Ebook: Information Technology and Intelligent Transportation Systems

Intelligent Transportation Systems (ITS) are the direction of development for future transportation systems. They are the model for integrating advanced information technology, data communication transmission technology, electronic sensing technology, control technology and computer technology into a comprehensive ground traffic management system. ITS can effectively utilize existing transport facilities, reduce traffic load and environmental pollution, ensure traffic safety and improve transport efficiency. For this reason, ITS have attracted a great deal of attention from many countries. The development of intelligent transportation is inseparable from the development of the internet of things (IoT). ITS is the embodiment of transportation, and means that the 21st century will become the century of intelligent traffic. The ITS that people are most likely to adopt is an advanced integrated traffic management system in which the vehicle runs freely on the road by virtue of its own intelligence, and the highway adjusts traffic flow to the best state by means of its own intelligence. With the help of this system, managers will have a clear understanding of both the road and the vehicle's whereabouts. Intelligent transportation is a service system based on modern electronic information technology for transportation. It collects, processes, publishes, exchanges, analyzes and uses information as the main basis for providing diverse services to traffic participants, Artificial Intelligence Information Technology (AIIT) is a general term for the management and processing of various information by means of AI technology. It uses computer science and communication technology to design, develop, install and implement information systems and application software. It is also known as Information and Communications Technology (ICT), and includes sensor technology, computer technology and communications technology.

With this in mind, the 3rd International Conference on Information Technology and Intelligent Transportation Systems (ITITS 2018) was held in Xi'an on 15–16 September 2018. It provided a platform for all professionals and researchers from industry and academia to present and discuss recent advances in the field of information technology and intelligent transportation systems. Intelligent transport systems vary in the technologies they apply, from basic management systems to more application-based systems. Information technology – including wireless communication, computational technologies, floating car data/floating cellular data, sensor technologies, and video vehicle detection – is also intrinsic to intelligent transportation systems, and the technologies of intelligent transportation systems also include aspects of theoretical and applied topics such as emergency vehicle notification systems, automatic road enforcement, collision avoidance systems and some cooperative systems. The conference also fosters cooperation between organizations and researchers involved in other emerging fields. Internationally known professors have been invited to further explore these topics and discuss technical presentations in depth. All papers have been reviewed by 3–4 referees, and the program chairs of the conference committee made their selections based on the score of each paper. This year, ITITS 2018 received more than 168 papers from 4 countries, of which 41 papers were accepted.

ITITS 2018 was sponsored by the Shaanxi Computer Society, Chang'an University and co-sponsored by Xi'an University of Technology, Northwestern Polytechnical

University, CAS, Shaanxi Sirui Advanced Materials CO., LTD, and the Special Aircraft Engineering Research Institute, among others.

Prof. Derong Liu, Fellow of Chinese Academy Science, China; Prof. Bogdan M. Wilamowski from Auburn University, USA; Prof. Michio Sugeno from the Tokyo Institute of Technology, Japan; and Prof. Chin-Chen Chang from Feng Chia University, Taiwan were invited as the International Advisory Board. ITITS 2018 also invited Dr. Sungwon Lee; Dr. Jaehyun So; Dr. Hong-Mo Seong from the Korean Transport Institute, Republic of Korea; Prof. Bazhanov Vladimir from Samara State Transport University, Russia; and Prof. Yonghui Li from the University of Sydney, Australia as keynote speakers. There were more than 23 oral presentations and 18 poster presentations in the conference.

We wish to express our gratitude to all those involved in the organization of the conference. We are also grateful to all our sponsors, keynote speakers, paper and poster presenters and reviewers for making ITITS 2018 a great success.

