Complex Traffic Environments Research Articles

Road Extraction from Remote Sensing Imagery with Spatial Attention Based on Swin Transformer

Road extraction is a crucial aspect of remote sensing imagery processing that plays a significant role in various remote sensing applications, including automatic driving, urban planning, and path navigation. However, accurate road extraction is a challenging task due to factors such as high road density, building occlusion, and complex traffic environments. In this study, a Spatial Attention Swin Transformer (SASwin Transformer) architecture is proposed to create a robust encoder capable of extracting roads from remote sensing imagery. In this architecture, we have developed a spatial self-attention (SSA) module that captures efficient and rich spatial information through spatial self-attention to reconstruct the feature map. Following this, the module performs residual connections with the input, which helps reduce interference from unrelated regions. Additionally, we designed a Spatial MLP (SMLP) module to aggregate spatial feature information from multiple branches while simultaneously reducing computational complexity. Two public road datasets, the Massachusetts dataset and the DeepGlobe dataset, were used for extensive experiments. The results show that our proposed model has an improved overall performance compared to several state-of-the-art algorithms. In particular, on the two datasets, our model outperforms D-LinkNet with an increase in Intersection over Union (IoU) metrics of 1.88% and 1.84%, respectively.

Game Theory-Based Interactive Control for Human–Machine Cooperative Driving

To address the safety and efficient driving issues of human–machine shared control vehicles (HSCVs) in future complex traffic environments, this paper proposes a game theory-based interactive control method between HSCVs and surrounding autonomous vehicles (SVs) and involves considering different driving behaviors. In HSCV, a comprehensive driver model integrating steering control and speed control is designed based on the brain emotional learning circuit model (BELCM), and the control authority between the driver and the automation system is dynamically allocated through the evaluation of the driving safety field. Factors such as driving safety and travel efficiency that reflect personalized driving style are considered for modeling the uncertain behavior of SVs. In the interaction between HSCVs and SVs, a method based on game theory and distributed model predictive control (DMPC) that considers the uncertainty of SVs’ driving behavior is established and is finally integrated into a multi-objective constraint problem. The driver control input in an HSCV will also be introduced into the solution process. To demonstrate the feasibility of the proposed method, two test scenarios considering the driving characteristics of different SVs are established. The test results show that automation control systems can promptly stop the human driver’s dangerous operations in an HSCV and safely interact with multiple AVs with different driving characteristics.

Energy-Efficient Speed Planning for Autonomous Driving in Dynamic Traffic Scenarios

In the field of autonomous driving, velocity planning is of paramount importance for handling dynamic obstacle scenarios. To avoid unnecessary acceleration and deceleration, self-driving vehicles need to find an energy-optimized velocity trajectory. Moreover, in complex traffic environments, the vehicle trajectory must consider the spatio-temporal coupling problem to avoid unrealistic driving paths. To address these challenges, this paper proposes a hierarchical planner that first plans the path and then performs speed planning based on the already planned path. Specifically, we focus on the energy consumption factor and use dynamic programming for speed planning while combining safety and comfort considerations. The optimal energy-saving trajectory is obtained by combining the speed profile with the optimal path. To cope with complex scenarios on real roads, we propose an adaptive trajectory adjustment strategy based on model predictive control to track by adaptively selecting tracking modes. Finally, hardware-in-the-loop experimental validation demonstrates that our proposed method significantly reduces energy consumption compared with the traditional decoupling method while ensuring that the autonomous vehicle adapts well to complex traffic scenarios.

MCFP: A multi-target 3D perception method with weak dependence on 2D detectors

3D perception for multi-target in complex traffic environments plays an important role in autonomous driving. Nowadays, some mainstream models use 2D detectors to provide auxiliary information for 3D perception. But it is a challenge that the accuracy of 2D detectors has a sensitive impact on the final results. In this paper, we propose MCFP, a multi-target 3D perception model with weak dependence on 2D detectors, to improve the 3D perception performance. We use the point cloud expansion as the main idea to provide more information about instance objects for model learning. Firstly, the centroid awareness strategy is integrated into the front-end part for downsampling and feature extraction to reduce the loss of object-related information. Secondly, the ball query operation with adaptive radius is designed to obtain more complete object information. Experiments show that our model outperforms baseline models in normal environments, and achieves detection performances close to those of SOTA models in extreme environments, which implies that our model has a strong ability to obtain semantic information and hence enhancing the perception ability. The code and data will be made available at https://github.com/MrSakasky/MCFP.

