End-to-end Driving via Conditional Imitation Learning

https://arxiv.org/abs/1710.02410

End-to-end Driving via Conditional Imitation Learning

Felipe Codevilla, Matthias Müller, Alexey Dosovitskiy, Antonio López, Vladlen Koltun

(Submitted on 6 Oct 2017)

Deep networks trained on demonstrations of human driving have learned to follow roads and avoid obstacles. However, driving policies trained via imitation learning cannot be controlled at test time. A vehicle trained end-to-end to imitate an expert cannot be guided to take a specific turn at an upcoming intersection. This limits the utility of such systems. We propose to condition imitation learning on high-level command input. At test time, the learned driving policy functions as a chauffeur that handles sensorimotor coordination but continues to respond to navigational commands. We evaluate different architectures for conditional imitation learning in vision-based driving. We conduct experiments in realistic three-dimensional simulations of urban driving and on a 1/5 scale robotic truck that is trained to drive in a residential area. Both systems drive based on visual input yet remain responsive to high-level navigational commands. Experimental results demonstrate that the presented approach significantly outperforms a number of baselines. The supplementary video can be viewed at this https URL

使用模拟器 carla也进行了训练,真实环境测试视频也有。

机器之心对carla的介绍: https://baijiahao.baidu.com/s?id=1586489273374923446&wfr=spider&for=pc

原文发布于微信公众号 - CreateAMind(createamind)

原文发表时间:2018-01-27

本文参与腾讯云自媒体分享计划,欢迎正在阅读的你也加入,一起分享。

发表于

我来说两句

0 条评论
登录 后参与评论

相关文章

来自专栏后端云

云环境中GPU配置

Ocata周期的科学技术重点之一是进一步扩展OpenStack中GPU支持的状态。这里的第一个问题是我们在讨论GPU支持时正在讨论的问题,因为使用现有的Open...

2373
来自专栏计算机视觉与深度学习基础

Leetcode 5 Longest Palindromic Substring

Given a string S, find the longest palindromic substring in S. You may assume ...

1885
来自专栏進无尽的文章

多媒体-图片、音频、视频压缩

很多时候我们需要把一些图片、音频、视频上传到服务器,于是就有了文件压缩这个问题了,这篇文章就小结一下具体的文件压缩实现吧。

2991
来自专栏linux驱动个人学习

高通Audio中ASOC的machine驱动

ASoC被分为Machine、Platform和Codec三大部分,其中的Machine驱动负责Platform和Codec之间的耦合以及部分和设备或板子特定的...

1.3K4
来自专栏专知

关于写论文说来简单但做起来难的三条建议

A few years ago, we prepared a series of workshops on writing research papers an...

2905
来自专栏机器人网

别让接线这件小事,拉开你与工程师的差距

导线与导线的连接、线头与接线桩的连接,事情小,责任大。本文图文并茂,让你清清楚楚看懂! 导线与导线的连接 导线的连接情况有:单股铜芯导线的直线连接、T字形连接;...

3437
来自专栏和蔼的张星的图像处理专栏

8.SSD目标检测之二:制作自己的训练集

最近秋色甚好,一场大风刮散了雾霾,难得几天的好天气,周末回家在大巴上看着高速两旁夕阳照射下黄澄澄的树叶,晕车好像也好了很多。 特地周六赶回来为了周末去拍点素材...

1634
来自专栏GIS讲堂

ArcGIS Image Server简介以及OL2中的加载

本文讲述Arcgis Image Server相关以及在OL2中如何加载Arcgis Server发布的影像服务。

1152
来自专栏CreateAMind

Ray RLlib: Scalable Reinforcement Learning

1612
来自专栏生信技能树

使用ESTIMATE来对转录组表达数据根据stromal和immune细胞比例估算肿瘤纯度

ESTIMATE (Estimation of STromal and Immune cells in MAlignant Tumor tissues usin...

4982

扫码关注云+社区

领取腾讯云代金券