Lakhmi C. Jain, University of Canberra, Australia, University of Technology Sydney, Australia and Liverpool Hope University, UK

Xiangmo Zhao, Chang'an University, China

Valentina E. Balas, University “Aurel Vlaicu” Arad, Romania

Fuqian Shi, Wenzhou Medical University, China

Nowadays, English as a Second Language (ESL) learners in China often make errors in grammar while writing English essay. Different approaches to high-quality Grammatical Error Correction (GEC) have been proposed recently. Two typical approaches are rule-based method and statistical-based method when correcting grammatical errors in English essays. The former can't cover all grammar errors in English essays of ESL learners and this led to low accuracy when the former corrects grammar errors in English essays of ESL learners. In contrast, the latter uses statistical approaches to train English essays of ESL learners and get the knowledge of their grammatical errors. This paper proposes GEC method for Chinese English learners and a hybrid approach which combined statistical-based method and rule-based method to GEC. On the one hand, some rules are introduced to solve most of grammar errors, for instance, subject-verb agreement is more complicated and is more effective to determine the errors by linguistic and grammatical rules. On the other hand, the statistical-based method has the advantage in the correction of replacement preposition. The experiment shows that it is able to detect and correct many kinds of grammatical errors and obtained a higher F0.5 score for correction.

Parking problem becomes one of the major issues in the city development. However, accuracy parking occupancy detection is one of the fundamental issues. This paper addresses this issue from the perspective of detecting the parking vehicles using magnetic sensors. Our research focuses on feature analysis of magnetic distortions induced by parking vehicles in different conditions. Besides, an accuracy parking detection solution based on signal feature and information collaborative mechanism has been proposed. By using signal feature of parking vehicle, more information of vehicle has been exploited to improve the performance of conventional state machine method. In order to improve the accuracy of detection, an information collaborative mechanism of adjacent magnetic sensors has been introduced to tackle the interferences from adjacent vehicles. To test our solution particularly in the challenging situations, real experiments were carried out in both under-ground and ground parking lot. The experiments reveal a high accuracy of about 99.77% in the un-ground and 99.54% in the ground parking plot. Our solution has a significant increase in detection accuracy.

Currently, the electric vehicles equipped with an automated mechanical transmission (AMT) are usually employed with the conventional two-parameter gear shift schedule to improve the dynamic and economic performance, especially for the pure electric bus (PEB). However, the conventional gear shift strategy cannot take full advantage of the electrical drivetrain of the PEB. Although the Dynamic Programming (DP) can acquire an optimal gear shift strategy during the whole driving profile, it requires abundant previous knowledge of the whole driving cycle. In this paper, an optimized method of gear shift schedule for the PEB based on a fixed city-bus route is proposed, for which the DP algorithm is employed for three different periods of historical driving conditions to obtain the optimal shift points. And then fitting by the orthogonal least squares polynomial, an optimized gear shift strategy of the PEB based on the city-bus route can be obtained and extracted for an actual application. After a verification of the optimized gear shift schedule by a stochastic driving cycle, the proposed gear shift schedule can be excellently improved than the conventional strategy, but no prominent than DP, compared with the gear shift schedule of the conventional and DP algorithm. However, it can be more easily acquired.

In order to improve the steering stability of four-in-wheel-motor-driven electric vehicle under complex working condition, this paper proposed a stability control strategy with hierarchical structure for direct yaw moment control of four-in-wheel-motor-driven electric vehicle. Based on the vehicle reference model and actual vehicle state parameters, the fuzzy controller was designed to calculate the additional yaw moment in the upper layer, and the sideslip angle was estimated by using the Unscented Kalman filter algorithm. In the lower layer, a torque optimal distribution algorithm was developed with the consideration of the motor limited condition and road adhesion constraint. Then the control strategy model was established and verified under double lane change and sine hysteresis condition in the co-simulation environment of Matlab/Simulink and Carsim. The simulation results indicate that the proposed control strategy, comparing with the uncontrolled condition, can well tracking the lane and improve the steering stability of vehicle.

The cooperation between port and logistics parks is crucial for port to build the collection and distribution system. Using the Decision Tree Learning method, this work establishes evaluation model of correlation between ports and logistics parks in the rear area. Firstly, the correlation factors are analyzed, including the spatial correlation, the functional matching and the service object matching. Secondly, based on a number of industry regulation we generate a training dataset of 105 samples. Thirdly, the Decision Tree is built using the classic ID3 algorithm. All the process is implemented by MATLAB. Last but not least, we analyze the actual statistical data of Shenzhen port and its logistics parks based on the Decision Tree we build, and verify the validity of the evaluation model.