Deep Learning-Based Computer Vision Methods for Complex Traffic Environments Perception: A Review

Deep Learning-Based Computer Vision Methods for Complex Traffic Environments Perception: A Review

TC-GCN: Triple cross-attention and graph convolutional network for traffic forecasting

With the rapid development of urbanization, increasingly more data are being acquired by intelligent transportation systems (ITSs), which is of great significance for traffic flow forecasting. Efficient intelligent traffic management systems (TMSs) depend on accurately forecasting traffic flows as well as reasonably researching and judging traffic states. However, the current prediction models used in transportation forecasting tasks are traditional temporal- and spatial-dimensional prediction approaches. Especially when faced with road channelization, the numbers of lanes and lane types increase significantly, causing significant increases in the dimensionality and complexity of traffic data, and research models utilizing the cross-interaction relationships among multiple dimensions have not been considered. This has led to unsatisfactory prediction results in complex traffic environments. Therefore, it is very important to explore the exchange–correlation of complex dimensional data and the mining of hidden attributes. This paper presents a method for performing multidimensional cross-attention and spatiotemporal graph convolution. This method fully considers cross-information and constructs an attention cross-view between every pair of dimensions among the channel, time and space domains to model the cross-dimensional dependencies of traffic data. We innovatively propose a triple cross-attention and graph convolutional network (TC-GCN), which can achieve further improved traffic forecasting performance. The TC-GCN is verified on two real-world traffic datasets, namely, METR-LA and PEMS-BAY, and the experimental results are compared with those of multiple advanced baselines, showing that the proposed approach outperforms most of these baselines, which proves the effectiveness of the method proposed in this paper.

Super-Resolution GAN and Global Aware Object Detection System for Vehicle Detection in Complex Traffic Environments

Super-Resolution GAN and Global Aware Object Detection System for Vehicle Detection in Complex Traffic Environments

Intelligent control of self-driving vehicles based on adaptive sampling supervised actor-critic and human driving experience.

Due to the complexity of the driving environment and the dynamics of the behavior of traffic participants, self-driving in dense traffic flow is very challenging. Traditional methods usually rely on predefined rules, which are difficult to adapt to various driving scenarios. Deep reinforcement learning (DRL) shows advantages over rule-based methods in complex self-driving environments, demonstrating the great potential of intelligent decision-making. However, one of the problems of DRL is the inefficiency of exploration; typically, it requires a lot of trial and error to learn the optimal policy, which leads to its slow learning rate and makes it difficult for the agent to learn well-performing decision-making policies in self-driving scenarios. Inspired by the outstanding performance of supervised learning in classification tasks, we propose a self-driving intelligent control method that combines human driving experience and adaptive sampling supervised actor-critic algorithm. Unlike traditional DRL, we modified the learning process of the policy network by combining supervised learning and DRL and adding human driving experience to the learning samples to better guide the self-driving vehicle to learn the optimal policy through human driving experience and real-time human guidance. In addition, in order to make the agent learn more efficiently, we introduced real-time human guidance in its learning process, and an adaptive balanced sampling method was designed for improving the sampling performance. We also designed the reward function in detail for different evaluation indexes such as traffic efficiency, which further guides the agent to learn the self-driving intelligent control policy in a better way. The experimental results show that the method is able to control vehicles in complex traffic environments for self-driving tasks and exhibits better performance than other DRL methods.

Enhancing traffic signal control with composite deep intelligence

Traffic signal control has always been a hot topic in the field of intelligent transportation. With the increasing complexity of urban traffic conditions due to urbanization, how to develop effective scheduling strategies to adapt to the changing traffic demands has become a key problem in current intelligent transportation. In light of this, this paper focuses on the traffic signal control problem at intersections and proposes a composite intelligent traffic signal control model based on heterogeneous graph neural networks with dual attention mechanisms and deep reinforcement learning. For the first time, the model incorporates the dual attention mechanism in graph neural networks into the traffic signal control, integrating graph neural networks with deep reinforcement learning techniques and traffic intersection scenarios. This allows for the construction of traffic condition models and the scheduling control of traffic resources, catering to the perception and decision-making needs in complex traffic environments. Firstly, the graph relationship representation of intersection resources is established, constructing the graph information structure for traffic flow and signal states. Then, a heterogeneous graph neural network is designed, incorporating both node-level and semantic-level dual attention mechanisms to characterize the traffic state and explore the relationships, enabling the extraction of explicit and implicit information in traffic intersections. Lastly, a deep reinforcement learning algorithm that combines Double Deep Q-Network (DDQN) and Dueling DQN is implemented to improve the algorithm's generalization and execution efficiency, enhancing the adaptability and stability of traffic signal scheduling in complex environments. Simulation tests are conducted on the SUMO simulation platform using real-world application datasets. Compared to four other similar traffic control model, the proposed model demonstrates performance advantages of more than 13% in terms of average reward, average delay, queue length, and waiting time. This validates the effectiveness of the proposed model.

