Complex Road Environment Research Articles

Trajectory Tracking Control for Small Electric Sweeper Based on the Hybrid Control Method

Background: Trajectory tracking, one of the critical automatic driving technologies, is significant for the operation of small electric sweepers working in complex road environments, such as campuses and closed parks. Methods: This paper investigates a trajectory tracking control system for small self-driving electric sweepers on roads with different operating speed requirements. A hybrid control method combining sliding mode control and model predictive control is proposed, with sliding mode control operating under low-speed (<2 m/s) conditions, and model predictive control operating under high-speed (>2 m/s) conditions. An improved sliding mode controller based on the kinematic model is designed to reduce chattering, and a model predictive controller based on the dynamics model is constructed. Result: Finally, the performance of the hybrid control method is verified using the MATLAB/ Simulink platform. Conclusion: The simulation results prove the effectiveness of the hybrid control method and the good tracking performance of the electric sweeper at different driving speeds.

Asphalt pavement characterization by GPR using an air-coupled antenna array

Ground penetrating radar (GPR) has become an effective tool for asphalt pavement inspection. However, a ground-coupled GPR system cannot facilitate a high-speed survey due to the complex road environment and traffic condition. In this paper, an air-coupled antenna array is designed for measurement of common midpoint (CMP) GPR data. From the CMP data, an improved velocity spectrum algorithm considering the refraction on the pavement surface is proposed to simultaneously estimate the thickness and dielectric permittivity of the pavement layer. The results of a laboratory experiment demonstrate that the improved velocity spectrum method can greatly enhance the accuracy of the velocity and thickness estimations, compared with the traditional inversion method. A field measurement conducted on a highway pavement shows that the maximum error of the thickness estimation of the asphalt layer is less than 10 mm (7.1%). It is concluded that the developed CMP technique can be used for quantitative characterization of asphalt pavement.

Research on global path planning of unmanned vehicles based on improved ant colony algorithm in the complex road environment

When planning the path of a non-urbanized road, the default ant colony optimization (ACO) algorithm does not consider complex road state function such as uneven surface, road attachment coefficient, and vehicle turning angle limit. Based on the actual situation of roads and vehicles, a pavement state function that considers uneven areas such as road bumps and pavement attachment is proposed to improve the description of path length. Then, a heuristic function based on the A* algorithm and an improved mechanism for the initialization of pheromone distribution is proposed, which changes the blindness of ant colony search, accelerates the convergence of the ACO, and improves the search efficiency. The global search capability of the algorithm is enhanced by improving the path selection strategy and path transition probability function. The pheromone updating method is improved by using the MAX-MIN Ant System, which increases the algorithm diversity and avoids local optima. Further, using the pruning algorithm to reduce the number of paths significantly increases the convergence speed. Simulation results show that the improved ACO algorithm has better convergence speed and global search ability. Combining road state processing with vehicle corner processing can effectively improve the safety, adaptability, and reliability of autonomous vehicles. And the global optimal path planning of unmanned vehicles on complex roads can also be realized.

Hybrid Autonomous Driving Guidance Strategy Combining Deep Reinforcement Learning and Expert System

The complex traffic and road environment pose considerable challenges to the accuracy, timeliness, and adaptive ability of connected and autonomous vehicles (CAVs) in making driving decisions. This paper uses vehicle collaboration and integrates the adaptive learning capabilities of machine learning and the interpretation capabilities of expert systems (ESs) in a unified architecture to form a hybrid autonomous driving guidance system, which not only solves the “bottleneck” of knowledge acquisition during the construction of expert systems but also solves the “black box” phenomenon of machine learning in the decision-making process. First, an autonomous driving strategy based on deep reinforcement learning (DRL) is proposed for CAVs to make decisions and extract corresponding rules. Next, we design an ES knowledge base expansion method including rule extraction, rule sharing, and rule test. Particularly, vehicular blockchain is adopted to ensure user privacy and data security during the rule-sharing process. Third, hybrid autonomous driving guidance combining ES and machine learning is proposed for CAVs to make accurate and efficient decisions in different driving environments. Once the strategy is well trained, it can effectively guide CAVs to cope with the complex traffic environment. Extensive simulations validate the performance of our proposal in terms of decision-making accuracy, effectiveness, and safety.

