Dynamic Traffic Situations Research Articles

Advancements in Reinforcement Learning Algorithms for Autonomous Systems

Reinforcement learning, often known as RL, has developed as a strong paradigm to teach autonomous software agents to make choices in contexts that are both complicated and dynamic. This abstract investigates recent developments and uses of RL in a variety of fields, showing both its transformational potential and the constraints that it faces at present. Recent developments in reinforcement learning (RL) algorithms, in particular deep reinforcement learning (DRL), have made it possible to make major advancements in autonomous decision- making tasks. DRL algorithms can learn complicated representations of state-action spaces by using deep neural networks. This allows for more efficient exploration and exploitation methods to be implemented. Additionally, advancements in algorithmic enhancements, such as prioritized experience replay and distributional reinforcement learning, have improved the stability and sample efficiency of reinforcement learning algorithms, which has made it possible for these algorithms to be used in real-world applications. Robotics, autonomous cars, game playing, finance, and healthcare are just a few of the many fields that may benefit from the use of RL. In the field of robotics, reinforcement learning (RL) makes it possible for autonomous agents to learn how to navigate, manipulate, and move about in environments that are both complicated and unstructured. To improve both safety and efficiency on the road, autonomous cars make use of reinforcement learning (RL) to make decisions in dynamic traffic situations. In finance, RL algorithms are used for portfolio optimization, algorithmic trading, and risk management. These applications serve to improve investment techniques and decision-making procedures. Furthermore, in the field of healthcare, RL supports individualized treatment planning, clinical decision support, and medical image analysis, which enables physicians to provide patients with care that is specifically suited to their needs. Despite the promising improvements and applications, RL is still confronted with several difficulties that restrict its capacity to be widely adopted and scaled. Among these problems are the inefficiency of the sample, the trade-offs between exploration and exploitation, concerns about safety and dependability, and the need for explainability and interpretability in decision-making processes. To effectively address these difficulties, it is necessary to engage in collaborative efforts across several disciplines, conduct research on algorithmic developments, and establish extensive assessment frameworks (Anon, 2022).

VISSIM Based Traffic Flow Simulation Analysis on Road Network

The urban transport system is a heterogeneous system and is in serious trouble. Traffic congestion has become a major socio-economic problem in both urban and rural areas. Traffic congestion is managed through signage, interchanges, and road widening. Traffic analysis remains a difficult task, especially in countries like India where the number of vehicles is constantly increasing and the roads are increasingly busy, because it is difficult to generate alternatives. Traffic simulation systems such as PTV VISSIM are tools dedicated to the analysis of traffic problems. VISSIM is a simulation software that is widely used to create simulations in dynamic traffic situations before evaluating traffic conditions and creating a realistic plan. Dynamic simulation of traffic flow using VISSIM and appropriate techniques to reduce traffic congestion. The main objective is to examine the study area and prepare a sample for it to find out the root causes of the traffic problem and give the necessary suggestions to solve the problem.

Vehicular Traffic Flow Prediction Model Deep Learning

Abstract: Efficient traffic flow prediction is crucial for effective traffic management and congestion reduction in urban areas. However, traditional statistical models often struggle to accurately capture the intricate dynamics of vehicular traffic flow, particularly under dynamic conditions. In this research project, we propose a novel approach that leverages deep learning techniques, specifically Long Short-Term Memory (LSTM) neural networks, AdaBoost, and gradient descent, to enhance the accuracy of traffic flow predictions .By harnessing historical traffic data, our model generates precise predictions for the next time step, empowering traffic managers to optimize signal timings and proactively reroute traffic. To boost the model's performance, we incorporate AdaBoost, which integrates LSTM predictions as additional input features. We evaluate the accuracy of our model using mean absolute error (MAE) and R2 score techniques, comparing the predicted traffic flow against the actual traffic flow .Experimental results demonstrate that our proposed model outperforms traditional statistical models, exhibiting lower MAE and higher R2 scores. This indicates its efficacy in accurately predicting traffic flow and presents a promising solution for traffic management and congestion reduction. Our research contributes to the advancement of traffic flow prediction models by offering a more reliable and accurate approach. Future work may explore the integration of real- time data streams and external factors, such as weather conditions and events, to further enhance prediction accuracy and effectively address dynamic traffic situations. By optimizing traffic management strategies, reducing congestion, and improving overall traffic flow efficiency, our proposed model holds significant potential for improving urban traffic conditions.