Research on energy-saving driving control of hydrogen fuel bus based on deep reinforcement learning in freeway ramp weaving area

In the intelligent network traffic scenario, the convenient acquisition of microscopic vehicle states and global traffic states can help solve the problems of vehicle driving and energy management in complex traffic environments. This paper proposes a new energy management method for the hydrogen fuel cell bus based on the double-layer deep deterministic policy gradient (DDPG). Combined with the SUMO simulation platform, a double-layer deep reinforcement learning (D-DRL) architecture based on DDPG is designed to improve control accuracy and training speed. In the upper D-DRL, the Agent handles the effects of complex traffic environments to control the reasonable speed of the vehicle and keep it running smoothly to reduce the loss of energy caused by speed changes, compared with the SUMO-IDM model, the maximum-minimum velocity difference was reduced by 21 % and the acceleration and acceleration change was reduced by 7.9 % and 19 %, After the Agent receives the output speed from the upper layer, it distributes the power between fuel cell and power cell. Compared with the DP algorithm, it keeps the SOC at a higher level, the hydrogen consumption level reaches 93.25 %, and the fluctuation amplitude decreases by 42.09 %, effectively improving fuel cell durability.

How Does Traffic Environment Quantitatively Affect the Autonomous Driving Prediction?

Accurate trajectory prediction is essential for safe and efficient autonomous driving in complex traffic environments. While artificial intelligence has shown great promise in improving prediction accuracy, its inherent uncertainty and lack of explainability may lead to unpredictable failures, creating challenges for safety-critical decision-making. This study aims to address these challenges by exploring the impact of traffic environment on prediction algorithms. The study proposes a trajectory prediction framework with epistemic uncertainty estimation ability to output high uncertainty when facing unforeseeable or unknown scenarios. The framework analyzes the environmental effect on the trajectory prediction by considering scenario features and shifts. Features are divided into kinematic features of a target agent, features of surrounding traffic participants, and other scenario features. Feature correlation and importance analyses are performed to study their influence on prediction error and epistemic uncertainty. The impact of unavoidable distributional shifts in the real world on trajectory predictions is investigated using multiple intersection datasets. The results indicate that deep ensemble-based methods have advantages in improving robustness while estimating epistemic uncertainty. Consistent conclusions were obtained from the correlation and importance analyses, indicating that kinematic features of the target agent have relatively strong effects on both prediction error and epistemic uncertainty. Finally, the study analyzes the accuracy deterioration caused by distributional shifts and the potential of the deep ensemble-based method. Through deep ensemble, the errors of the prediction methods based on GRIP++ and Trajectron++ have been improved by 6.4% and 10.8% in the same-dataset test, and 6.3% and 10.8% in the cross-dataset test.

Car-Following Models for Human-Driven Vehicles and Autonomous Vehicles: A Systematic Review

The focus of car-following models is to analyze the microscopic characteristics of traffic flows, with particular attention given to the interaction between adjacent vehicles. This paper presents a systematic review of existing studies on car-following models, with an emphasis on the behavior of both human-driven vehicles (HDVs) and autonomous vehicles (AVs). We classify car-following models based on their applicable background and structure. By considering driving behavior and specific model parameters, we identify the advantages and limitations of each microscopic simulation model in terms of accuracy and continuity. We also discuss model calibration methods, stability analysis, and the impact of complex traffic environments on the car-following process. Finally, we present detailed discussions of each model’s features and provide recommendations based on reviewed works and development trends for future research, including mixed traffic flow composed of AVs and HDVs.

