A novel vehicle lateral positioning methodology based on the integrated deep neural network

作者：

Highlights：

• A novel lateral positioning methodology using the information of road areas.

• Designing an efficient deep convolutional neural network for road detection.

• Developing a multilayer neural network for the prediction of lateral position.

• The methodology obtains lane-level accuracy and applies to many road scenes.

摘要

•A novel lateral positioning methodology using the information of road areas.•Designing an efficient deep convolutional neural network for road detection.•Developing a multilayer neural network for the prediction of lateral position.•The methodology obtains lane-level accuracy and applies to many road scenes.

论文关键词：Integrated deep neural network,Road detection,Lateral position prediction,Vehicle lateral positioning

论文评审过程：Received 18 November 2018, Revised 14 July 2019, Accepted 28 September 2019, Available online 1 October 2019, Version of Record 10 October 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.112991