Distributed Multi‐MMW Radar Fusion for Target Detection and Tracking in Highway Traffic Environment

High‐resolution millimeter‐wave (MMW) radar is viewed as a low‐cost and highly reliable sensor compared to camera, lidar, etc., in moving scenarios and thus has been selected by highway stakeholders as an important roadside detector to detect the movement of traffic vehicles and monitor traffic flow in real time. However, the echo signal of MMW radar in complex highway environment contains not only the signal reflected by target but also spurious signals and other interference signals, which significantly affects the estimation of the target movement state. To solve this problem, an improved vehicle tracking method is designed to simultaneously estimate the polar angle and polar radius in coordinator of MMW radar. Moreover, considering the movement patterns of target vehicles in dynamic uncertain traffic situations, a set of state space models, such as CA, CV, and CT are combined to represent the vehicle movement. In addition, based on the enhanced detection performance of a single radar, the combination of multiple MMW radars’ information was performed to determine the sequential trajectory of the target vehicle on the continuous road sections; then, the historical trajectory of the target vehicle was correlated and fused. Real experiments in highway scenarios show that the method used in this study is effective in deriving the trajectory of the vehicle and improving the positioning accuracy and reliability when the vehicle performs heavy maneuvers.

Accurate Location in Dynamic Traffic Environment Using Semantic Information and Probabilistic Data Association

High-accurate and real-time localization is the fundamental and challenging task for autonomous driving in a dynamic traffic environment. This paper presents a coordinated positioning strategy that is composed of semantic information and probabilistic data association, which improves the accuracy of SLAM in dynamic traffic settings. First, the improved semantic segmentation network, building on Fast-SCNN, uses the Res2net module instead of the Bottleneck in the global feature extraction to further explore the multi-scale granular features. It achieves the balance between segmentation accuracy and inference speed, leading to consistent performance gains on the coordinated localization task of this paper. Second, a novel scene descriptor combining geometric, semantic, and distributional information is proposed. These descriptors are made up of significant features and their surroundings, which may be unique to a traffic scene, and are used to improve data association quality. Finally, a probabilistic data association is created to find the best estimate using a maximum measurement expectation model. This approach assigns semantic labels to landmarks observed in the environment and is used to correct false negatives in data association. We have evaluated our system with ORB-SLAM2 and DynaSLAM, the most advanced algorithms, to demonstrate its advantages. On the KITTI dataset, the results reveal that our approach outperforms other methods in dynamic traffic situations, especially in highly dynamic scenes, with sub-meter average accuracy.

Dynamic Routing Algorithm for Hazmat Transportation Problems

After a series of gas pipeline explosions in Kaohsiung City, Taiwan, in 2014, the Environmental Protection Administration (EPA) issued regulations in 2017 requiring all chemical tank trucks to be equipped with GPS for real-time monitoring and control. Hazmat transportation problems, along with solution algorithms and applications, have been studied extensively. However, most of the research has focused on the route planning level. Dynamic characteristics should be incorporated to reflect possible dynamic traffic situations with more advanced information and communication capabilities for chemical tank trucks. This research formulates a bi-objective dynamic model and proposes a solution algorithm based on a genetic algorithm (GA) for dynamic hazmat route optimization. A dynamic bi-objective model, including transportation cost and risk, is developed to reflect time-varying traffic conditions. Dynamic traffic characteristics are reflected through traffic volume and travel time, simulated from simulation models. Numerical experiments are conducted in the Kaohsiung City network. The results show that dynamic routes for hazmat transportation vary with respect to traffic flow conditions. The discussions of this research are expected to provide some insights into policy implications.

