基于多模态深度学习的汽车虚拟驾驶环境生成方法

打开文本图片集

中图分类号：TP391.41 文献标志码：B

Generation method for automobile virtual driving environment based on multimodal deep learning

ZHANG Shusheng'， ZHU Xuefeng ，YE Qian’

1.School of Automobile Engineering，Dalian Universityof Technology，Dalian116O24，Liaoning，China： 2.Ningbo Research Institute of Dalian University of Technology，Ningbo 315ooo，Zhejiang，China）

Abstract：To facilitate the development of autonomous driving technology，a generation method for automobiles virtual driving environments based on deep learning of multimodal images is used to generate multimodal images which contain multiple physical scenes simultaneously. The encoder and generator are constructed using a partially shared hidden space，and the perceptual loss of domain-invariant properties is used，and the comparison experiments of Cityscapes and Comma2k19 datasets are carried out. The diversity evaluation metrics are adopted for evaluation. The results show that the virtual driving environment images generated by multimodal deep learning is high realism and diversity，which are important for the rapid construction of the virtual simulation platform for autonomous driving.

Key words： virtual driving environment； deep learning；multimodal； autonomous driving； imagegeneration；simulation platform

0引言

安全性是汽车工业中必须要考虑的关键问题，开发高标准的自动驾驶车辆更需要大量的行驶测试，而传统的道路行驶测试需要花费数十年甚至上百年的时间[1]。（剩余6842字）