It is anticipated that the intelligent transportation system (ITS) will continue to have a mix of manual and autonomous cars for many years to come. Therefore, before driverless vehicles become completely mainstream, it is essential to address the concerns about safety raised by this mix of manual and autonomous vehicles. Blended traffic structures: security as well as comfort considerations are greatly impacted by the problems autonomous cars present, such as the ITS system has grown more complicated. There has been a low intentions identification rate and poor immediate performance when anticipating the driving path. Consequently, research is needed to create an increasingly advanced Intelligent Transportation System (ITS), and self-driving cars need to be able to predict their path in real time depending on the traffic conditions surrounding them. In this study, we describe a deep learning-based traffic safety method for an ITS that combines human drivers with autonomous vehicles. In 5G-enabled ITS, this technique uses two information sets as network inputs to Bidirectional Long-short term memory (BiLSTM) networks: one for driving trajectories and the other for normal driving. The softmax function computes the probability matrix for each aim. The average of each rule in a decision layer is then fused to get the final intended probabilities. With intentions identification rates of 91.58% and 90.88% for left and right lanes moveings, accordingly, the experimental outcomes demonstrate that the suggested strategy satisfactorily resolves the lanes shift problem in a mixed traffic scenario and enhances accuracy and real-time intention detection.
Autonomous and Manual Vehicles System Based on 5G-Enabled Intelligent Transportation System
28.11.2024
4495391 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Autonomous Vehicles in Intelligent Transportation Systems
Springer Verlag | 2021
|Autonomous Vehicles and Intelligent Transportation Systems—A Framework of Intelligent Vehicles
Springer Verlag | 2022
|