Sensing fusion in vehicular network digital twins for 6G smart city

The aims demonstrated in this article are to effectively monitor the complex road environment in smart city transportation using the sixth generation mobile communication technology (6G) Digital Twins (DTs), to perceive the complex road environment of smart city traffic. Vehicular Networks (VN) in the smart transportation system have been selected as the research object, and the multi-sensor collaboration and fusion technology in the network is explored, so as to meet the active control requirements of intelligent vehicles. A lidar and camera fusion-based segmentation network C-LNet is proposed. The structure of a C-LNet multi-sensing data fusion segmentation network is double encoder-single decoder. Two encoders are used to extract image features and lidar features respectively. The same heterogeneous data is realized through the synchronization of lidar point cloud data and image data in sensor space. For multimodal information, a multiscale feature fusion-based vehicle collaboration method is designed. In the simulation experiment part, the C-LNet multi-sensing data fusion segmentation network is verified on the KITTI data set. The accuracy, F1 value, and MIoU of C-LNet are 98.4%, 96.7%, and 94.51%, respectively, which are better than those of an RGB network and lidar network. In summary, the smart transportation system supported by DTs in a 6G environment is explored. The proposed VN sensing fusion method can effectively realize the collaborative positioning perception of multiple vehicles, which lays the foundation for the realization of complex collaborative decision-making and control in smart transportation.

Review of Recent Automated Pothole-Detection Methods

Potholes, a kind of road defect, can damage vehicles and negatively affect drivers’ safe driving, and in severe cases can lead to traffic accidents. Efficient and preventive management of potholes in a complex road environment plays an important role in securing driver safety. It is also expected to contribute to the prevention of traffic accidents and the smooth flow of traffic. In the past, pothole detection was mainly performed via visual inspection by human experts. Recently, automated pothole-detection methods apply various technologies that converge basic technologies such as sensors and signal processing. The automated pothole-detection methods can be classified into three types according to the technology used in the pothole-recognition process: a vision-based method, a vibration-based method, and a 3D reconstruction-based method. In this paper, three methods are compared, and the strengths and weaknesses of each method are summarized. The detection process and technology proposed in the latest research related to automated pothole detection are described for each method. The development plans of future technology that is connected with those studies are also presented in this paper.

Lane-GAN: A Robust Lane Detection Network for Driver Assistance System in High Speed and Complex Road Conditions.

Lane detection is an important and challenging part of autonomous driver assistance systems and other advanced assistance systems. The presence of road potholes and obstacles, complex road environments (illumination, occlusion, etc.) are ubiquitous, will cause the blur of images, which is captured by the vision perception system in the lane detection task. To improve the lane detection accuracy of blurred images, a network (Lane-GAN) for lane line detection is proposed in the paper, which is robust to blurred images. First, real and complex blur kernels are simulated to construct a blurred image dataset, and the improved GAN network is used to reinforce the lane features of the blurred image, and finally the feature information is further enriched with a recurrent feature transfer aggregator. Extensive experimental results demonstrate that the proposed network can get robust detection results in complex environments, especially for blurred lane lines. Compared with the SOTA detector, the proposed detector achieves a larger gain. The proposed method can enhance the lane detail features of the blurred image, improving the detection accuracy of the blurred lane effectively, in the driver assistance system in high speed and complex road conditions.

Urban Traffic Congestion State Recognition Supporting Algorithm Research on Vehicle Wireless Positioning in Vehicle–Road Cooperative Environment

Vehicle–road cooperative technology applies wireless communication and a new generation of internet technology to urban traffic management, providing an effective way to solve urban traffic congestion and improve traffic efficiency. This article researches the vehicle wireless positioning fusion algorithm, suitable for the actual vehicle–road collaborative environment, which is an important step of urban traffic congestion state recognition. First, based on the error correction of existing wireless positioning algorithms, a weighting indicator considering distance and positioning compound errors is designed, and a vehicle wireless positioning fusion algorithm based on error weighting to eliminate line-of-sight (LOS) and non-line-of-sight (NLOS) error is proposed. Secondly, the wireless positioning fusion algorithm is verified based on accuracy evaluation indicators such as root mean square error (RMSE), Cramer Rao lower bound (CRLB), geometric differentiation of precision (GDOP), and cumulative distribution probability (CDP), and the sensitivity of the distance propagation model parameters to the positioning error is analyzed. The verification results show that the local vehicle wireless positioning fusion algorithm proposed in this article could be useful to locate vehicles in an actual vehicle–road collaborative environment. The positioning accuracy could reach 46.31 m with 67% probability, while the positioning accuracy could reach 122.53 m with 95% probability. The average positioning accuracy could reach 39.97 m. Compared with the two types of wireless positioning methods based on ranging and non-ranging methods, the positioning accuracy is improved by 7.74% and 17.69%. The algorithm can either use the roadside base stations to carry out the individual vehicle positioning or cooperate with GPS positioning and trilateral positioning to make up for the positioning blind spots caused by the lack of signal, interference, or base station overload in the urban complex road environment and, furthermore, improves the robustness of vehicle positioning. The results could assist in all-day real-time traffic congestion state recognition and other actual scenarios.

