Reinforcement learning based edge computing in B5G

The development of communication technology will promote the application of Internet of Things,and Beyond 5G will become a new technology promoter.At the same time,Beyond 5G will become one of the important supports for the development of edge computing technology.This paper proposes a communication task allocation algorithm based on deep reinforcement learning for vehicle-to-pedestrian communication scenarios in edge computing.Through trial and error learning of agent,the optimal spectrum and power can be determined for transmission without global information,so as to balance the communication between vehicle-to-pedestrian and vehicle-to-infrastructure.The results show that the agent can effectively improve vehicle-to-infrastructure communication rate as well as meeting the delay constraints on the vehicle-to-pedestrian link.

Study on pedestrian thorax injury in vehicle-to-pedestrian collisions using finite element analysis

Objective： To explore the relationship between the collision parameters of vehicle and the pedestrian thorax injury by establishing the chest simulation models in car-pedestrian collision at different velocities and angles. Methods： 87 cases of vehicle-to-pedestrian accidents, with detailed injury information and determined vehicle impact parameters, were included. The severity of injury was scaled in line with the Abbreviated Injury Scale （AIS）. The chest biomechanical response parameters and change characteristics were obtained by using Hyperworks and LS-DYNA computing. Simulation analysis was applied to compare the characteristics of injuries. Results： When impact velocities at 25, 40 and 55 km/h, respectively, 1） the maximum values of thorax velocity criterion （VC） were for 0.29, 0.83 and 2.58 m/s; and at the same collision velocity, the thorax VC from the impact on pedestrian＇s front was successively greater than on his back and on his side; 2） the maximum values of peak stress on ribs were 154,177 and 209 MPa; and at the same velocity, peak stress values on ribs from the impact on pedestrian＇s side were greater than on his front and his back. Conclusion： There is a positive correlation between the severity and risk of thorax injury and the collision velocity and angle of car-thorax crashes. At the same velocity, it is of greater damage risk when the soft tissue of thorax under a front impact; and there is also a greater risk of ribs fracture under a side impact of the thorax. This result is of vital significance for diagnosis and protection of thorax collision injuries.

Simulation Evaluation of Filtering Method for Improving Pedestrian Positioning Accuracy Using Signal Strengths

In recent years, we have been able to use various services using the position information of smartphones and tablets. In addition, research on intelligent transport systems (ITS) has been actively conducted. To consider reducing traffic accidents by exchanging position information between pedestrians and vehicles by vehicle-to-pedestrian communication, we require accurate position information for pedestrians and vehicles. The GPS (global positioning system) is the most widely used method for acquiring position information. However, in urban areas, the GPS signal is affected by the surrounding buildings, which increases the positioning error. In this study, a method to improve the positioning accuracy of pedestrians using the signal strengths from vehicles and beacons was proposed. First, a Kalman filter was applied to the signal strength. Then, the path loss index was dynamically calculated using vehicle-to-vehicle communication. Finally, the position of a pedestrian was obtained using weighted centroid localization (WCL) after filtering the nodes. The positioning accuracy was evaluated using a simulator and demonstrated the superiority of the proposed method.

Pedestrian trajectory prediction method based on the Social-LSTM model for vehicle collision

Techniques for predicting the trajectory of vulnerable road users are important to the development of perception systems for autonomous vehicles to avoid accidents.The most effective trajectory prediction methods,such as Social-LSTM,are often used to predict pedestrian trajectories in normal passage scenarios.However,they can produce unsatisfactory prediction results and data redundancy,as well as difficulties in predicting trajectories using pixel-based coordinate systems in collision avoidance systems.There is also a lack of validations using real vehicle-to-pedestrian collisions.To address these issues,some insightful approaches to improve the trajectory prediction scheme of Social-LSTM were proposed,such methods included transforming pedestrian trajectory coordinates and converting image coordinates to world coordinates.The YOLOv5 detection model was introduced to reduce target loss and improve prediction accuracy.The DeepSORT algorithm was employed to reduce the number of target transformations in the tracking model.Image Perspective Transformation(IPT)and Direct Linear Transformation(DLT)theories were combined to transform the coordinates to world coordinates,identifying the collision location where the accident could occur.The performance of the proposed method was validated by training tests using MS COCO(Microsoft Common Objects in Context)and ETH/UCY datasets.The results showed that the target detection accuracy was more than 90%and the prediction loss tends to decrease with increasing training steps,with the final loss value less than 1%.The reliability and effectiveness of the improved method were demonstrated by benchmarking system performance to two video recordings of real pedestrian accidents with different lighting conditions.