Article output of Chinese transportation engineering and its international effect are the objects of this paper. The SCI online version database (SCI Expanded) as statistical original data to inspect the article output of transportation engineering issued by Chinese author. Having been data cleaning and carding, the science of bibliometric parameters and the cointegration theory are used to calculate the following key elements of research output of transportation engineering, including overall proportion, issuing trends over the years, types of WOS article, and international cooperation relationship of articles, funding institutions, issuing agency and article impact factors. A scient metric analysis of the articles published by Chinese author in the field of transportation engineering is employed. As a consequence, there is marked features in term of quantity and quality of output in transportation engineering. The paper also has quantitative analysis of the general situation of article output of Chinese transportation engineering to provide reference of scientific positioning of development level. In conclusion, this analysis could contribute to the knowledge of scientific output related to transportation engineering field in China.

Structural failure classification for the reinforced concrete (RC) buildings is one of the machine learning challenging tasks. Several successful studies were conducted to train the Neural Network (NN) with well-known optimization techniques. In the current work, a cuckoo search (CS) based classification model of structural failure of the RC buildings was proposed. The proposed NN-CS system was compared to well-known models, namely the Multilayer perceptron feed-forward network (MLP-FFN) trained with scaled conjugate gradient descent and the NN supported by the Particle swarm optimization algorithm (NN-PSO). The performance metrics, including the accuracy, precision, recall, and F-measure were calculated. The experimental results established the superiority of the proposed NN-CS with reasonable improvement (93.33% accuracy) compared to the other models.

With the rapid development of GPS and communication technology, floating car data (FCD), which is also called probe vehicle, become the essential composition of traffic information. In this paper, we propose the model based on the speed-density relationship, to estimate the vehicles between the floating cars at signalized intersection, and then we can obtain the traffic volume at intersection and other evaluation index such as queue length, which is crucial to signal optimization and control. The model is tested using SUMO simulation data. We use the error of traffic volume at intersection as the evaluation index to estimate trajectory data with penetration rates of 20% to 50%. The results indicate that the proposed model is very sensitive to penetration rates, the success rate increases with the growth of penetration rates.

In order to meet various passengers' travel demand, improve passenger service levels and get the optimal running plan foe fast and slow trains, which make the minimum travel time of all passengers, this paper establishes an optimal fast and slow train stop model with passengers' transfer behavior and solves the model and analyzes the data with MATLAB. Under the premise of this article, compared with the traditional plan with the same departure frequency, the optimal plan can save travel time of 10980 minutes, that is, 15.3 s/per. After analyzing the proportion of passengers with different riding distances, it is concluded in this paper, the optimal proportion of passengers is, short distance: middle distance: long distance=20.9%:32.8%:46.4%, and the saving time is 29.3s/per.

In order to estimate the parameters of the vehicle under cornering and steering conditions, the paper proposes two EKF-based observer models. The models consider the influence of the braking/driving force of each individual tire on the whole vehicle. For the purpose of improving the accuracy of the lateral force estimation, the lateral relaxation length is introduced and then observe the influence of relaxation length' change on observer through simulation. Finally, the estimated effect of the observers are verified by double lane-change test under the Co-simulation experiment of Carsim/Simulink. The simulation results show that the proposed method can effectively estimate the vehicle state parameters.

Aiming at the low efficiency of traffic at intersections in urban road networks and the traffic signal control strategy hard to meet changes in upstream traffic flow, a method of intersection traffic signal control and optimization based on timed colored Petri net (TCPN) is presented in this paper. Firstly, the timed colored Petri net models of roads, intersection and signal control system are established, and then an optimization formula for traffic flow with the minimum number of vehicles at the intersection is proposed. Using the number of vehicles on the input and output sections of intersections acquired in 15 phase cycles, the phase timing that satisfies the optimization goal is solved to ensure that the number of output vehicles at the intersection is the largest and the average number of vehicles to be passed on the input road section is the minimum. The simulation results show that the traffic capacity at the intersection is significantly improved, and the average number of vehicles on each output road section increased by 13.3%, 9.7%, 9.8% and 4.3%, respectively.