Data Fusion Model of Road Sensors based IoT Feature Clustering

The collection of traffic data can play a role in analyzing and predicting highway design, planning, and real-time traffic management. The accuracy requirements for road dynamic data collection are low, and the accuracy is usually 3%-5%. However, it is required that vehicles can pass at high speed and obtain traffic information such as vehicle classification and vehicle speed. The prerequisite for the application of Internet of Things (IoT) technology to road information monitoring lies in the research and development of sensor technology in the perception layer and communication technology in the network layer, so that can obtain a large amount of perception data to serve the development and application of algorithms. To achieve the goal of low-cost and long-term monitoring of comprehensive traffic information and road service status information, this paper constructs a road vibration monitoring system, carries out road vibration monitoring under complex road environments, and proposes a traffic information monitoring method driven by road vibration data. By deploying the pavement vibration monitoring system in the actual road, the original signal of pavement vibration under the action of vehicle moving load is obtained. Through smooth processing and eigenvalue extraction, the monitoring of vehicle speed, wheelbase driving direction, vehicle load position and traffic flow is realized. The experimental results prove that the analysis of the road dynamic response under working conditions, as well as smoothing processing and eigenvalue extraction, the numerical modeling method in this paper realizes the monitoring of the position of the vehicle load and the traffic flow. The calculation error of vehicle speed and wheelbase is within ±4%, which is helpful to find the characteristic index of road vibration signal for evaluating road service status, and provides a reference for the application of road vibration response in road damage early warning and scientific maintenance.

The Value and Path of the Application of Artificial Intelligence Technology in Automatic Guided Vehicles from the Perspective of Commercialization

Intelligence is an important direction for the transformation and development of modern logistics systems. Automatic guided vehicles are the key equipment for the intelligent operation of modern logistics systems. In order to improve the commercialization capability of automatic guided vehicles, the functions of automatic guided vehicles must be continuously improved, so that automatic guided vehicles can adapt to more complex road environments and be applied to more application scenarios, so as to make the commercialization process of automatic guided vehicles more smoothly. Based on the perspective of the commercial operation of automatic guided vehicles, this paper uses the relevant theories of management science and computer technology to discuss the value and path of applying artificial intelligence technology to automatic guided vehicles from the perspective of commercialization. The conclusion is that artificial intelligence technology can enrich the function of the automatic guided vehicle, the optimization of the mechanical design of the automatic guided vehicle, and the improvement of the performance of the automatic guided vehicle, thereby expanding the application scenarios of the automatic guided vehicle, reducing the manufacturing cost of the automatic guided vehicle, and reducing the maintenance cost of the automatic guided vehicle. The path of applying artificial intelligence technology to automatic guided vehicles mainly includes the optimization of six fields including motion module, communication module, anti-collision module, power supply module, vehicle-mounted module, and AGVS monitoring and dispatching management system.

Lane Line Detection Based on Improved Semantic Segmentation in Complex Road Environment

Lane Line Detection Based on Improved Semantic Segmentation in Complex Road Environment

A continuum model considering the uncertain velocity of preceding vehicles on gradient highways

Gradient roads are not uncommon in mountainous areas and freeways. The driving field of vision will be limited under such a complex road environment, which hinders the drivers from accurately perceiving the speed of the preceding vehicle. Accurate modeling of traffic flow in such an environment is therefore paramount to make accurate predictions and to effectively control the system. To this end, we present a modified continuum model accounting for the uncertain velocity of preceding vehicles on gradient highways. We derive the stability criterion and the KdV–Burgers equation of the proposed model via linear and nonlinear stability analysis. The density wave solution obtained by solving the above KdV–Burgers equation can explain the propagation mechanism of traffic jams near the stability curve. Numerical examples reveal that the slope information and uncertainty term exert great influence on traffic jams and energy consumption. Specifically, the effect of the slope information is positive, whereas the effect of the uncertainty term is negative.

