The advantages of UAV video in flexibility, traceability, easy-operation, and abundant information make it a popular and powerful aerial tool applied in traffic monitoring in recent years. This paper proposed a systematic approach to detect and track vehicles based on the YOLO v3 model and the deep SORT algorithm for further extracting key traffic parameters. A field experiment was implemented to provide data for model training and validation to ensure the accuracy of the proposed approach. In the experiment, 5400 frame images and 1192 speed points were collected from two test vehicles equipped with high-precision GNSS-RTK and onboard OBD after completion of seven experimental groups with a different height (150 m to 500 m) and operating speed (40 km/h to 90 km/h). The results indicate that the proposed approach exhibits strong robustness and reliability, due to the 90.88% accuracy of object detection and 98.9% precision of tracking vehicle. Moreover, the absolute and relative error of extracted speed falls within ±3 km/h and 2%, respectively. The overall accuracy of the extracted parameters reaches up to 98%.

Polyvinylidene fluoride (PVDF) nanofiber mats have played a significant role in wearable electronic devices that have been in great demand in recent decades. Although manifold PVDFbased freely stacked or well-aligned nanofiber mats created via the electrospinning process have been demonstrated to achieve multisensory capabilities with high sensitivity and long detection range, rarely have any of them proved their ability with a stable process and accurate processing parameters. In this work, we successfully developed freely stacked and well-aligned PVDF nanofiber mats with diameters ranging from micrometers to nanometers, providing stable performance for wearable electronic devices. Through in-depth investigations into material preparation, electrospinning, and fiber collection processes, we revealed the relationship between the nanofiber morphology, β-phase fraction, and piezoelectric output with various process parameters. Characterized by analytical methods, we have established a mature, reliable nanofiber mat fabrication system capable of mass-producing PVDF nanofibers with the required diameter and consistent properties. At 18 kV voltage and 60% RH humidity, the uniformity of the fiber diameter and β-phase content was maintained in a favorable range. When the drum speed increased to 2000 r/s, the fiber orientation and β-phase content increased. We assembled aligned PVDF nanofiber mats with conductive fabric in a flexible piezoelectric sensor that successfully monitored different body movements and produced an output voltage of 0.1 V. This study provides the necessary process parameters for the large-scale production of high-quality PVDF nanofiber mats and provides clear guidance for beginners in the field of nanofiber mat manufacturing.

Driving behavior recognition can provide an important reference for the intelligent vehicle industry and probe vehicle-based traffic estimation. The identification of driving behavior using mobile sensing techniques such as smartphone- and vehicle-mounted terminals has gained significant attention in recent years. The present work proposed the monitoring of longitudinal driving behavior using a machine learning approach with the support of an on-board unit (OBU). Specifically, based on velocity, three-axis acceleration and three-axis angular velocity data were collected by a mobile vehicle terminal OBU; through the process of data preprocessing and feature extraction, seven machine learning algorithms, including support vector machine (SVM), random forest (RF), k-nearest neighbor algorithm (KNN), logistic regression (LR), BP neural network (BPNN), decision tree (DT), and the Naive Bayes (NB), were applied to implement the classification and monitoring of the longitudinal driving behavior of probe vehicles. The results show that the three classifiers SVM, RF and DT achieved good performances in identifying different longitudinal driving behaviors. The outcome of the present work could contribute to the fields of traffic management and traffic safety, providing important support for the realization of intelligent transport systems and the improvement of driving safety.

Pavement condition monitoring is an important task in road asset management and efficient abnormal pavement condition detection is critical for timely conservation management decisions. The present work introduces a mobile pavement condition monitoring approach utilizing low-cost sensor technology and machine-learning-based methodologies. Specifically, an on-board unit (OBU) embedded with an inertial measurement unit (IMU) and global positioning system (GPS) is applied to collect vehicle posture data in real time. Through a comprehensive analysis of both time domain and frequency domain data features for both normal and abnormal pavement conditions, feature engineering is conducted to identify how the most important features affect abnormal pavement condition recognition. Six machine learning models are then developed to identify different types of pavement conditions. The performance of different algorithms and the significance of different features are then analyzed. Moreover, the influence of vehicle speed on pavement condition assessment is further examined and classification models for different speed intervals are developed. The results indicate that the random forest (RF) model that considers vehicle speed achieves the best performance in pavement condition monitoring. The outcomes of the present work would contribute to cost-effective pavement condition monitoring and provide an important reference for pavement maintenance sectors.

The development of smart vehicles has increased the demand for high-definition road maps. However, traditional road maps for vehicle navigation systems are not sufficient to meet the requirements of intelligent vehicle systems (e.g., autonomous driving). The present work comes up with a method of generating high-definition map models based on the geographic information system (GIS). A systematic map construction framework including the road layer, intersection connection layer, and lane layer is proposed based on the GIS database. Specifically, the constrained Delaunay triangular network method is applied to extract road layer network models, which are then used as linear reference networks to construct lane-level road maps. To further examine the feasibility of the proposed framework, a field experiment is then conducted to build a high-definition road map. Furthermore, a lane-level map matching test is conducted in the constructed road map using the trajectory data collected from a probe vehicle. The results show that the proposed method provides an efficient way of extracting lane-level information from urban road networks and can be applied for lane-level map matching with good performance.