Real-time Road Research Articles

ITSMEI: An intelligent transport system for monitoring traffic and event information

The disorderly growth of urban centers can lead to serious socioeconomic disadvantages, such as health problems, due to long-term exposure to toxic gases and also monetary losses due to time stopped in congestion. Thus, there is a need for systems that help in the management and control of the flow of vehicles on the roads, seeking to reduce the damage resulting from a faulty transportation system and also avoiding the use of an inefficient system of information dissemination of urban roads. In this scenario, innovative systems are being developed to analyze the conjunction of road conditions to supervise and provide routes as needed for drivers to provide greater comfort and safety to vehicle traffic on urban roads. Thus, in this work, we propose the development of a system to monitor vehicle traffic, informing about events that are taking place on the roads in real time. The system can recommend new routes to drivers or allow drivers to take action based on information received from a particular road. As well as, the system uses driver location information for traffic monitoring, which will later be available for any devices, either a mobile device (smartphone) or a desktop. For the evaluation of the proposed system, a user case was developed for the Catanduva city in which we performed a test with the proposed system and was possible to verify a reduction in vehicle stopping time by 42% and a shorter travel time of 50% with an average speed of 33 km/h.

Analysis of Emergency Medical Vulnerability and Survival Rates Following Real-Time Traffic Information

Providing rapid access to emergency medical services (EMS) within the “golden time” for survival is important to improve the survival rate of emergency patients. This study analyzes the accessibility of EMS based on driving speed changes following real-time road traffic conditions by time to estimate vulnerable areas for EMS and survival rates of emergency patients. The key results of the network analysis based on real-time road speed and this evaluation of vulnerable areas by village level across South Korea reveal the different characteristics of urban and rural areas to access emergency medical facilities. In urban areas, road traffic congestion during rush hour delays the patients’ access time to EMS. In contrast, in rural areas, the long geographical distance to an emergency medical facility is a hurdle for receiving care from an EMS during the “golden time” because emergency medical facilities are mostly located in urban areas. The existing standard to assess vulnerable areas of EMS accessibility is based on the speed limit of roads, but the time may be underestimated because the speed limit alone does not reflect the real road conditions. The study results show that an effective way to increase the survival rate is receiving immediate first aid treatment, which means that the government should continuously train the public to perform cardiopulmonary resuscitation (CPR) as well as install automated external defibrillators (AEDs) in populated places, and train the public to use them. Reducing assess time to emergency medical centers in urban areas and providing additional manpower to help with first aid in rural areas are reasonable ways to improve the survival rate of emergency patients.

Monitoring Vehicle Communication and Road Condition in VANET

The connected vehicular ad-hoc network (VANET) and cloud computing technology allows entities in VANET to enjoy the advantageous storage and computing services offered by some cloud service provider. However, the advantages do not come free since their combination brings many new security and privacy requirements for VANET applications. In this article, we investigate the cloud-based road condition monitoring (RCoM) scenario, where the authority needs to monitor real-time road conditions with the help of a cloud server so that it could make sound responses to emergency cases timely. When some bad road condition is detected, e.g., some geologic hazard or accident happens, vehicles on site are able to report such information to a cloud server engaged by the authority. We focus on addressing three key issues in RCoM. First, the vehicles have to be authorized by some roadside unit before generating a road condition report in the domain and uploading it to the cloud server. Second, to guarantee the privacy against the cloud server, the road condition information should be reported in ciphertext format, which requires that the cloud server should be able to distinguish the reported data from different vehicles in ciphertext format for the same place without compromising their confidentiality. Third, the cloud server and authority should be able to validate the report source, i.e., to check whether the road conditions are reported by legitimate vehicles. To address these issues, we present an efficient RCoM scheme, analyze its efficiency theoretically, and demonstrate the practicality through experiments