Passenger traffic volume as the basic data of traffic science management, has extremely high feedback ability and research value. The timely and accurate prediction of passenger traffic volume plays an important role in intelligent traffic management and control. With the rapid increase of traffic data volume in today society, along with highly nonlinear and random traffic systems, the prediction of passenger traffic volume is still a problem. This paper uses the long-term and short-term memory (LSTM) neural network, combined with actual passenger traffic data, through experiments and comparison with other representative forecast models validates that the proposed LSTM network can achieve a better performance, simultaneously analyzed the effects of various input settings on the LSTM forecast performance. In the passenger environment with large quantity and periodic regularity, the calculation time is saved and the prediction precision is improved, and the calculation speed of the same scale prediction problem is greatly improved. It can effectively provide the reference of traffic scheduling and planning during peak hours for the main transport stations with large passenger traffic.

This work integrated several features extraction techniques for a raw sensor data from a wearable assistant consists of three accelerometers sensors for patients suffered from Parkinson's disease (PD) and Freezing of Gait (FOG) symptom during their movement. We considered three types of transformation, namely the one-dimensional Discrete Wavelet Transform (1D DWT), the two-dimensional Discrete Wavelet Transform (2D DWT), and the Fast Fourier Transform (FFT). The extracted features from these transformations were applied to machine learning methods, such as Artificial Neural Network (ANN) to detect FOG. The proposed hybrid system integrates the extracted features from 1D DWT and FFT concluding a total of fifteen extracted features. These hybrid features are then used during the classification process using the ANN that established accuracy of 96.28% for PD detection. The results established that the hybrid 1D DWT-FFT features has the ability to build a robust classification model to detect the FOG accurately.

This research paper proposes a image denoising algorithms by combining two well-known image denoising techniques: i.e. robust bilateral filter and total variation filter, to obtain a better result of image denoising compared to using an individual filter. From the experimental results on standard images data sets, it is shown that the proposed algorithm results in a better-quality image, both numerically and visually. The image denoising results are compared quantitatively in terms of peak signal to noise ratio (PSNR) and it can be seen that combination of robust bilateral filter and total variation filter results in an improved image as compare to only robust bilateral filter or only total variation filter.

Image Denoising method is developed according to the characteristics of energy distribution of noise and wavelet transform. In the first step, through wavelet transform with higher scale, noisy image is decomposed, further in the second step, the square margin of White Gaussian Noise (WGN) or Additive White Gaussian Noise (AWGN) and threshold in high frequency coefficient of wavelet transform with dissimilar scale are shown separately. The coefficients are compared with different values of threshold. At the end, after taking inverse wavelet transform for all coefficient, reconstructed image has been achieved. Experimental results show that the noise is removed from image efficiently and the maximum image information is kept saved.

This paper develops a cooperative adaptive cruise control (CACC) model based on the intelligent driver model (IDM), which incorporates the communicative information of both the leading vehicle and the immediate predecessor. Then, the linear stabilities of the proposed model are theoretically analyzed and numerically simulated. The result shows that compared to models proposed by previous studies, the improvement of traffic flow stability is obtained by taking into consideration the communicative information of the leading vehicle in our paper. The numerical simulations also verify the theoretical results, which indicates the proposed CACC model benefits for linear stability.

The open-circuit voltage (OCV) of batteries is a crucial characteristic parameter which reflects many aspects of a battery's performance, such as capacity, state of charge (SOC) and state of health (SOH). In order to improve and better the applicability of Li-NiCoMn lithium-ion batteries in electric vehicles, a series of charge and discharge experiments were carried out in constant current mode. The battery capability of Li-NiCoMn lithium-ion batteries had been obtained from the experiment results with multiple different type models. Based on the analysis on the charge and discharge curves, the battery capacity as well as SOC can be deduced according to the corresponding OCV. The results of charge and discharge tests indicate that Li-NiCoMn lithium-ion batteries have heavy load charge and discharge performance.

