Ebook: Intelligent Transportation and Smart Cities

Welcome to the proceedings of the 2024 International Conference on Intelligent Transportation and Smart Cities (ICITSC2024). This prestigious gathering of researchers, scholars, practitioners, and industry experts, delves into the dynamic intersection of intelligent transportation systems and the advancement of smart cities. The conference serves as a platform for knowledge exchange, the sharing of innovative ideas, and the cultivation of collaborations that will shape the future of transportation and urban development.

The ICITSC2024 conference brings together experts from diverse disciplines, including computer science, urban planning and data science. It serves as a catalyst for interdisciplinary discussion, aiming to bridge the gap between theory and practice in the realm of intelligent transportation and smart cities. The growing interest in this field is propelled by the need to address the challenges posed by rapid urbanization, increasing traffic congestion, environmental concerns, and the demand for efficient and sustainable transportation systems.

This book of proceedings presents research articles covering topics within the domain of Intelligent Transportation and Smart Cities, which is a multidisciplinary issue, Researches in this field not only focus on the development of basic theories in computer science (IoT, Cloud Computing, Artificial Intelligence, among others), but also on the transfer of advanced theory to engineering applications. We have collected articles which blend different disciplines, such as communication technology, artificial intelligence, edge computing, traffic management, and smart cities, with a focus on articles with practical applications in different engineering fields. These proceedings showcase cutting-edge advancements and breakthroughs, and offer valuable insights into the latest trends, challenges, and opportunities in both fields.

We extend our sincere gratitude to those authors who have contributed their work to this volume, as well as to the reviewers who have dedicated their time and expertise to ensuring the quality and relevance of the presented research. We would also like to express our appreciation to the conference organizers, sponsors, and participants for their unwavering support and commitment to advancing the fields of intelligent transportation and smart cities.

It is our hope that these proceedings will serve as a valuable resource for researchers, practitioners, policymakers, and industry professionals, inspiring further exploration and collaboration in the pursuit of intelligent, sustainable, and inclusive urban mobility solutions within the context of smart cities.

With warm regards,

Conference Chairs

Prof. Vitaliy Mezhuyev, FH JOANNEUM University of Applied Sciences, Austria

Prof. Carlos Becker Westphall, Federal University of Santa Catarina, Brazil

The urban rail transit industry in modern society is growing at an alarming rate, and in the rail transit of the network is the schedule is the basis of the train operation. Current research focuses on designing and optimizing train schedules to better meet passenger demands. In this paper, a multi-objective programming model is established on the basis of objective functions that are coupled and aimed at minimizing operating costs and passenger travel efficiency in order to optimize the train timetable, while considering meeting the demand of passenger flow. One objective function is kept in the original problem according to the characteristics of the model, and other objective functions are transformed into constraint conditions by adding restricted domains, thus turning them into single-objective programming models. Genetic algorithm is used to obtain results and train operation is simulated through the dynamic programming algorithm to carry out dynamic search. The CSMA/CD (Carrier Sense Multiple Access/Collision Detection) protocol is introduced to optimize the constraint conditions. The waiting time is transformed into the minimum tracking time interval by sending the carrier monitoring code. As such, the departure time data of large and small routes are calculated dynamically, and equal interval parallel operation diagrams are drawn. The calculation results indicate that the multi-objective optimization model improved by genetic algorithm can effectively solve practical cases, and its train timetable can highly match the spatiotemporal distribution of passenger flow demand and obtain satisfactory feasible solutions within a reasonable time.