Toward real-time road detection for autonomous vehicles

Road detection is a vital task for autonomous vehicles, as it has a direct link to passengers’ safety. Given its importance, researchers aimed to improve its accuracy and robustness. We look at the task from a holistic point of view, where we aim to balance computation and accuracy. A multimodal road detection pipeline is proposed, which fuses the camera image with the preprocessed LIDAR input. First, the LIDAR input is preprocessed using three-dimensional models inspired from computer graphics to generate image-like representations. Then, the preprocessed LIDAR input is combined with the camera image using a fusion module named inputs cross-fusion module, to reduce the computation amount required by other fusion strategies. To prevent the accuracy loss caused by the computation gain, we introduce the surface normal information to add distinctiveness. Furthermore, we propose a cost/benefit metric to evaluate the trade-off between computation cost and accuracy of road detection approaches. Several tests were conducted using the KITTI road detection benchmark based on deep convolutional neural networks, the obtained results were considered very satisfactory. In particular, the robustness of the proposed approach resulted in accuracies higher than 95% on different road types, comparable to those of the state-of-the-art techniques. In addition to marginally reducing the inference time of the used DCNN on images with a resolution of 1248 × 352 pixels to 130 ms using an NVIDIA GTX-1080TI.

Design and performance of a cantilever piezoelectric power generation device for real-time road safety warnings

This paper presents a batch-processable cantilever piezoelectric device suitable for road traffic conditions, and the power it converts can provide a real-time safety warning for automobiles in hazardous areas. Based on road traffic characteristics, environment characteristics and construction convenience, a multilayer vertical butterfly array piezoelectric device with high sensitivity is designed. Furthermore, to verify the reliability of the device for the early warning in the laboratory stage, the power generation rules under different traffic conditions are studied systematically, and the applicable situations in different combinations of automobile types and speeds are clarified. Finally, the effects of instantaneous safety warning are verified, and the installation measure of device is determined. The results indicate that the power effect of the device first increases and then decreases with the increase in the speed (Corresponding to the vibration frequency generated by the axles of automobile). The power effect of a single device is optimal when simulating a light traffic conventional speed of 50–90 km/h (7–11 Hz), the maximum electrical output is 26 V-22.09 mW; while the power under heavy traffic is relatively poor, which is 20.8 V-7.84 mW. Moreover, the flashing effect of warning sign powered by a single device is better under the speeds, and the devices will be installed in front of the hidden section in piece-by-piece. The clean electric energy converted by the piezoelectric device will be helpful for real-time road safety early warning, reduce power loss for long-distance transmission of power grids, and promote the process of intelligent transportation applications.

Adaptive road detection method combining lane line and obstacle boundary

In order to realise autonomous navigation of unmanned platforms in urban or off-road environments, it is crucial to study accurate, versatile and real-time road detection methods. This study proposes an adaptive road detection method that combines lane lines and obstacle boundaries, which can be applied to a variety of driving environments. Combining multi-channel threshold processing, it is robust to lane feature detection under various complex situations. Obstacle information extracted from the grid image constructed by 3D LIDAR point cloud is used for lane feature selection to avoid interference from pedestrians and vehicles. The proposed method makes use of adaptive sliding window for feature selection, and piecewise least squares method for road line fitting. Experimental results on dataset and in real-world environments show that the proposed method can overcome illumination changes, shadow occlusion, pedestrian, vehicle interference and so on in a variety of scenes. The proposed method has good enough efficiency, robustness and real-time performance.

Model Reference Adaptive Control of Vehicle Slip Ratio Based on Speed Tracking

This paper investigates a slip ratio control problem of a vehicle braking system in which real-time road conditions are unavailable. To overcome the nonlinearity and uncertainties, the slip ratio control problem is converted to a speed tracking control problem. By introducing a road-condition observer, a model reference adaptive control (MRAC) method is proposed to overcome modeling errors and system uncertainties. The road condition observer based on the Lyapunov theory is designed and a sliding mode observer is also constructed to estimate the unknown vehicle velocity. The asymptotic and robustness property of the adaptive controller is derived by the Popov hyperstability theory. Simulations under different conditions including different road adhesion coefficient, system uncertainties and modeling errors are provided to illustrate the effectiveness and robustness of the proposed method.