Vehicle Trajectory Prediction Method Driven by Raw Sensing Data for Intelligent Vehicles

Vehicle trajectory prediction plays a vital role in intelligent driving modules and helps intelligent vehicles travel safely and efficiently in complex traffic environments. Several learning-based prediction methods have been developed that accurately identify vehicle behaviour patterns in actual driving data. However, these methods rely on manually curated structured data and are difficult to deploy in intelligent vehicles. In addition, modular information channels that perform vehicle detection, tracking, and prediction tasks encounter error propagation issues and insufficient computing resources. Therefore, this paper proposes a new multitask parallel joint framework in which vehicle detection, state assessment, tracking, and trajectory prediction are performed simultaneously according to raw LIDAR data. Specifically, a multiscale bird's eye view (BEV) backbone feature extraction model is proposed and combined with the designed vehicle state identification branch to distinguish dynamic and static vehicles, which is used as a strong prior for trajectory prediction. In addition, a spatiotemporal pyramid model with convolutions and a backbone residual network is used to generate high definition (HD) maps with strong constraints and guidance capabilities, thereby improving the trajectory prediction accuracy. The experimental results on the real-world dataset nuScenes show that the proposed multitask joint framework outperforms state-of-the-art vehicle detection and prediction schemes, including ES3D and PnPNet.

Vehicle detection and tracking using low-channel roadside LiDAR

Traffic object detection and tracking is one of the fundamental tasks in processing point cloud data to collect high-resolution vehicle trajectories using Roadside LiDAR sensors. Comprehensive, accurate, and complete traffic object tracking within the sensor’s scanning range still remains elusive, especially in complex traffic environments, mainly due to challenges like occlusion. To solve these issues, a novel vehicle detection and tracking method for low-channel roadside LiDAR in complex environments is proposed. Firstly, we use the L-shape fitting method to get a more accurate bounding box to extract the object features. Then, the decision tree with bagging algorithm is used to classify the traffic objects based on the selected features. Next, an improved Hungarian algorithm with the Kalman filter is used to predict the vehicle’s path considering the conditions of complete and partial occlusion. Finally, the effectiveness of the proposed framework is evaluated by comparing the roadside LiDAR data collected from four sites with the ground truth data obtained either from video-cameras or VeloView (software for point cloud data visualization). The result shows that the detection and tracking accuracy of the proposed method can reach up to 99.50% and 97%, respectively, which outperforms the state-of-the-art algorithms.

Collaborative Intelligence Enabled Routing in Green IoV: A Grid and Vehicle Density Prediction-Based Protocol

Green Internet of Vehicles (IoV) is a newly-emerged research area which focuses on reducing networking overhead and improving communication efficiency in future vehicular cyber-physical systems (VCPS). A critical requirement to realize such tasks in Green IoV is the achievement of effective routing protocols for data dissemination. However, maintaining end-to-end communication and reducing the communication overhead at the same time is quite challenging due to high vehicle dynamics and complex traffic environments. In this paper, we propose a vehicle density prediction-based routing protocol called VDPGrid. Firstly, we introduce a vehicle density prediction model according to the spatio-temporal features of vehicle trajectories. Then, to reduce the communication overhead and improve communication efficiency, we divide the map into grids and present a routing path evaluation scheme that jointly considers the vehicle density, link quality, and routing length. Moreover, a grid-based routing method is proposed to select the optimal relay node according to real-time traffic information. Finally, extensive experiments using real-world vehicle trajectories are conducted to validate the effectiveness of our method.

An object detection approach with residual feature fusion and second‐order term attention mechanism

AbstractAutomatically detecting and locating remote occlusion small objects from the images of complex traffic environments is a valuable and challenging research. Since the boundary box location is not sufficiently accurate and it is difficult to distinguish overlapping and occluded objects, the authors propose a network model with a second‐order term attention mechanism and occlusion loss. First, the backbone network is built on CSPDarkNet53. Then a method is designed for the feature extraction network based on an item‐wise attention mechanism, which uses the filtered weighted feature vector to replace the original residual fusion and adds a second‐order term to reduce the information loss in the process of fusion and accelerate the convergence of the model. Finally, an objected occlusion regression loss function is studied to reduce the problems of missed detections caused by dense objects. Sufficient experimental results demonstrate that the authors’ method achieved state‐of‐the‐art performance without reducing the detection speed. The mAP@.5 of the method is 85.8% on the Foggy_cityscapes dataset and the mAP@.5 of the method is 97.8% on the KITTI dataset.