Traffic Dataset for Dynamic Routing Algorithm in Traffic Simulation

The purpose of this research is to create a simulated environment for teaching algorithms, big data processing, and machine learning. The environment is similar to Google Maps, with the capacity of finding the fastest path between two points in dynamic traffic situations. However, the system is significantly simplified for educational purposes. Students can choose different traffic patterns and program a car to navigate through the traffic dynamically based on the changing traffic. The environments used in the project are VIPLE (Visual IoT/Robotics Programming Language Environment) and a traffic simulator developed in the Unity game engine. This paper focuses on creating realistic traffic data for the traffic simulator and implementing dynamic routing algorithms in VIPLE. The traffic data is generated from the recorded real traffic data published on the Arizona Maricopa County website. Based on the generated traffic data, VIPLE programs are developed to implement the traffic simulation with support for dynamic changing data.

IPDALight: Intensity- and phase duration-aware traffic signal control based on Reinforcement Learning

Reinforcement Learning (RL) has been recognized as one of the most effective methods to optimize traffic signal control. However, due to the inappropriate design of RL elements (i.e., reward and state) for complex traffic dynamics, existing RL-based approaches suffer from slow convergence to optimal traffic signal plans. Meanwhile, to simplify the traffic modeling, most optimization methods assume that the phase duration of traffic signals is constant, which strongly limits the RL capability to search for traffic signal control policies with shorter average vehicle travel time and better GreenWave control. To address these issues, this paper proposes a novel intensity- and phase duration-aware RL-based method named IPDALight for the optimization of traffic signal control. Inspired by the Max Pressure (MP)-based traffic control strategy used in the transportation field, we introduce a new concept named intensity, which ensures that our reward design and state representation can accurately reflect the status of vehicles. By taking the coordination of neighboring intersections into account, our approach enables the fine-tuning of phase duration of traffic signals to adapt to dynamic traffic situations. Comprehensive experimental results on both synthetic and real-world traffic scenarios show that, compared with the state-of-the-art RL methods, IPDALight can not only achieve better average vehicle travel time and greenwave control for various multi-intersection scenarios, but also converge to optimal solutions much faster.

Advancing Knowledge on Situation Comprehension in Dynamic Traffic Situations by Studying Eye Movements to Empty Spatial Locations.

This study used the looking-at-nothing phenomenon to explore situation awareness (SA) and the effects of working memory (WM) load in driving situations. While driving, people develop a mental representation of the environment. Since errors in retrieving information from this representation can have fatal consequences, it is essential for road safety to investigate this process. During retrieval, people tend to fixate spatial positions of visually encoded information, even if it is no longer available at that location. Previous research has shown that this "looking-at-nothing" behavior can be used to trace retrieval processes. In a video-based laboratory experiment with 2 (WM) x 3 (SA level) within-subjects design, participants (N = 33) viewed a reduced screen and evaluated auditory statements relating to different SA levels on previously seen dynamic traffic scenarios while eye movements were recorded. When retrieving information, subjects more frequently fixated emptied spatial locations associated with the information relevant for the probed SA level. The retrieval of anticipations (SA level 3) in contrast to the other SA level information resulted in more frequent gaze transitions that corresponded to the spatial dynamics of future driving behavior. The results support the idea that people build a visual-spatial mental image of a driving situation. Different gaze patterns when retrieving level-specific information indicate divergent retrieval processes. Potential applications include developing new methodologies to assess the mental representation and SA of drivers objectively.

An Analysis of Air Traffic Controllers’ Job Satisfaction

The air traffic controllers' job is one of the most hectic in today’s world, predominantly due to its safety-critical operations and altering expectations. The primary purpose of this paper is to provide a holistic directory of determinants and synthesized reinforcements for air traffic controllers' job satisfaction. Researchers in the past have put the spotlight on individual air traffic controller’s technical job satisfaction factors, such as impacts from remote tower operation, airplane trajectory changes, and dynamic air traffic situations. However, none described the connection among those factors and how adjusting those factors can enhance the cognitive components related to their job satisfaction. This paper's in-depth analysis identified factors contributing to air traffic controllers' job satisfaction based on past literature. It is intended to increase understanding and improve knowledge for future researchers and practitioners. The five predominant factors identified for air traffic controllers’ job satisfaction are ambiguity of job functions, overwhelming workload, complex task performance and uncertain work demand, job fatigue, and work-family conflict. Some effective methods to increase air traffic controllers’ job satisfaction are regular break between shifts, technological advancement to facilitate jobs, and sound insulations.