Self-delayed feedback car-following control with the velocity uncertainty of preceding vehicles on gradient roads

Uphill and downhill roads are prevalent in mountainous areas and freeways. Despite the advancements of vehicle-to-vehicle (V2V) communication technology, the driving field of vision could be still largely limited under such a complex road environment, which hinders the sensors from accurately perceiving the speed of front vehicles. As such, a fundamental question for autonomous traffic management is how to control traffic flow associated with the velocity uncertainty of preceding vehicles? This paper aims to answer this question by devising a cooperative control method for autonomous traffic to stabilize the traffic flow under such a complex road environment. To this end, this paper first develops a traffic flow model accounting for the uncertainty of preceding vehicle’s velocity on gradient roads and further devises a new self-delayed feedback controller based on the velocity and headway differences between the current time step and historical time step. The sufficient condition where traffic jams do not occur is derived from the perspective of the frequency domain via Hurwitz criteria and \(H_{\infty }\) norm of transfer functions. The Bode diagram reveals that the robustness of the closed-loop traffic flow model can be significantly enhanced. Simulation results show that the key parameters (control gain coefficient and delay time) of the designed controller contribute to the stability of traffic flow, which is consistent with the theoretical analysis conclusion.

YOLOv5s-FCG : An Improved YOLOv5 Method for Inspecting Riders’ Helmet Wearing

Manual inspection of riders’ helmets is time-consuming and labor-intensive, low in efficiency and small in coverage. Aiming at this shortcoming, this paper proposed an improved YOLOv5s-FCG(FourLayers, CBAM attention, GhostBottleneck) helmet wearing detection method based on YOLOv5 (You Only Look Once). Based on the smallest volume of YOLOv5s in YOLOv5 series, the network was improved, the shallow feature detection layer was added, the three-scale feature detection was changed to four-scale feature detection, and the up sampling was increased by four times. Add the CBAM attention module; Use lightweight GhostBottleneck instead of Bottleneck structures. The results in our experiments show that YOLOv5s-FCG raises the average detection accuracy (mAP) by 2.0% compared with YOLOv5s on the ourselves’ riding safety helmet data set and 1.5% on the NWPU-VHR-10 public data set. The proposed algorithm not only ensured the detection rate, volume, computation and number of parameters, but also improved the detection accuracy. And it had good adaptability and generalization ability in complex road environments such as poor light and small targets.

Multi-scale spatial convolution algorithm for lane line detection and lane offset estimation in complex road conditions

Deep learning has made remarkable progress in the field of image classification and object detection. Nevertheless, in the autonomous driving research, the real-time lane line detection and lane offset estimation in complex traffic scenes have always been challenging and difficult tasks. Traditional detection methods need manual adjustment of parameters, they face many problems and difficulties and are still highly susceptible to interference caused by obstructing objects, illumination changes, and pavement wear. It is still challenging to design a robust lane detection and lane offset estimation algorithm. In this paper, we propose a convolutional neural network for lane offset estimation and lane line detection in a complex road environment, which transforms the problems of lane line detection into the instance’s segmentation. In response to a change in the method of lane processing, the network will form its example to each line. The global scale perception optimization mechanism is designed to solve the issue, especially where the lane line width is gradually narrowing at the vanishing point of the lane. At the same time, to realize multi-tasking processing and improve performance, and end-to-end lane offset estimation network is used in addition to the lane line detection network.

SaccadeFork: A lightweight multi-sensor fusion-based target detector

Commercialization of self-driving applications requires precision and reliability of the perception system due to the highly dynamic and complex road environment. Early perception systems either rely on the camera or on LiDAR for moving obstacle detection. With the development of vehicular sensors and deep learning technologies, the multi-view and sensor fusion based convolutional neural network (CNN) model for detection tasks has become a popular research area. In this paper, we present a novel multi-sensor fusion-based CNN model–SaccadeFork–that integrates the image and upsampled LiDAR point clouds as the input. SaccadeFork includes two modules: (1) a lightweight backbone that consists of hourglass convolution feature extraction module and a parallel dilation convolution module for adaptation of the system to different target sizes; (2) an anchor-based detection head. The model also considers deployment of resource-limited edge devices in the vehicle. Two refinement strategies, i.e., Mixup and Swish activation function are also adopted to improve the model. Comparison with a series of latest models on public dataset of KITTI shows that SaccadeFork can achieve the optimal detection accuracy on vehicles and pedestrians under different scenarios. The final model is also deployed and tested on a local dataset collected based on edge devices and low-cost sensor solutions, and the results show that the model can achieve real-time efficiency and high detection accuracy.

