To widen the range of deployment of autonomous vehicles, we need to develop more secure and intelligent systems exhibiting higher degrees of autonomy and able to sense, plan, and operate in unstructured environments. For that, the vehicle must be able to predict the intention of other traffic participants to interact coherently with its world. This paper addresses the prediction of lane change maneuver prediction of surrounding vehicles on highways. Two lane change prediction approaches based on machine learning are presented, the first is based on Support Vector Machine and the second on Artificial Neural Network, NGSIM dataset is used for training and testing. Used features are extracted from this dataset. The proposed approaches achieve a good performance, the results show improvement over the state of art in terms of prediction time and accuracy.
Prediction of Surrounding Vehicles Lane Change Intention Using Machine Learning
01.09.2019
1100323 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Two-Stream Networks for Lane-Change Prediction of Surrounding Vehicles
ArXiv | 2020
|Europäisches Patentamt | 2024
|