provide reference for the planning department.

Yu-ping Xu, Zheng Zhang, Tian-tian Wu

Deterrent Effect of Fixed-Site Speed Enforcement in Freeways

Abstract

Speeding is common in freeways. Speed cameras have limited coverage and the effect of them on speed behavior is unclear. This may lead to irrational decision making on the investment and configuration of enforcement facilities. In this study, required sample speed data were collected in five cross-sections around a speed camera in G45 (Daguang Freeway) of China. Statistics showed that the distance effect model of the speed camera is a concave parabola function. Speeding drivers begin to slow down to speed limits when they approach advance warning signs and then immediately speed up after passing the enforcement site. The deterrent effect of fixed-site speed enforcement is limited to a 300-m range. It was concluded that there is an obvious distance halo effect of fixed-site speed enforcement by deterrence theory. The findings imply that the deterrent effect of fixed-site speed enforcement is weak due to its short effective distance and overt form of control. Reasonable configuration of enforcement facilities and combination with mobile speed enforcement will be appropriate to reduce speeding.

Mengdie Yang, Jun Ma, Qiang Chen, Yichi Yang, Ning Shen

Weather-Responsive Freeway Speed-Limits Using Approximated Friction Coefficient of Road Surface

Abstract

The regular weather-responsive design of variable speed limit (VSL, as a subsystem of Intelligent Transport System) of freeway relys on expensive realtime detection of road surface friction coefficient, not available for economy-underdeveloped regions. A low-cost approach is proposed to design 2 types of weather-responsive VSL (straight-running VSL subject to visible distance, and curve-running VSL aimed at skid-avoidance) in the absence of friction coefficient, using equivalently-approximated discrete friction coefficient (thresholds) depending on the types and intensities of adverse weather elements (i.e. the road surface conditions of dry or wet, further of water or snow/ice covering adhered on road surface), in stead of expensive realtime-detected continuous friction coefficient.

Jianhua Yuan, Yang Wu, Zhiyong Zhang, Qing Ma

Algorithm of Speed-up Turnout Fault Intelligent Diagnosis Based on BP Neural Network

Abstract

Based on analysis of action current curves change law when the speed-up turnout is normal and fault, this paper summarized the current curve eigenvalues, proposed the turnout fault intelligent self-diagnostic algorithms based on change characteristics of the turnout action current curve. Then mapping sample set between action current curve eigenvalues and turnout fault types, and using BP neural network to establish speed-up turnout fault intelligent diagnosis algorithm. The results show that: the fault diagnosis algorithm of the speed-up turnout is high precision and adaptability.

Kai Zhang, Yongfeng Ju, Kai Du, Xu Bao

Research and Application of Traffic Visualization Based on Vehicle GPS Big Data

Abstract

GPS data of vehicle on the road can reflect the actual road status, its analysis can be help to urban road planning, but the multi-source, mass and high dimension features of GPS big data has restricted its application for road traffic. Aiming that, a GPS big data visual computation architecture has been design in this paper, speed attribute and two-pass corner detection has been introduce to improve map matching and clustering analysis methods. Vehicle GPS data of Zibo city has been selected as the case, the relevant result shows that the improved methods can be effective and visual, and can get a better effect than flow map method for vehicle GPS big data processing.

Xin Wang, Shuxu Zhao, Liang Dong

A CA Model with Variable Cell Size for Passengers Behavior in Subway

Abstract

The Cellular Automaton (CA) is a kind of discrete system whose dynamic performance depends on simple reactions among single cells. It has been widely applied in many fields nowadays because it can mimic some intricate situations. The present designs of cellular automaton, however, are mostly such systems with stationary cell size. In particular cases, there could be some huge differences between simulation and real statistics. The CA model with variable cell size (CA-VCS) then becomes significant, and the design for extended CA based on the existing model is proposed to simulate the movement of passengers in subway. In most cases, the cell size is referred to psychological size not actually physical body size. With the consideration of high density in subway in rush hours, Cells could alter their cell size based on the normal CA models to describe passengers much accurately according to the condition around. When meeting highly dense population, cells will be compressed and transformed into small size until they could find enough space to enlarge their size. The simulation based on the defined movement rules of cells shows the necessary processes for the cell transformation. The relative research and its results provide the proof to apply the proposed model in analysis of subway.

Yifan Zhuang, Yichen Zheng, Yi Zhang, Xudong Xie

Erratum to: Proceedings of the Second International Conference on Intelligent Transportation

Huapu Lu (Ed.)