http://info.xcar.com.cn/201906/news_2039496_1.html

https://www.tesla.cn/careers/job/autopilot-simulationtoolsgineer-50003

The Role

Autopilot is of critical importance to Tesla’s mission. It is safer, makes driving more enjoyable, and will ultimately deliver on the promise of self-driving cars. As a member of Tesla’s Autopilot Simulation team, you will be in a unique position to accelerate the pace at which Autopilot improves over time. The main ways in which the simulation team realizes this include:

• building tools that enable Autopilot software developers to perform virtual test drives instead of real ones
• testing all Autopilot software releases for regressive behavior
• generating synthetic data sets for neural network training

The foundation on which we build these elements is our simulation environment. We develop photorealistic worlds for our virtual car to drive in, enabling our developers to iterate faster and rely less on real-world testing. We strive for perfect correlation to real-world vehicle behavior and work with Autopilot software engineers to improve both Autopilot and the simulator over time. Our group is a cross-disciplinary team with people from various backgrounds and fields of expertise, like CS generalists, mechanical engineers, game developers, and computer graphics artists. As a member on this team, you will be challenged to learn about all these different disciplines and use your own expertise to further improve the scope and usability of our tools.

As an Autopilot Simulation Tools Engineer, you will contribute to the development of the simulation by building tools that enable rapid prototyping/testing of Autopilot. You will be expected to contribute to such projects as photorealistic world generation, simulation scenario authoring, performance evaluation, and content validation tools. These simulation tools will both assist the simulation team and the broader Autopilot organization in both feature creation and performance improvement.

Responsibilities

• • Design and develop various content creation, simulation, reporting, and build tools
  • Contributing to the development of a modular, scalable simulation platform
  • Work with all members of the Autopilot team to understand simulation tooling requirements and champion good development/testing processes
  • Leverage and improve existing tools

Requirements

• • Experience developing and delivering engineering tools
  • Excellent software development skills in C++ & Python
  • BS/MS in Engineering, Computer Science, or the equivalent in experience and evidence of exceptional ability
  • Comfortable developing with Linux & Git
  • Knowledge of software life cycle, modern testing methods, and solid understanding of continuous integration
  • Able to work independently and proactively
  • Desirable to have experience working with and developing tools for Unreal Engine 4, Maya, and/or Houdini

ICCV——国际计算机视觉大会，英文全称International Conference on Computer Vision。

被誉为计算机视觉领域三大顶级会议之一，与CVPR和ECCV并列

1.  透镜畸变及校正模型

2. 计算机视觉：相机成像原理：世界坐标系、相机坐标系、图像坐标系、像素坐标系之间的转换

3. 相机畸变与标定

4. 使用OpenCV进行标定（Python）

https://blog.csdn.net/zhangkaixu321/article/details/79176923

1.  How to Install and Configure VNC on Ubuntu 16.04
2. https://github.com/deepdrive/deepdrive#dataset
3. https://deepdrive.io/
4. https://deepdrive.berkeley.edu/
5. ubuntu16.04挂载windows NTFS磁盘方法
6. sudo fdisk -l 查看所有磁盘分区
7. mount -t ntfs /dev/sdb3 /mnt/share -o iocharset=utf8,umask=0
8. Windows进入tensorflow env，打开anaconda prompt，activate tensorflow
9. Linux进入tensorflow env， source activate tensorflow

10. Key binds

    • Esc – Pause
    • Alt+Tab – Control other windows* ; – Toggle FPS
    • m – Toggle manual driving (control the car yourself)
    • 1 – Chase cam
    • 2 – Orbit (side) cam
    • 3 – Hood cam
    • 4 – Free cam (use WASD to fly)
    • WASD or Up, Down, Left Right – steer / throttle
    • Space – Handbrake
    • Shift – Nitro
    • H – Horn
    • L – Light
    • R – Reset
    • E – Gear Up
    • Q – Gear down
    • Z – Show mouse
    • </kbd><kbd> – Unreal console (first press releases game input capture)

http://www.leiphone.com/

Percy Liang 和团队针对这个问题提出的方法是尝试沿着模型的学习算法追踪模型的预测，一直反向追踪到模型参数产生的源头。他们希望这种方法——从训练数据的视角审视模型——可以成为开发、理解、诊断机器学习的标准方法的一部分。