According to the multiple configuration and fault propagation, embryonic electronic cell fault detection is hard and detection rate is low. Therefore, an on-line self-check method for embryonic electronic cell based on encoding fault detection is advanced. First, each module of embryonic electronic cell is detected with parity codes by being compared with configuration information. Next, comparison results are detected. A self-adaption self-check structure based on embryonic electronic cell structure and configuration is designed. The results of simulations in a sequential circuit with single bit fault of the fault set verify the effectiveness of the method.

Sensors used in structural health monitoring (SHM) are key units for data collection. However, their performance cannot be easily confirmed in situ for the absence of proper excitation sources. We present a calibrating method by which the load combination by traffic vehicles is recommended as an excitation source. We then demonstrate the stability of calibrating results under a random excitation source, i.e., random frequency and random amplitude. The stability experiment is performed with a simply supported beam bridge model under random excitation simulated by a programmable linear module. We conduct a dependency analysis by two-way ANOVA, which shows that the frequency and the amplitude of excitation source have no significant effect on calibrating results of sensors.

This study investigates the impact of green supply chain practices including ecological design, green logistics, green purchasing etc. A sample of 231 manufacturing firms' data were collected. By using SEM modeling, the results revealed that except green purchasing remaining green practices have significant and positive effect on the organizational performance. In addition, green purchasing is insignificant effect on organizational performance due to lack of environmental policies, government no seriousness, no financial subsidies on green materials, scarcity of green suppliers, and heavy import duties.

When explosive dust concentrations reach a certain value near an ignition source, it creates an explosion hazard. Therefore, we propose an explosive dust detection and characteristic analysis method using dust particle imagery. First, the Fourier transform domain of the fractional derivative is used to define the image filtering framework. The Rudin – Osher – Fatemi (ROF) model, in a bounded variation imagery function space, is selected to obtain prior noise knowledge with noise variance. Then, the imagery texture region and the nontexture region are divided according to the statistical information of the image local variance. A modified differential evolution particle swarm optimization algorithm is then used to identify the complex dust particles and to determine and update the fitting parameter optimal values, which can separate the overlapping particle intersection points. The model and algorithm are compared and analysed experimentally. The influence of the dust particle parameters is then obtained. We thus demonstrate that the noise suppression and staircase effect are better for the dust images, and that the overlapping particles are effectively separated. The research results indicate the correctness and feasibility of the proposed model, which provides the theoretical and experimental basis for the design of dust explosion concentration intervals.

The WeChat public platform is a Web App that is developed based on the developer mode. It can realize the interaction between the public number developers and individual users. This paper uses the message request & response interface provided in the developer mode to embed business information such as product information, popular products, cultural promotion, and product sales into the WeChat application. This paper successfully tested the public number's provision of location information, message reply and merchandise ordering functions on the mobile phone side. It has also been used by relevant commercial enterprises to solve the problem of the development of commercial enterprise marketing on mobile terminals.

In this paper, an improved LDA topic model called SLDA is proposed for analyzing English essay of Chinese learners, which is used to analysis sentiment of English essay and judge whether the content of the English essay is on-topic, and it performs well combined with English essay automatic correction system. The corrected LDA topic model is added a sentence level analysis process to make up for the shortcoming of the method based on bag of words that ignoring the relationship between words, and it overcomes the shortcoming that model needs lots of training text. The model proposed in this paper mainly aims at the automatic analysis of the overall thought expression of Chinese English learners, which can be applied to the automatic correction system for analyzing English essay to serve Chinese English learners, and experiments show that it performs well in practice.

Speech Emotion Recognition (SER) has become a hot topic recently. In this paper, Back Propagation Neural Network (BPNN) was used as a training system for SER classification, and four emotional speeches of the German Berlin Emotional Database (EMO-DB) were selected as the experimental data-set. The recognition accuracy was compared under different number of nodes in the hidden layer, and the best classification model was determined by combining the training time and the mean squared error (MSE). The experimental results showed that when the number of nodes in the hidden layer is 14, the MSE is minimum, and the average recognition rate of BPNN reaches 98.81%. By compared with different number of nodes in the hidden layer, the average recognition rate increased by 0.2% to 23.5%.