Lane Change Classification and Prediction with Action Recognition Networks

arXiv:2208.11650v3 Announce Type: replace
Abstract: Anticipating lane change intentions of surrounding vehicles is crucial for efficient and safe driving decision making in an autonomous driving system. Previous works often adopt physical variables such as driving speed, acceleration and so forth for lane change classification. However, physical variables do not contain semantic information. Although 3D CNNs have been developing rapidly, the number of methods utilising action recognition models and appearance feature for lane change recognition is low, and they all require …

abstract action recognition arxiv autonomous autonomous driving autonomous driving system change classification cs.cv decision decision making driving however making networks prediction recognition safe semantic speed type variables vehicles