Search

With the continuous development of socio-economics, the rapid increase in the use of road vehicles has led to increasingly severe issues regarding traffic accidents. Timely and accurate detection of road traffic accidents is crucial for mitigating casualties and alleviating traffic congestion. Consequently, road traffic accident detection has become a focal point of research recently. With the assistance of advanced technologies such as deep learning, researchers have designed more accurate and effective methods for detecting road traffic accidents. However, deep learning models are often constrained by the scale and distribution of their training datasets. Presently, datasets specifically tailored for road traffic accident detection suffer from limitations in scale and diversity. Furthermore, influenced by the recent surge in research on intelligent driver assistance systems, datasets from the surveillance perspective (the third-person viewpoint) are fewer than those from the driver's perspective (the first-person viewpoint). Considering these shortcomings, this paper proposes a new dataset, Traffic Accident Detection from the Perspective of Surveillance (TADS). To the best of our knowledge, we are the first to attempt to detect traffic accident under the surveillance perspective with the aid of eye-gaze data. Leveraging the special data components within this dataset, we design the RF-RG model (input: the RGB and optical flow values of the frames; output: the RGB and gaze values of the predicted frame) for detecting road traffic accidents from a surveillance perspective. Comparative experiments and analyses are conducted with existing major detection methods to validate the efficacy of the proposed dataset and the approach. The TADS dataset has been made available at: https://github.com/cyc-gh/TADS/. [ABSTRACT FROM AUTHOR]

Bike-sharing is a powerful solution to urban challenges (e.g., expanding bike communities, lowering transportation costs, alleviating traffic congestion, reducing emissions, and enhancing health). Accurately predicting bike-sharing demand not only ensures the system meets community needs but also optimizes resource allocation, reduces operational costs, and enhances the user experience, thereby increasing the system's sustainability and city-wide benefits. However, prediction is complicated in low-computing environments with insufficient data due to privacy regulations or policy constraints. This study proposes an online learning-based two-stage forecasting model based on a low-computing environment with insufficient data for robust, fast multistep-ahead prediction for bike-sharing demand in Seoul. The model was applied with exploratory data analysis (EDA) to the Seoul Bike-sharing Demand dataset, split into insufficient training and sufficient testing sets. First, we generated prediction values for the random forest, extreme gradient boosting, and Cubist methods in training and testing. Second, we used the Ranger package trained with external factors and prediction values using time-series cross-validation for multistep-ahead prediction 1 h to 1 day later. We compared the model performance with that of 23 machine and deep learning models to verify its superiority. Using interpretability methods and EDA, we reported the relationships between external factors and bike-sharing demand. [ABSTRACT FROM AUTHOR]

With advancing vehicular technology, there are challenges related to the computing capabilities of the deployed infrastructure. In a dense vehicular network, the system performance quickly degrades due to the scarcity of computing capacity and the heavy workload on the coordinating nodes. In situations of a road accident or slow-moving traffic, many trigger messages are generated to enable situational awareness. Aforesaid events may lead to network congestion across the vehicular environment resulting in higher packet loss. Moreover, emergency messages incur increased service delays rather than getting preference for servicing. In this work, we propose a clustering-based re-routing framework for network traffic congestion avoidance on urban vehicular roads. We use queue optimization to categorize and execute tasks based on their priority level at each fog node. Moreover, threshold-based congestion detection is used to determine congested nodes, which, alongside clustering-based suitable node selection, is used for workload sharing. Furthermore, a re-routing mechanism is implemented to decrease packet loss, improving the delivery rate. The simulation results show that the proposed technique is reliable, effective, and robust in terms of packet loss, throughput, and delivery of emergency messages. [ABSTRACT FROM AUTHOR]

Intelligent transportation systems (ITS) have advanced significantly over the past years as an incredible technology for averting congested traffic and enhancing traffic safety. Recent researchers show that incorporating Wireless Sensor Networks (WSN) into ITS can decrease the necessary investment and permits the creation of intelligent collaborative applications that enhance traffic efficiency and driver safety. In this paper, we propose a Trust Aware Clustering Technique based on Rat Swarm Optimizer for WSN-based Intelligent Transportation System (TACTIRSO), which is a secure method for selecting cluster heads (CHs) based on nodes' trust value. We employed the Rat Swarm Optimizer (RSO), one of the most recent swarm-based optimization methods, to efficiently choose CHs. For the selection of CH, the proposed fitness function takes into account the node remaining energy and trust value. Moreover, the exponential moving average model is employed to dynamically change the predefined threshold values according to the network state. In order to enhance the performance of RSO, we applied different local search strategies in addition to an energy and trust-aware method of initializing the rat population. The simulation results reveal that TACTIRSO outperforms existing studies in terms of energy efficiency, selection of most trustworthy nodes, and average network lifetime. Numerical results indicate that TACTIRSO improves average network stability by at least 1.13 times, average trust value by at least 3.31 times, and reliability by at least 4.45 times over other schemes in a heterogeneous network, while these improvements are by 1.02 times, 0.33 times, and 3.52 times, respectively, in a homogeneous network. [ABSTRACT FROM AUTHOR]