A Road Monitoring Approach with Real-Time Capturing of Events for Efficient Vehicles Safety in Smart City

Nowadays, Internet-of-things (IoT) provides various services to the users by equipped with the smart devices. In this regard, the next generation of vehicles collaborates with the features of IoT to provide the safety and security on the roads. To achieve this, it equipped with short-range communication advances and establishes Vehicle-to-Vehicle (V2V) connectivity. The standardized V2V connectivity and communication are termed as IEEE 802.11p. Later, an alternative named as (LTE-V2V) has been introduced. However, both the technologies are only concerned with the continuous broadcast of information and cooperative awareness. It only takes information from one vehicle in the format of text message and sends to another. In this regard, an efficient and satisfactory safety is not provided by these technologies for analysis of real-time road traffic monitoring. A clear observation of road conditions and awareness of event triggering is a necessary task rather than broadcasting of information provided by the vehicles. Therefore in this paper, we proposed a solution by providing the real-time information about road conditions and traffic scenarios to the users. We utilized the capturing images of road conditions by the positioned cameras and Global Positioning System in order to extract the information regarding vehicle and camera position. The proposed work provides a better security rather than message passing system in V2V communication. The users in our anticipated scenarios can extract and see a clear view of road conditions by use of captured videos/images. We offered an approach named as Smart Road Monitoring for IoT based smart cities to provide an effective solution for the real-time analysis of traffic road conditions. The proposed approach gives a clear view of conditions of the road to the users such that they can make a decision according to their suitability Our proposed solution well copes with the moderate traffic conditions and provides high satisfaction score. The performance analysis is done in a comparison of V2V and LTE-V2V in terms of various metrics. The simulation results show that our proposed work can achieve high performance in the provision of providing safety compared to other scheme introduced in this field.

Embedded Real-Time System for Traffic Sign Recognition on ARM Processor

This article proposes the design of a novel hardware embedded system used for automatic real-time road sign recognition. The algorithm used was implemented in two main steps. The first step, which detects the road signs, is performed by the maximally stable extremal region method on HSV color space. The second step enables the recognition of the detected signs by using the oriented fast and rotated brief features method. The novelty of the embedded hardware system, on an ARM processor, leads to a real-time implementation of the ADAS applications. The proposed system was tested on the Belgium Traffic Sign Detection and Recognition Benchmark and on the German Traffic Signs Datasets. The proposed approach attained a high detection and recognition rate with real-world situations. The achieved results are acceptable when compared to state-of-the-art systems.

Visualizing Potential Transportation Demand From ETC Log Analysis Using ELK Stack

Traffic conditions are among the issues most concerned with the general public, and the freeway is a large-scale Internet-of-Things application. In addition to obtaining real-time road usage information, analysis of local road usage habits is crucial in evaluations of government policy implementation. Using road usage data provided by the electronic toll collection (ETC) system, we investigated the data on road usage history on the freeways. The ELK stack was employed to construct a platform for visualizing real-time road usage information and history in this article; the platform is named the local transportation knowledge (LTK) platform. By analyzing more than 500 million pieces of data, the LTK platform proposed in this article efficiently visualized road usage data and facilitated the acquirement of local road usage knowledge. We verified that residents of other counties and cities commuted to Taichung each day. We also discovered that a considerable number of Taichung city residents were employed in Hsinchu Science Park and commuted between the two places. The LTK platform can present real-time freeway traffic conditions, facilitate in-depth analysis of local road usage data, and provide data to verify the relevant information.

Sparse Bayesian Learning Assisted Approaches for Road Network Traffic State Estimation

Real-time (online) road network traffic state estimation plays a vital role in enhancing the services offered by Intelligent Transportation Systems (ITS). Spatio-temporal data vacancies of contemporary acquisition systems considerably limit the reliability of online traffic state estimation. Online estimation insists smaller temporal frame widths inhibiting the accuracies of low rank matrix reconstruction approaches, while matrix imputation methods do not capture the non-trivial traffic data relationships. Alternatively, this paper investigates traffic state estimation by constructing sparse representations via Sparse Bayesian Learning (SBL) and Block SBL (BSBL) approaches to accommodate under-sampled data, independent of the acquisition type. Appropriate kernel matrices are determined by leveraging historical spatio-temporal correlations among the road network traffic data. Subsequently, unavailable traffic states are estimated from predictive distributions. The estimates are further pruned by kalman filter that corroborates online processing. With the SBL approach, experiments on PeMS traffic data demonstrate less than 6% Normalized Mean Absolute Error (NMAE) for a Signal Integrity (SI) of 0.5. Compared to the state-of-the-art approaches, this value is significantly better. BSBL approach gives similar error performance as SBL despite a SI of 0.26, at the cost of increased computational time. The NMAE from kalman filtered SBL (SBL+K) approach is less than 3.5% in contrast to 4.5% from kalman filtered BSBL (KBSBL) approach, thus demonstrating SBL+K approach as a good compromise between NMAE and computational time, facilitating more accurate online traffic state estimation.