The driver monitoring system constitutes an essential element in the realm of human-machine interaction within intelligent vehicles, primarily tasked with overseeing and promptly alerting deviations from standard driving behaviors that could potentially lead to traffic accidents. Presently, the evolution of driver monitoring systems in China is at a nascent stage. Challenges persist due to constraints in hardware equipment, resulting in relatively simplistic terminal devices for detecting driver fatigue features. Consequently, this has led to frequent occurrences of false positives and missed detections. This study centers around a non-contact vehicle monitoring device employing machine vision. It delves into the development of a rational, effective, real-time, and accurate mechanism for continuously monitoring driving duration while concurrently recording driving behavior data. Within this framework, the research focuses on the utilization of a modified version of MTCNN for real-time driver face detection. It involves extracting critical driver head posture features, encompassing the pitch angle, yaw angle, and roll angle of the driver’s head. By comparing the positions of obtained driver face images, a fatigue driving feature index system is established. This system facilitates the identification, analysis, and discrimination of a driver’s fatigue state based on the extracted head feature values. The ultimate aim is to realize a fatigue warning function within the driver monitoring system, thereby enhancing the detection speed of facial fatigue recognition. This endeavor holds paramount significance for road traffic safety, contributing to the continual improvement of driver monitoring systems and consequently mitigating potential risks on the road.

This paper investigates the fusion analysis of complex multi-directional traffic flows in urban areas based on macroscopic fundamental diagrams with the objective of serving urban traffic management subsystems in Intelligent Transportation Systems (ITS) at rush hour or under special circumstances. Therefore, travel efficiency and regional vehicle accumulation are used as basic parameters to model the macroscopic fundamental diagram based on city-regional road network and traffic flow data. According to the directional influx traffic flow, the city area division is defined, thus a multi-directional traffic flow fusion analysis model can be established. The multi-directional traffic flow fusion analysis model is used to accurately describe the changing characteristics of multi-directional traffic flow and the complex integration between them under the influence of directional incoming traffic in urban road network areas. The analytical model also reveals that in addition to the imposition of boundary control management to limit the influx of traffic, the internal flow in the peripheral area also has a non-negligible impact on the overall traffic system operation.

This paper focuses on the research of longitudinal and lateral instability of dual motor electric vehicles during driving. The main objective is to ensure that the vehicle maintains both longitudinal and lateral stability during acceleration and steering, and also to ensure dynamic performance. The study involves the design of a drive anti-skid steering control system and a handling stability control system, with coordination between the two. To achieve this, the study analyses the vehicle’s different performance and stability requirements at different speeds. It sets a speed threshold where performance is prioritized at low speeds and stability is prioritized at high speeds. Based on these requirements, the activation conditions of the control system are designed for different speed ranges. Results from simulation verification under two different operating conditions - acceleration and steering - show that the control systems can effectively manage wheel slip rates to stay within 0.02 of the optimum slip rate for the road surface when the wheels slip. In addition, the deviation of the actual yaw rate from the setpoint is kept below 16% of the setpoint when the vehicle is experiencing lateral instability. The control strategy developed not only improves the stability and dynamics of the dual-motor electric vehicle but also has academic implications for improving vehicle stability and dynamics in general.

In view of the image information collected by hardware devices under low-light conditions, the signal and noise are relatively low, which will lead to the inability to accurately analyze and process image details and colors in the future. In order to explore the license plate recognition technology based on CNN, this paper improves some recognition steps to strengthen the license plate recognition technology in low-light conditions. The authors use histogram correction and histogram equalization methods in the image pre-processing stage to enhance the image and improve the image quality, and applies the algorithm is applied in low light conditions, which can solve the related problems of low license plate recognition accuracy due to weather or environmental influence, and make the license plate recognition algorithm more suitable for low light environments.

With the developing economy and consumption level, the market demand for agricultural products is getting bigger and bigger, and the requirements for the level of agricultural logistics and distribution are also getting higher and higher, which also brings great opportunities and challenges to China’s agricultural logistics and distribution. Swarm intelligence optimization algorithm includes ant colony genetic algorithm, so this paper is based on ant colony genetic algorithm for logistics distribution path optimization for application research. Agricultural products are perishable, and in the process of logistics distribution, the distributor has to meet the customer’s demand, but also to achieve the minimization of product distribution time and cost. This paper firstly constructs the distribution model through the basic constraints and objective functions of the model; then designs the vehicle path dynamic optimization algorithm and sets the initial population size so as to achieve the optimization of the distribution path; finally, by comparing and analysing the ant colony genetic algorithm and the improved ant colony genetic algorithm, the improved ant colony genetic algorithm is used to analyse different distribution modes.