In recent years, a new paradigm of Content-Centric Networks has emerged, which employs the prefix content name for addressing in the context of Named Data Networking (NDN). In conventional networks, subscribers send interest packets, and publishers generate related content and publish it to the requesters. In NDN, Caching Routers (CRs), as well as publishers, can also propagate data packets toward the requesters. The NDN network flow has two opposite sides: (1) the interest packet flow, known as interest forwarding on the upstream side, (2) the content flow, known as data publishing, on the downstream side. High traffic volume can be created by the increasing number of requested packets and associated data that are leading to bottlenecks in some parts of the networks. Therefore, congestion control and avoidance are significant issues in NDN. A considerable number of congestion control methods employ forwarding interest rate adjustment coming through the forwarding side. Nevertheless, the congestion can be prevented on the data publishing side using an efficient cache placement method. Congestion avoidance is a process for controlling congestion and balancing traffic loads to make the network effective. In this paper, a Dynamic Cache Placement (DCP) method is proposed to provide congestion avoidance by dynamically relocating the content of the CRs according to the traffic volume pattern and the link capacity. The DCP method distributes the popular data to the network regions with less traffic load and more accessible routers to balance the congested routers' traffic load. The DCP is implemented in the ndnSIM network simulator, and its performance is compared to the conventional method. Simulation results show that DCP is a fair and robust cache placement design, which successfully avoids congestion inflows with highly varying traffic demand. The proposed techniques employed in the DCP method improve the network performance in dynamic circumstances of the network. [ABSTRACT FROM AUTHOR]

Traffic lights are an important controlling factor in traffic flows, and good policies will facilitate traffic congestion. A car's waiting time is highly related to the period in which the traffic lights are green or red; thus, a proper controlling policy reduces the waiting time and speeds up the car movement. Emergency vehicles always have a higher priority than other cars and need to reach their destination faster. In the last decade, a lot of research has been done to solve the traffic light scheduling issue using artificial intelligence techniques. Reinforcement learning with simulating human learning has the outstanding performance to solve complex problems, such as traffic lights control. In this study, a deep reinforcement learning traffic light control was implemented at a crossroad using real-time traffic information with an emphasis on emergency vehicles. The agent learns to adjust its policies to prioritize the emergency cars over the other cars. The results show less waiting time and speed up the cars' movement. It also indicates the emergency vehicles will cross the intersection with the least delay. With 200 to 600 cars and one emergency vehicle, the average delay decreases 2–16% in total using reinforcement learning. Considering more emergency vehicles, there is a 27–40% decrease in average delay for emergency vehicles. [ABSTRACT FROM AUTHOR]

As the global economy develops rapidly, traffic congestion has become a major problem for first-tier cities in various countries. In order to address the problem of failed real-time control of the traffic flow data by the traditional traffic light control as well as malicious attack and other security problems faced by the intelligent traffic light (ITL) control system, a multi-agent distributed ITL control method was proposed based on the fog computing platform and the Q learning algorithm used for the reinforcement learning in this study, and the simulation comparison was conducted by using the simulation platform jointly constructed based on the VISSIM-Excel VBA-MATLAB software. Subsequently, on the basis of puzzle difficulty of the computational Diffie–Helleman (CDH) and Hash Collision, the applicable security control scheme of ITL under the fog computing was proposed. The results reveal that the proposed intelligent control system prolongs the time of green light properly when the number of vehicles increases, thereby reducing the delay time and retention rate of vehicles; the security control scheme of ITL based on the puzzle of CDH is less efficient when the vehicle density increases, while that based on the puzzle of Hash collision is very friendly to the fog equipment. In conclusion, the proposed control method of ITL based on the fog computing and Q learning algorithm can alleviate the traffic congestion effectively, so the proposed method has high security. [ABSTRACT FROM AUTHOR]

In recent years, due to the drastic rise in the number of vehicles and the lack of sufficient infrastructure, traffic jams, air pollution, and fuel consumption have increased in cities. The optimization of timing for traffic lights is one of the solutions for the mentioned problems. Many methods have been introduced to deal with these problems, including reinforcement learning. Although a great number of learning-based methods have been used in traffic signal control, they suffer from poor performance and slow learning convergence. In this paper, a transfer learning-based method for traffic signal control has been proposed. Multi-agent system has also been used for modelling the traffic network and transfer learning has been used to make reinforcement learning agents transfer their experience to each other. Furthermore, a classifier has been utilized to classify the transferred experiences. The results show that using the proposed method leads to a significant improvement on average delay time and convergence time of the learning process. [ABSTRACT FROM AUTHOR]