Human-Machine Cooperation and Path Planning for Complex Road Conditions

With the rapid development of the information age, the development of industrial robots is also advancing by leaps and bounds. In the scenes of automobile, medicine, aerospace, and public service, we have fully enjoyed the convenience brought by industrial robots. However, with the continuous development of industrial robot-related concepts and technologies, human-computer interaction and cooperation have become the development trend of industrial robot. In this paper, the human-machine cooperation and path optimization of industrial robot in a complex road environment are studied and analyzed. At the theoretical modeling level, firstly, the industrial robot is modeled and obstacle avoidance is analyzed based on the kinematics of industrial robot; thus, an efficient and concise collision detection model of industrial robot is proposed. At the algorithm level, in view of the complex road conditions faced by industrial robots, this paper will study and analyze the obstacle avoidance strategy of human-computer cooperation and real-time path optimization algorithm of industrial robots. Based on the virtual target point algorithm, this paper further improves the problem that the goal of the traditional path planning algorithm cannot be fully covered, so as to propose the corresponding improved path planning algorithm of industrial robots. In the experimental part, based on the existing industrial robot system, the human-machine cooperation and path planning system proposed in this paper are designed. The experimental results show that the algorithm proposed in this paper improves the accuracy of obstacle avoidance by about 10 points and the corresponding convergence speed by about 5% compared with the traditional algorithm and the experimental effect is remarkable.

Seatbelt detection in road surveillance images based on improved dense residual network with two-level attention mechanism

Seatbelt detection is an important topic in intelligent transport systems. The accuracy of seatbelt detection in traffic video surveillance is affected by many factors, such as complex road environments, lighting, weather, direction of the camera, etc., which bring difficulties for traditional image processing methods. We propose a seatbelt detection method based on classification using convolutional neural networks. Dense residual blocks are used to avoid gradient dispersion and information loss during the network training. A two-level attention mechanism is also introduced to further improve the network performance by simultaneously modeling the correlation among different feature channels using channel attention mechanism, and correlation among different pixels within the feature map using pixel attention mechanism. To make the whole system complete, we added the driver area localization module, which is accomplished using the YOLOV3 model. Small images concentrated on the driver location, which are obtained through the localization operation from larger images, are passed into the classification network for seatbelt detection. We compared the proposed method with several other methods. Comparative results show that the proposed method has higher accuracy and is more robust for seatbelt detection for complex environments.

Influence of prompt timing and messages of an audio navigation system on driver behavior on an urban expressway with five exits

An audio navigation system is a very useful tool for driving path guidance with less distraction. However, the influence of the audio navigation system on driver behavior, especially in complicated road environments, is still not entirely clear. This study aims to investigate navigation prompt timing (NPT), navigation prompt message (NPM), and their combination in an audio navigation system on driving behavior on an urban expressway with five exits. Driving simulator technology was used to reproduce the experimental environment and obtain driver behavior data. Four indicators-speed, standard deviation of speed, absolute values of acceleration, and depth of accelerator-were selected to examine the main and interactive effects of NPT, NPM, and their combination based on repeated measures analysis of variance. The results show that the driver’s psychological state and operation of the vehicle on the urban expressway were affected by the prompt timing and messages of the audio navigation system. An interaction effect existed between prompt timing and prompt messages of the system, and this effect adjusted the effect on the driver’s psychological state and vehicle operation caused by these two important factors. This study contributes to our understanding of the influence of audio navigation systems on driver behavior in complex road environments and thus lays a foundation for developing standard audio navigation broadcast guidelines to improve drivers’ acceptance of navigation systems, reduce drivers’ cognitive workload, and improve the level of vehicle operation safety.

Effects of Vehicle Type and Inter-Vehicle Distance on Aerodynamic Characteristics during Vehicle Platooning

Considering the future development in vehicle platooning technology and the multiple models pertaining to complex road environments involving freight cars and general vehicles, the speed and distance of a vehicle model were set as variables in this study. This study aimed at analyzing the effect of currents acting differently using SolidWorks Flow Simulation tool for the vehicle platooning between different models of trucks that are currently being studied actively and sports utility vehicle (SUV) whose market share has been increasing, in order to evaluate the changes in the drag coefficient and their causes. Additionally, purpose-based vehicle (PBV) presented by Hyundai Motor (Ulsan, Korea) during the CES 2020 was considered. In this study, we found that the shape of the rear side of the leading vehicle reduces the drag coefficient of the following vehicle by washing the wake, similar to a spoiler at the rear. The rear side area of the leading vehicle forms a wide range of low pressures, which increases the drag coefficient effect of the following vehicle. The overall height of the leading vehicle also generates a distribution of low pressures above the rear of the vehicle. This reduces the impact of low pressures on the overall height of the following vehicle. The shape of the front of the following vehicle enables the wake of the leading vehicle, which involves low pressures, to inhibit the Bernoulli effect of the following vehicle. Furthermore, the front of the following vehicle continues to be affected by the wake of the leading vehicle, resulting in an increase in the drag coefficient reduction.