Bus priority signal control can effectively alleviate traffic congestion problems, but existing theoretical methods have many limitations in engineering practice, which makes them difficult to apply in actual signal machines or have poor control efficiency. Therefore, this article focuses on engineering practice and conducts research on bus priority methods for induction and coordination scenarios, as well as hardware in the loop simulation. In various traffic scenarios, a comprehensive control method for bus priority is established, including dynamic priority, bus priority strategy, green light compensation mechanism, etc. The Jerry traffic signal control machine is combined with the traffic simulation platform SUMO to build a hardware in the loop simulation system. The simulation results show that under the background of induction control and coordinated control, the delay of public transportation can be reduced by more than 26.26% and 20.64%, respectively, indicating that this method can effectively improve the efficiency of public transportation priority control in engineering practice.

With the acceleration of urbanization, traffic congestion has become an important factor affecting urban economic development and the quality of life of residents. How to use technological means to improve the transportation system has become an important task in the construction of smart cities. The emergence of digital twin technology provides a shortcut for the development of smart cities. As a current research hotspot, the construction of smart transportation city systems based on digital twins aims to improve traffic efficiency, reduce traffic congestion, enhance urban traffic safety, and provide personalized strategies through intelligent and data-driven methods, thereby improving the travel experience of urban residents. This paper adopts a comprehensive research method, combining literature review, empirical analysis, and case study, to deeply explore the smart transportation city system based on digital twins. At the same time, advanced technological means such as perception devices, data collection technology, digital twin simulation technology, and intelligent analysis algorithms are also utilized to effectively simulate and restore complex traffic scenes, monitor and analyze urban traffic data in real-time, and provide scientific basis for traffic management and decision-making. Build an efficient and intelligent urban transportation system through the integration of a series of technologies and methods. The research results indicate that the application of digital twin technology can achieve comprehensive perception, accurate simulation, and intelligent regulation of urban transportation systems. This not only helps to solve traffic congestion problems, improve transportation efficiency, but also enhances the travel experience of urban residents.

Mountain roads are the channels for survival in mountainous areas and communication with cities. They are an important support for promoting economic development in mountainous areas. Therefore, it is necessary to establish a prediction model for the accessibility capacity of mountain roads. Combination model refers to the combination of multiple network models to improve prediction accuracy by extracting the advantages of different models. In order to predict the accessibility of mountainous highways, this paper proposes a CNN LightGBM prediction model that combines CNN and LightGBM. This model first utilizes the CNN model to explore the correlation between factors that affect the accessibility of mountainous highways, and constructs an evaluation index system for the accessibility of mountainous highways. Then, this study inputs the extracted feature vectors into the LightGBM model to achieve accessibility prediction. The experimental results indicate that this model has better predictive performance than previous models. This research can help traffic planners make more informed decisions in determining the priority of maintenance and upgrading projects by more accurate assessment of road conditions. And this method can be applied to other regions and road types, which provides valuable insights into the potential of in-depth learning methods in transport infrastructure management.

This article aims to evaluate the application effect of Python in intelligent transportation system (ITS) data analysis. With the development of informatization and digitalization, ITS has become the core of urban traffic management. The article first introduces the concept of ITS, technical framework and principles of Python data analysis, and then analyzes in detail the application of Python in data collection, processing, traffic mode prediction and traffic safety analysis. The research adopted a case analysis method and selected the intelligent transportation system of a certain city for empirical research. Through tools such as Python’s Pandas library and SciPy library, the collected traffic data is processed and analyzed, and models such as linear regression and time series prediction are used to predict and optimize traffic flow and safety. The results show that data analysis using Python effectively improves traffic circulation, reduces accident rates, and optimizes the traffic signal system.

