The traffic safety problem in the hybrid traffic scenario is a very important issue, among which, how to avoid traffic accidents by unmanned vehicles is particularly important. In this paper, Gaussian Mixed Hidden Markov Models (GMM-HMM) is used to predict the driving intention of human-driven vehicles in mixed traffic scenarios. Next Generation Simulation (NGSIM) dataset is used to train and test HMM representing different driving intentions. Experimental results show that the proposed method has better prediction effect than Support Vector Machine (SVM).
A Driving Intention Prediction Method for Mixed Traffic Scenarios
11.11.2022
539276 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Intention-Enhanced Vehicle Trajectory Prediction for Autonomous Driving in Highway Scenarios
Springer Verlag | 2025
|