Improve adaptive cruise control under driver distraction: Time headway compensation based on a random‐effect crash risk model

Human drivers conduct compensatory behaviour to counteract the increased risk while being distracted. This kind of compensation strategy should be learned for a safer and smart design of adaptive cruise control system (ACC). Hence, a simulator study was conducted, requiring performance visual, cognitive, and combined secondary tasks during a car following scenario. An increased time headway (THW) was found in all of the three distraction conditions, which confirmed that drivers compensated an extra THW to counteract the increased crash risk. Furthermore, crash probability models using binary logistic regression with random intercept were constructed where driver distraction and dynamic traffic situations were embodied as inputs. Results showed that crash risk increased with reduced THW, increased lead vehicle deceleration, and unopened brake light of the lead vehicle. Besides, visual-related distractions increased crash risk, while pure cognitive distraction lowered crash risk in low THW (lower than 1.8 s) condition and increased crash risk in high THW (larger than 1.8 s) condition. Based on the authors' proposed models, theoretical compensation in THW to fully counteract the increased crash risk by distraction was derived, which could be used for the design of a human-like ACC with automatic adjustment of THW setting considering driver distraction.

Look ahead: understanding cognitive anticipatory processes based on situational characteristics in dynamic traffic situations

Understanding the intentions of other road users increases subjective certainty, improves one's own performance and contributes to enhanced traffic safety. The perception and processing of situational characteristics are pivotal to develop an appropriate situation model as the basis of anticipation in dynamic traffic situations. To gather further insights into the process of situation comprehension, different situational characteristics were tested in a laboratory experiment using urban traffic scenarios. Participants had to indicate, in a two-step reaction method, the initial supposition and certain anticipation of another car merging into their own lane while being cognitively distracted (high versus low). Retrieval cues that enable the formation of causal relations and the development of a coherent mental representation increased the probability of anticipation. Furthermore, different situational characteristics were identified based on distinct underlying processing patterns. Cues that imply a close relationship with the target's behaviour (target cues) are assumed to require fewer inferences. They were less affected by increased cognitive load compared to other situational characteristics (context cues). These findings enhance the understanding of human anticipation and are discussed regarding the methodological approach and the generalisation of the proposed situational cues.

Ontology-based adaptive testing for automated driving functions using data mining techniques

This paper presents an adaptive verification framework for automated driving functions based on ontologies and data-mining techniques. Despite the recent rapid growth of driver assistance systems to consolidate road safety, they still have various challenges coping with dynamic traffic situations of daily life. Therefore, automotive systems engineering has established data- and knowledge-driven test methods to assure the required functional safety and reliability in a highly safety-critical context. However, the reliance on field testing is inadequate and, in particular, time- and cost-intensive when applied to the next generation of automated driving functions, e.g. collision-free emergency braking and vehicle platooning. The presented framework utilises an ontology-based test scenario synthesis to identify criticality margins using a Hardware-in-the-Loop co-simulation platform for automated driving functions. Additionally, we demonstrate a systematic process to complement virtual testing by extracting insights from field testing database using event-based time-series analysis. To this end, data mining techniques are used to obtain representative scenarios witnessed in real-world traffic. Agglomerative hierarchical clustering is performed to extract homogeneous groups (clusters) from recorded triggering events by proximity metrics using normalised cross-correlations. Extracted scenarios are subsequently used at earlier stages of development to effectively and efficiently ensure reliability and safety. In summary, the results show the benefits and some of the challenges of using the industry-proven framework, which enables a cost-effective extension of test domain vaidility throughout software product engineering.