Smart transportation is a new development model for intelligent traffic based on the combination of the Internet, the Internet of Things, and other networks. Road detection is one of the foundational and core technologies in smart transportation. It provides accurate, reliable, and real-time road information for applications such as autonomous driving, intelligent traffic management, and urban planning. This contributes to the improvement of traffic safety, efficiency, and sustainability. However, relying solely on visual images for road detection still faces numerous challenges, such as changes in lighting conditions, image blurring, and occlusions. To enhance the robustness and accuracy of road detection, leveraging LiDAR (Light Detection and Ranging) data as supplementary information is considered. LiDAR data is not affected by visual noise and can offer more precise depth and height information. Nevertheless, effectively integrating and adapting to different data and feature spaces remain crucial challenges. This paper proposes an improved road detection method based on the PLARD (Progressive LiDAR Adaptation for Road Detection) network. It introduces the U-Net network as an additional branch, collaborating with the image and point cloud branches for feature fusion and road detection. The U-Net network utilizes information from the images for initial segmentation, providing a better prior, and enhances the feature fusion capability of the PLARD network, thereby improving segmentation accuracy and robustness. Experiments conducted on the KITTI Road Dataset demonstrate that the proposed method outperforms the PLARD network and other baseline methods in terms of the Dice coefficient. Particularly, it excels in complex urban scenarios and exhibits strong generalization capabilities. The introduced road detection approach, combining LiDAR and visual data, provides an efficient, robust, and versatile solution for smart transportation. It adapts to different scenarios and environments, thereby enhancing the precision and stability of road detection in smart transportation systems.

In order to study the accurate extraction of vehicle traffic parameter information from drone aerial videos to improve urban traffic intelligent management and auxiliary modeling of car-following behavior. A new method for lightweight extraction of vehicle following behavior parameters from drone videos is proposed in this article. The improved ShuffleNet network and GSConv module were introduced into the Yolov7-tiny neural network model as the target detection stage. HOG features and IOU motion metrics are introduced into the DeepSort multi-object tracking algorithm as the tracking matching stage. By building a self-built UAV aerial traffic data set, experiments were conducted to prove that the new method improved a few detection and tracking indicators. In addition, it improves the false detection, missed detection, wrong ID conversion and other phenomena of the previous algorithm, and improves the accuracy and lightweight of multi-target tracking. Finally, the velocity and headway parameters extracted from the car-following behavior using the new method were compared with GPS/INS and proved that the errors were within an acceptable range. The newly proposed traffic flow parameters can be used in traffic flow modeling and simulation to improve the dynamic characteristics and safety of the car-following model, thereby alleviating traffic congestion and improving driving safety.

The high-speed maglev traffic is a new transportation system that combines high speed, safety, green, and intelligence. The test and verification of the high-speed maglev transportation system mainly depends on the test line, which cannot meet the current test and verification requirements due to the large investment amount and long cycle. Semi-physical simulation is a real-time simulation technology that combines physical hardware and simulation software. The characteristics of semi-physical simulation technology include effectiveness, repeatability, economy, and safety. With the development of intelligent transportation, especially in the trend of intelligence and electronics, in recent years, semi-physical simulation technology has also played a crucial role in the development of intelligent traffic control systems. This paper introduces the design and implementation of a high-speed maglev traffic Semi-physical operation simulation system. Based on this system, it can realize the functional performance test and algorithm verification of key components such as suspension, guidance, and eddy brake controller in the laboratory environment, and can also conduct the whole system integration simulation of Shanghai maglev line, 4 stations and 3 sections line,6 stations and 5 sections line, and so on. Verify the rationality and reliability of the system design and support the research and optimization of the key technologies of the maglev transportation system. The simulation results show that the proposed semi-physical operation simulation system can meet the test and verification requirements of high-speed Maglev traffic system.

Traditional fixed signal timing often leads to issues such as green light empty and residual queue at intersections. Traffic signal induction control can automatically adjust signal timing based on vehicle arrivals, but it still struggles to handle immediate control of special traffic events like long queue, exit overflow and intersection deadlock. Furthermore, it does not account for the impact of intelligent connected new energy vehicles (ICNEVs). So we propose an improved method for intersection multi-events traffic signal induction control in ICNEVs. The method involves deploying various types of detectors at intersections to identify events like green light empty, long queue, exit overflow, and intersection deadlock. Establish the induction control logic rules corresponding to different events, and consider the permeability optimization control parameters of ICNEVs at intersection. Finally, the proposed method is validated using real intersection data through the VISSIM traffic simulation software. The results demonstrate that the method effectively reduces the average queue length and delay at intersections. It also reveals that the higher the permeability of ICNEVs, the better the performance of intersection induction control.