PerspectiveNet: An Object Detection Method With Adaptive Perspective Box Network Based on Density-Aware

Aiming at the problems of object mutual occlusion and inaccurate confidence in complex traffic environments, we propose an adaptive network object detection algorithm with a perspective box based on density-aware. Firstly, an adaptive convolutional neural network is designed according to the complexity of image density. Res2Net-101 is used as the backbone. For the images with objects concentrated on both sides and occluded each other more, a high-density network is selected to improve the accuracy of object detection, and a low-density network is selected for the images with significant and simple objects. Secondly, in the high-density network, to improve the accuracy of object detection and reduce the missed detection problem caused by occlusion, a perspective box is added to the detection head, and the transparent box is fused with the prediction box to obtain more accurate positioning. Finally, we propose the perspective loss function, which is based on the repulsion loss of focusing smooth function and combined with feature loss and classification loss to form the overall loss of the model. The experimental results show that the model has a good detection effect compared with the state-of-the-art object detection model in complex traffic environments. Without reducing the detection speed, the mAP of PerspectiveNet on the KITTI dataset is 86.2%, which is 3.6% higher than that of VarifocalNet. On the Cityscapes dataset, the mAP is 96.3%, which is 1.7% higher than that of VarifocalNet.

Traffic Accident Detection Method Using Trajectory Tracking and Influence Maps

With the development of artificial intelligence, techniques such as machine learning, object detection, and trajectory tracking have been applied to various traffic fields to detect accidents and analyze their causes. However, detecting traffic accidents using closed-circuit television (CCTV) as an emerging subject in machine learning remains challenging because of complex traffic environments and limited vision. Traditional research has limitations in deducing the trajectories of accident-related objects and extracting the spatiotemporal relationships among objects. This paper proposes a traffic accident detection method that helps to determine whether each frame shows accidents by generating and considering object trajectories using influence maps and a convolutional neural network (CNN). The influence maps with spatiotemporal relationships were enhanced to improve the detection of traffic accidents. A CNN is utilized to extract latent representations from the influence maps produced by object trajectories. Car Accident Detection and Prediction (CADP) was utilized in the experiments to train our model, which achieved a traffic accident detection accuracy of approximately 95%. Thus, the proposed method attained remarkable results in terms of performance improvement compared to methods that only rely on CNN-based detection.

Modeling and Simulation of Crowd Pre-Evacuation Decision-Making in Complex Traffic Environments.

Human movements in complex traffic environments have been successfully simulated by various models. It is crucial to improve crowd safety and urban resilience. However, few studies focus on reproducing human behavior and predicting escape reaction time in the initial judgement stage in complex traffic environments. In this paper, a pedestrian pre-evacuation decision-making model considering pedestrian heterogeneity is proposed for complex environments. Firstly, the model takes different obvious factors into account, including cognition, information, experience, habits, stress, and decision-making ability. Then, according to the preference of the escapees, the personnel decision-making in each stage is divided into two types: stay and escape. Finally, multiple influencing factors are selected to construct the regression equation for prediction of the escape opportunity. The results show that: (1) Choices of escape opportunity are divided into several stages, which are affected by the pedestrian individual risk tolerance, risk categories strength, distance from danger, and reaction of the neighborhood crowd. (2) There are many important factors indicating the pedestrian individual risk tolerance, in which Gen, Group, Time and Mode are a positive correlation, while Age and Zone are a negative correlation. (3) The analysis of the natural response rate of different evacuation strategies shows that 19.81% of people evacuate immediately. The research in this paper can better protect public safety and promote the normal activities of the population.

Autonomous driving planning and decision making based on game theory and reinforcement learning

AbstractAutonomous driving technology is one of the important methods that avoid the hidden dangers of traffic safety. Although the existing autonomous driving technology can meet some needs of real traffic scenarios, the ability of multi‐agents to effectively generate autonomous driving strategy in complex traffic environments remains to be improved. Aiming at this problem, the automatic drive model based on game theory and reinforcement learning is proposed by combining these two technologies and applying them in multi‐agent cooperative driving, which enables multi‐agents to carry out strategic reasoning with negotiation in traffic scenarios by extending the game description language, and puts forward the constrained multi‐agents deep deterministic policies gradient algorithm. Finally, the related experiments are conducted through the autonomous driving simulation platform, and the experimental results show that the proposed model can effectively generate driving strategies for multi‐agents in complex traffic environments, which verifies the validity and feasibility of the proposed model, and provides the general research basis for multi‐agents autonomous driving.