The efficacy of a brief hazard perception interventional program for child bicyclists to improve perceptive standards

IntroductionEven though child bicyclists are highly vulnerable in traffic only few studies focused on providing child bicyclists with means to enhance their abilities to deal with the complexity of dynamic traffic situations. The current study therefore evaluated whether a brief hazard perception intervention might be effective to improve hazard perception skills in child bicyclists towards a level more comparable to adult bicyclists. MethodsEighty children of the fourth grade (9.03 ± 0.43 years; 34 girls) and forty-six adults (34.67 ± 14.25 years age; 24 woman) first performed a Hazard Perception test for bicyclists. Response rate, reaction times, first fixation, duration of the first fixation, dwell time and total number of fixations on the events were measured. Next, the children took part in the HP intervention in which video clips of dangerous traffic situations were presented. The intervention comprised two classroom sessions of one hour (1/week). A post-test was performed one day after and the retention-test three weeks after the intervention. ResultsChildren responded to more covert hazards immediately after the intervention (p < 0.05), but did not improve their response rate for overt hazards. Reaction times for the covert hazards improved on the post-test (p < 0.001) compared to the pre-test but this effect was reduced on the retention test. There was no effect of the intervention for entry time of the first fixation but the duration of the first fixation increased for the covert hazards (p < 0.05). Children made fewer fixations on the event compared to adults (p < 0.001), except for the covert hazards on the retention-test. The training also increased the number of fixations for the overt hazards on the post-test (p < 0.001) and the retention-test (p < 0.001) but only increased on the retention test for the covert hazards (p < 0.001). ConclusionThe results demonstrated that a brief intervention for training hazard perception skills in child bicyclists is able to improve children’s situation awareness and hazard perception for potential dangerous situations. The training, however, was too short to improve children to higher adult levels.

A probabilistic model of pedestrian crossing behavior at signalized intersections for connected vehicles

Active safety systems which assess highly dynamic traffic situations including pedestrians are required with growing demands in autonomous driving and Connected Vehicles. In this paper, we focus on one of the most hazardous traffic situations: the possible collision between a pedestrian and a turning vehicle at signalized intersections. This paper presents a probabilistic model of pedestrian behavior to signalized crosswalks. In order to model the behavior of pedestrian, we take not only pedestrian physical states but also contextual information into account. We propose a model based on the Dynamic Bayesian Network which integrates relationships among the intersection context information and the pedestrian behavior in the same way as a human. The particle filter is used to estimate the pedestrian states, including position, crossing decision and motion type. Experimental evaluation using real traffic data shows that this model is able to recognize the pedestrian crossing decision in a few seconds from the traffic signal and pedestrian position information. This information is assumed to be obtained with the development of Connected Vehicle.

Toward Environmental-Adaptive Visible Light Communications Receivers for Automotive Applications: A Review

Since more and more vehicles are using the transportation infrastructure, the safety and the efficiency of roads are of absolute importance. A solution to enhance the security of road transportation is by enabling wireless communications between vehicles and also with the traffic infrastructures (e.g., traffic lights, street lighting, or traffic signs). In this area, visible light communications (VLC) is considered of great perspective. A fundamental problematic in vehicular VLC is the design of a proper sensor capable to support long distance communications in dynamic traffic situations and in unfriendly atmospheric conditions. Therefore, this paper is focused on the design of the VLC sensors intended for vehicular communication applications, offering a review of the solutions found to mitigate the effect of the problematic conditions. Furthermore, this paper summarizes these solutions and proposes an environmental adaptive VLC receiver that would be capable to optimally adjust its settings in order to maximize the communication efficiency, but without affecting the communication robustness to noise.