At present, the global urban traffic system generally faces multiple challenges such as congestion problems and environmental loads. Reasonable traffic planning by predicting traffic flow is conducive to solving such problems. As for traffic flow forecasting, we need to establish a forecasting model first. The big data technology is used to process the previous historical traffic data as input, and the traffic data is found through the prediction model to predict the future traffic. The prediction of future traffic is used as the output of the model. In this paper, a new spatio-temporal fusion traffic prediction model based on input traffic signal decomposition is proposed, which is called Decomposed Spatial-Temporal Graph Convolutional Network (DSTGCN), in order to better capture global spatial correlation and time dependence. It captures the global spatial information through the diffusion graph convolution network in the spatial module, and then captures the time information through the designed time module. The spatial module and the time module are organically connected by the forgetting gate and the update gate to decompose and predict the input information in multi-level and two dimensions, and capture temporal and spatial correlation of traffic. Moreover, the residual learning framework is used to enhance the model simulation ability. Experiments on PEMS04 and PEMS08 data sets indicate that the proposed method is more excellent than ASTGCN and Graph WaveNet.

China has an extensive road network, thus requiring daily maintenance. During the collection of road defect data sets, scenarios with low illumination or extreme lack of light may occur. In such environments, the quality of image collection can be significantly reduced due to the low visibility of the surrounding environment, leading to the absence of intricate details. These challenges can negatively impact subsequent intricate visual tasks, including object detection and defect recognition utilizing road dataset images, thereby diminishing the precision and efficiency of the recognition process. Therefore, this paper adopts the Retinex algorithm to enhance low-illumination road images and performs road defect detection on the enhanced images, significantly improving the detection accuracy. An enhanced YOLOv8 detection algorithm is introduced to boost the detection efficiency of road defects. This updated algorithm incorporates FasterNet as the feature extraction network, resulting in improved computational speed. Integrating the Deformable-LKA module in the Neck layer can enhance the detection capability for small and irregularly shaped objects. RFAConv, a combination of spatial attention mechanisms and convolution operations, is employed to enhance the detection head and amplify the model’s feature extraction capabilities.The introduced algorithm model undergoes training, validation, and testing procedures on road defect datasets, while also being evaluated against other algorithm models for comparison. The improved algorithm model can achieve accurate recognition of road defects.

The new generation of information technology has had a significant impact on people’s behavior, and making airports more intelligent and intelligent. Airport intelligence is an important component of smart cities and intelligent transportation. Accurate prediction of passenger throughput is not only one of the keys to achieving a smart airport, but also of great significance for airport management and operation. The artificial intelligence prediction method SVR has advantages such as fast convergence speed and suitability for small sample data. The BA algorithm has advantages in parameter optimization, and the combination of the two can effectively improve prediction performance. This article takes Shuangliu Airport in Sichuan Province, China as the research object. In response to the nonlinear characteristics of the airport’s monthly passenger throughput from 2001 to 2019, based on the analysis of the correlation between passenger throughput with historical data and takeoff and landing sorties, a prediction model SVR-BA was established using SVR and BA, while BP-BA and ARMA models were also established. The prediction performance of several types of models was compared from MAPE and MAE, and it was found that the SVR-BA model had better prediction performance and higher robustness. This indicates that after optimizing the parameters of the BA algorithm, the predictive performance of the SVR prediction model can be effectively improved.

