Search

In this research, an optimal transit priority signal timing decision method was proposed by minimizing the personal average delay, aiming at resolving multiple conflicting priority requests at an isolated intersection. Based on connected vehicle technologies, it is available to obtain detailed information including lane-level locations and speeds of connected vehicles, signal plans and light states of traffic lights, traffic flow rates and queueing lengths at the intersection. The proposed method reallocates green time of every two signal stages, where transits are weighted by their on-time performance, loading factors and passenger priority. The delay of transit is estimated in five scenarios according to bus arrival time and queueing behavior, while the delay of other traffic flows is calculated with respect to the adjustment of green time. The result shows that the transit signal priority control achieved 8%-40% reductions of average personal delay and balance of travel completion, depending on various volumes of transits and general traffic, compared against the cases without transit signal priority control or with fixed green extensions.

Heavy use by cars and trucks leads to huge damages, so road damage detection is an essential task to road maintenance. Traditional road damage detection has to require a huge amount of manual effort, it is therefore of great interest to propose vision-based systems that can automatically detect the road damages. In this work, we use deep learning models to detect road damages efficiently. Specifically, we apply a segmentation method to detect the road areas and build a road-interest map for the raw images. Then we adopt the state-of-the-art deep objective detection model including Faster-RCNN and YOLOv4 for completing detection. Experiments convey that the proposed model achieves good detection performance on the IEEE Global Road Damage Detection Challenge 2020.

Using vehicles and roads to perceive pedestrians cooperatively can eliminate the visual limitation of a single vehicle, thus improve the pedestrian collision warning capability of the driving assistance system. In this paper, we propose a vehicle-road pedestrian re-identification (ReID) system for vehicle-road cooperative perception of pedestrians, which can provide real-time and reliable ReID results under the premise of minimizing the occupation of communication resources. Among the system, we propose the CNN-LSH feature extraction algorithm and adaptive feature dimension reduction algorithm to adjust the dimension of pedestrian images features effectively according to the status of wireless channel, and use Hamming distance metric to simplify the calculation of feature similarity, which accelerates the speed of image feature matching in ReID. Experimental results demonstrate that the vehicle-road ReID system can significantly reduce the communication overhead and computation time while providing reliable ReID results in real-time.