Inverse algorithm for real-time road roughness estimation for autonomous vehicles

Human drivers take instant decisions about their speed, acceleration and distance from other vehicles based on different factors including their estimate of the road roughness. Having an accurate algorithm for real-time evaluation of road roughness can be critical for autonomous vehicles in order to achieve safe driving and passengers comfort. In this paper, we investigate the problem of interactive road roughness identification. We propose a novel inverse algorithm based on the knowledge of a vehicle dynamic characteristics and dynamic responses. The algorithm construct the road profile in time using one-iteration to update the wheels forces which are then used to identify the road roughness. The relation between the forces and the road profile is defined by a system of ordinary differential equations that are solved using the composite Gaussian quadrature. To reduce the error accumulation in time when noisy data is used for the vehicle response, a bidirectional filter is also implemented. We assume a simple model that is based on four degrees-of-freedom system and vibration acceleration measurements to evaluate the road roughness in real time. Although we present the results for this specific model, the algorithm can also be utilised with models of any number of degrees of freedom and can deal with models where the dynamic response is only available at some of the degrees of freedom. This is achieved by introducing a matrix reduction technique that is discussed in details. Furthermore, we evaluate the impact of uncertainty in the vehicle parameters on the algorithm estimation accuracy. The proposed algorithm is evaluated for different types of road roughness. The simulation results show that the proposed method is robust and can achieve high accuracy. The algorithm offers excellent potential for road roughness estimation not only for autonomous vehicle but also for vehicles and roads designing purposes.

Cartographer SLAM Method for Optimization with an Adaptive Multi-Distance Scan Scheduler

This paper presents the use of Google’s simultaneous localization and mapping (SLAM) technique, namely Cartographer, and adaptive multistage distance scheduler (AMDS) to improve the processing speed. This approach optimizes the processing speed of SLAM which is known to have performance degradation as the map grows due to a larger scan matcher. In this proposed work, the adaptive method was successfully tested in an actual vehicle to map roads in real time. The AMDS performs a local pose correction by controlling the LiDAR sensor scan range and scan matcher search window with the help of scheduling algorithms. The scheduling algorithms manage the SLAM that swaps between short and long distances during map data collection. As a result, the algorithms efficiently improved performance speed similar to short distance LiDAR scans while maintaining the accuracy of the full distance of LiDAR. By swapping the scan distance of the sensor, and adaptively limiting the search size of the scan matcher to handle difference scan sizes, the pose’s generation performance time is improved by approximately 16% as compared with a fixed scan distance, while maintaining similar accuracy.

Moving Target Detection and Tracking Based on Improved FCM Algorithm

With the rapid development of computer intelligence technology, the majority of scholars have a great interest in the detection and tracking of moving targets in the field of video surveillance and have been involved in its research. Moving target detection and tracking has also been widely used in military, industrial control, and intelligent transportation. With the rapid progress of the social economy, the supervision of traffic has become more and more complicated. How to detect the vehicles on the road in real time, monitor the illegal vehicles, and control the illegal vehicles effectively has become a hot issue. In view of the complex situation of moving vehicles in various traffic videos, the authors propose an improved algorithm for effective detection and tracking of moving vehicles, namely improved FCM algorithm. It combines traditional FCM algorithm with genetic algorithm and Kalman filter algorithm to track and detect moving targets. Experiments show that this improved clustering algorithm has certain advantages over other clustering algorithms.