To meet the demand for rapid development of intelligent cockpit software in intelligent connected cars, this paper proposes a convenient and reliable development approach. This approach, through the unified definition of interface standards, effectively separates the OTA main control program from vehicle interaction information and user interaction information, thereby promoting the platformization and standardization of OTA UE module core code. In this mode, for the development of different vehicle models, only the IVI (In-Vehicle Infotainment System) needs to implement the vehicle interfaces and interaction interfaces defined by UE, and the UE module code can be reused without the need for repetitive development and debugging work. This improvement not only enhances the stability and reliability of UE code but also significantly increases development efficiency, providing strong support for the rapid development of the intelligent connected car industry.

A digital twin is a system that maps real objects to the virtual world using real-time simulation technology. The application of digital twin technology can enable the self-driving test of real intelligent cars in virtual road scenarios. A digital twin-based in-the-loop test system for intelligent driving vehicles was designed and applied to the development and validation of the Autonomous Emergency Braking system. A vehicle-in-the-loop test system is formed by combining simulation test tools, vehicle to everything (V2X) communication equipment, real test vehicles, and other functional units under the framework of digital twin technology. This system maps virtual and real environments to each other. A mirror simulation scene of the real driving environment is constructed using Unreal Engine 4. V2X communication technology is utilized to transmit real-time state data from the real vehicle to the virtual AirSim vehicle controller. Additionally, the environment perception data of the simulation scene is fed back to the test vehicle. Considering the impact of signal delay in the test system, the AEB control algorithm is optimized and improved through in-loop testing of the entire vehicle under various operating conditions. Real vehicle tests have verified the effectiveness and superiority of the digital twin-based intelligent vehicle test method. The results demonstrate that this test method maintains the authenticity of real vehicle tests and effectively avoids the risks associated with them.

This paper presents the design of a cloud platform for the measurement and control of agricultural UAV. The plateform aims to streamline the management and deployment of UAV by enterprises, help agricultural scientific research institutions in conducting scientific research work, data collection and analysis, and provide basic agricultural data support for the government to making agriculture-related decisions. Based on the characteristics of high reliability, high security and high expansibility of linux on Z, this system adopts openstack, z/VM, hadoop, tomcat server, J2EE and other technologies to build a fast, stable and reliable cloud platform. Beidou navigation system was used to locate the UAV in real time. This system has the characteristics of convenience, fast, real-time and high efficiency, and provides students with a comprehensive application of cloud technology experiment case.

Driven by the new infrastructure of smart city, in order to explore the sustainable development of smart city, this paper starts from the reuse of waste concrete, and studies the mechanical properties of recycled concrete short columns of ribbed thin-walled square steel tube, taking the number of stiffeners and the ratio of height to thickness of stiffeners as parameters. Two types of square steel tube recycled concrete short columns were designed: one with ribs and the other with ribs. The final load capacity and failure mode of ribbed square steel tube recycled concrete short columns were primarily investigated. The mechanical properties of double-ribbed, single-ribbed and non-ribbed square steel tube recycled concrete axial compression short columns were analyzed from the load-displacement relationship. The test results show that the setting of stiffening ribs not only improves the bearing capacity of the specimen, but also effectively slows down the local buckling of the steel pipe. When the width-to-thickness ratio of the specimen is small, increasing the number of stiffeners has little effect on the bearing capacity when the steel ratio remains unchanged. With the increase of the height-thickness ratio of the stiffener, the bearing capacity of the specimen and the stability of the tube wall have an increasing trend. On this basis, the finite element software ABAQUS was used to establish the analysis model. The analysis’s findings demonstrate that the finite element model’s failure mode and load-displacement curve mostly agree with the findings of the experiments.

The utilization of semi-rigidly connected T-nodes in architectural engineering projects within smart cities presents advantages such as improved fatigue life of the structure, enhanced adaptability to the long-term development needs of the urban environment, and increased flexibility and construction convenience for smart city development. Additionally, this discussion delves into the application and arrangement of smart sensors on T-nodes, along with the implementation of a stress-strain cloud monitoring system for T-nodes. To comprehensively understand the impact of T-node flange plate thickness and bolt preload on the internal force of T-node bolts, a series of tensile tests and finite element simulation analyses are conducted on the T-node. These analyses aim to derive the load-strain curve of the T-node, providing valuable insights into the influence laws associated with these design parameters.