Cloud computing is attractive mostly because it allows companies to increase and decrease available resources, which makes them seem limitless. Although cloud computing has many advantages, there are still several issues such as unpredictable latency and no mobility support. To overcome these problems, fog computing extends communication, storage, and computation toward the edge of network. Therefore, fog computing may support delay-sensitive applications, which means that the application latency from end users can be improved, and it also decreases energy consumption and traffic congestion. The demand for performance, availability, and reliability in computational systems grows every day. To optimize these features, it is necessary to improve the resource utilization such as CPU, network bandwidth, memory, and storage. Although fog computing extends the cloud computing resources and improves the quality of service, executing capacity planning is an effective approach to arranging a deterministic process for web-related activities, and it is one of the approaches of optimizing web performance. The goal of capacity planning in fog computing is preparing the system for an incoming workload, so we are able to optimize the system's utilization while minimizing the total task execution time, which happens before sending the load toward the cloud environment or not sending it at all. In this paper, we evaluate the performance of a web server running in a fog environment. We also use QoS metrics to plan its capacity. We proposed performance closed-form equations extracted from a continuous-time Markov chain model of the system. [ABSTRACT FROM AUTHOR]

Dockless bike-sharing is more and more popular all over the world, which enables passengers to choose destinations more flexibly, so that the transfer radius between stations is further expanded. The transfer behavior triggered by dockless bicycle sharing has aroused many researchers' interest. Given that this paper aims to measure travelers' transfer tolerance to space interval, waiting interval and their perceived factors between stations, in which both dockless bike-sharing and walking are considered as feasible transfer mode. First, the transfer data collected by GPS application are uploaded to the server database for data mining and processing. Then, we propose a recursive fuzzy linear regression model to explore the functional relationships among passengers' tolerances to distance, waiting time and their perceived factors. Results indicate that passengers who transfer by walk will be more sensitive to distance, while dockless bike-sharing users concern more about the time cost of finding available bike and the delay in traffic signals. Furthermore, our result also proves that pedestrian's tolerance to environment is higher than cyclist who uses dockless bike-sharing system. [ABSTRACT FROM AUTHOR]

In many applications, it is critical to guarantee the in-order delivery of requests from the master cores to the slave cores, so that the requests can be executed in the correct order without requiring buffers. Since in NoCs packets may use different paths and on the other hand traffic congestion varies on different routes, the in-order delivery constraint cannot be met without support. To guarantee the in-order delivery, traditional approaches either use dimension-order routing or employ reordering buffers at network interfaces. Dimension-order routing degrades the performance considerably while the usage of reordering buffers imposes large area overhead. In this paper, we present a mechanism allowing packets to be routed through multiple paths in the network, helping to balance the traffic load while guaranteeing the in-order delivery. The proposed method combines the advantages of both deterministic and adaptive routing algorithms. The simple idea is to use different deterministic algorithms for independent flows. This approach neither requires reordering buffers nor limits packets to use a single path. The algorithm is simple and practical with negligible area overhead over dimension-order routing. The concept is investigated in both 2D and 3D mesh networks. [ABSTRACT FROM AUTHOR]

With advances in technology and the economy, people are more inclined to buy cars. However, as a result, people forfeit their lives and property through accidents due to traffic congestion and traffic chaos. To ensure safe driving, anti-collision warning applications have been widely discussed. Most vehicular safety applications use vehicular positioning and vehicular communication technologies to avoid car accidents. With Vehicular Ad Hoc Networks, it is common to use the global positioning system (GPS) as the vehicular positioning tool owing to its comprehensive development in vehicular navigation. However, there are factors that affect GPS positioning accuracy (e.g., multipath effect, atmospheric effect, etc.). To improve GPS vehicular positioning accuracy, we propose a cooperative positioning algorithm that can improve localization accuracy. We also demonstrate that our algorithm can work well in a real city environment. [ABSTRACT FROM AUTHOR]

A key element in any system based on several interconnected computing and/or storage nodes is the interconnection network. Currently, one of the main concerns of high-speed interconnection network designers is how to improve network performance while using the minimum number of network resources. In that sense, in this paper we describe an efficient switch architecture suitable for any interconnect technology implementing deterministic source-based routing. This switch architecture uses the same network resources to provide two issues that improve network performance: Congestion Management and QoS support. We also present results to compare the effectiveness of this architecture to those of other proposals typically used to provide these issues in this context. These results have been obtained for synthetic traffic and for traces from parallel benchmarks and video frames. From the results, we can conclude that in any traffic scenario, our proposal is as effective as the previous ones, while requiring fewer resources and thus being much more cost-effective. [ABSTRACT FROM AUTHOR]