IMAGE BASED RECOGNITION OF DYNAMIC TRAFFIC SITUATIONS BY EVALUATING THE EXTERIOR SURROUNDING AND INTERIOR SPACE OF VEHICLES

Abstract. Today, cameras mounted in vehicles are used to observe the driver as well as the objects around a vehicle. In this article, an outline of a concept for image based recognition of dynamic traffic situations is shown. A dynamic traffic situation will be described by road users and their intentions. Images will be taken by a vehicle fleet and aggregated on a server. On these images, new strategies for machine learning will be applied iteratively when new data has arrived on the server. The results of the learning process will be models describing the traffic situation and will be transmitted back to the recording vehicles. The recognition will be performed as a standalone function in the vehicles and will use the received models. It can be expected, that this method can make the detection and classification of objects around the vehicles more reliable. In addition, the prediction of their actions for the next seconds should be possible. As one example how this concept is used, a method to recognize the illumination situation of a traffic scene is described. This allows to handle different appearances of objects depending on the illumination of the scene. Different illumination classes will be defined to distinguish different illumination situations. Intensity based features are extracted from the images and used by a classifier to assign an image to an illumination class. This method is being tested for a real data set of daytime and nighttime images. It can be shown, that the illumination class can be classified correctly for more than 80% of the images.

A Fast and Scalable Re-routing Algorithm based on Shortest Path and Genetic Algorithms J. Lee, J. Yang Jungkyu Lee

This paper presents a fast and scalable re-routing algorithm that adapts to dynamically changing networks. The proposed algorithm, DGA, integrates Dijkstra’s shortest path algorithm with the genetic algorithm. Dijkstra’s algorithm is used to define the predecessor array that facilitates the initialization process of the genetic algorithm. Then the genetic algorithm keeps finding the best routes with appropriate genetic operators under dynamic traffic situations. Experimental results demonstrate that DGA produces routes with less traveling time and computational overhead than pure genetic algorithm-based approaches as well as Dijkstra’s algorithm in largescale routing problems.

MARV-X: Applying Maneuver Assessment for Reliable Verification of Car-to-X Mobility Data

Advanced driver-assistance systems (ADASs) employ single-object information to provide safety, comfort, or infotainment features. While today's systems use common sensors, such as radars or cameras, to recognize and predict the future states of relevant traffic participants, next-generation ADASs will also use data from additional sources such as Car-to-X (C2X) communication networks. We present a method that uses information on other traffic participants and furthermore recognizes and considers their interactions in terms of traffic maneuvers. For this purpose, a probabilistic approach is presented, which identifies object interactions and different road characteristics. This method may find a particular application in the C2X domain for evaluating the mobility of neighboring vehicles based on received messages. In this paper, we present M aneuver Assessment For R eliable Verification of C2X mobility data (MARV-X), which is a tool embodying a two-stage process for reliable C2X mobility data verification. The first stage consists of a dedicated mobility estimator realized by a Kalman filter (KF). In the second stage, a plausibility check for highly dynamic traffic situations is applied using the advocated probabilistic traffic maneuver recognition. MARV-X is fully integrated into the vehicle's C2X architecture. Its effectiveness is demonstrated by means of extensive real-world experiments.

An Experimental Space for Conducting Controlled Driving Behavior Studies based on a Multiuser Networked 3D Virtual Environment and the Scenario Markup Language

We present a new framework for conducting controlled driving behavior studies based on multiuser networked 3-D virtual environments. The framework supports: 1) the simulation of multiuser immersive driving; 2) the visualization of surrounding traffic; 3) the specification and creation of reproducible traffic scenarios; and 4) the collection of meaningful driving behavior data. We use our framework to investigate the “rubbernecking” phenomenon, which refers to the slowing down of a driver due to an accident on the opposite side of the road, and its effect on the following drivers. The main contribution of the paper is the Scenario Markup Language (SML) framework, which is composed of: 1) the SML as a practical tool to specify dynamic traffic situations (e.g., an accident) and 2) the Scenario Control System to ensure the reproducibility of particular traffic situations, so that traffic engineers can obtain comparable data and draw valid conclusions. To demonstrate the effectiveness of our framework, we specified the traffic accident scenario in SML and conducted a study about the rubbernecking phenomenon. We report on the results of our study from two viewpoints: 1) the reproducibility of the traffic accident situation (i.e., state variables of interest are recreated successfully in 78% of the cases); and 2) the interactive car-following behavior of human subjects embedded in the traffic situation of the virtual environment.