Real Time Road Blocker Detection and Distance Calculation for Autonomous Vehicle Based on Camera Vision

Real Time Road Blocker Detection and Distance Calculation for Autonomous Vehicle Based on Camera Vision

面向无人机飞控平台的实时道路目标深度神经网络检测方法

面向无人机飞控平台的实时道路目标深度神经网络检测方法

EO-CNN: An Enhanced CNN Model Trained by Equilibrium Optimization for Traffic Transportation Prediction

Abstract Time-ordered data are widely available in many real-life areas like traffic transportation, economic growth, weather prediction, as well as in monitoring and distributed system workloads and many more. Recently, deep learning models are often applied to solve time-series prediction due to their quality. While deep learning models such as recurrent neural networks are the most well-known in this direction, convolutional neural networks (CNNs) is more known for image processing. However, CNNs are also a strong candidate for sequence modeling as well as time-series forecasting. In general, deep learning models are often trained by backpropagation using an optimization algorithm like gradient descent. In this paper, we design a novel variant for training CNN based on meta-heuristic algorithm Equilibrium Optimization (EO). The proposed model, therefore, is called by EO-CNN is consequently applied to traffic transportation envisioning. To evaluate our model, we employ real-time road traffic data, including occupancy, speed, and travel time datasets collected from specialized traffic sensors at the Twin Cities Metro area in Minnesota. The experimental results proved that our design works effectively in application domains such as transportation with excellent performance in comparison with existing well-known approaches.

Multi-Feature Fusion and Enhancement Single Shot Detector for Traffic Sign Recognition

Road traffic sign detection and recognition play an important role in advanced driver assistance systems (ADAS) by providing real-time road sign perception information. In this paper, we propose an improved (Single Shot Detector) SSD algorithm via multi-feature fusion and enhancement, named MF-SSD, for traffic sign recognition. First, low-level features are fused into high-level features to improve the detection performance of small targets in the SSD. We then enhance the features in different channels to detect the target by enhancing effective channel features and suppressing invalid channel features. Our algorithm gets good results in domestic real-time traffic signs. The proposed MF-SSD algorithm is evaluated with the German Traffic Sign Recognition Benchmark (GTSRB) dataset. The experimental results show that the MF-SSD algorithm has advantages in detecting small traffic signs. Compared with existing methods, it achieves higher detection accuracy, better efficiency, and better robustness in complex traffic environment.

Deep-Learning-Based Real-Time Road Traffic Prediction Using Long-Term Evolution Access Data.

In this paper, we propose a method for deep-learning-based real-time road traffic predictions using long-term evolution (LTE) access data. The proposed system generates a road traffic speed learning model based on road speed data and historical LTE data collected from a plurality of base stations located within a predetermined radius from the road. Real-time LTE data were the input for the generated learning model in order to predict the real-time speed of traffic. Since the system was developed using a time-series-based road traffic speed learning model based on LTE data from the past, it is possible for it to be used for a road where the environment has changed. Moreover, even on roads where the collection of traffic data is invalid, such as a radio shadow area, it is possible to directly enter real-time wireless communications data into the traffic speed learning model to predict the traffic speed on the road in real time, and in turn, raise the accuracy of real-time road traffic predictions.

Efficient Outdoor Video Semantic Segmentation Using Feedback-Based Fully Convolution Neural Network

In this article, we focus on efficient semantic segmentation problem from sequential two-dimensional images, in which all pixels are classified into certain classes for scene understanding. Such problem is challenging because it involves constraints of both spatial and temporal consistencies, which have large difficulties in explicitly determining such structural constraints. Traditionally, such a problem is tackled using structured prediction method, such as conditional random field (CRF). However, pure CRF method suffers from very high complexity in computing high-order potentials and slow performance during inference step, which is unsuitable for efficient video segmentation in real scenario. In this article, a novel feedback-based deep fully convolutional neural network (CNN) is proposed to inherently incorporate spatial context through appending output feedback mechanism. The proposed method has the following contributions: 1) spatial context in images are easily captured through iterative feedback refinement, without the expensive postprocess step such as CRF refinement; 2) easily integrated with generic deep CNN structure; and 3) the inference time is greatly reduced for efficient image segmentation. Compared to current state-of-the-art methods, our proposed method was shown to provide up to 14% better accuracy on semantic segmentation task in challenging Camvid and Cityscapes datasets, while taking up to relatively 980% shorter inference time. The proposed method also shows its effectiveness for real-time road detection task of autonomous driving.