[AV][EN]Aurora.tech

0. Abstract

  • Aurora.tech

  • Found by Chris Urmson, Aurora offers self‑driving platform combines hardware, software, and data services that allow vehicles to move people and goods safely through the world.

  • He was at TED, talking about autonomous vehicle, giving big impression to audience regarding future ‘moving’ scheme. Below video is worth spending some minutes.



1. Team

  • Aurora’s leadership stands out for their depth of experience in artificial intelligence, robotics, and computer vision.

KEY Figures

  • Chris Urmson, CEO
    • Led the Google self-driving car team
    • Carnegie Mellon’s technology director when it won the 2007 DARPA Urban Challenge
  • Sterling Anderson
    • Developed MIT’s Intelligent CoPilot
    • Led the teams that launched Tesla’s Model X and Autopilot.
  • Drew Bagnell
    • Spent two decades at the intersection of machine learning and robotics both as a faculty member at Carnegie Mellon University and in engagements with industry from self-driving haul trucks to perception architecture for Uber’s self-driving cars.


2. Product


  • The Aurora Driver
    • It enables vehicles powered by it to move safely and efficiently through the world.
    • Comprised of the sensors, software and computer that powers and intergrates them both with the vehicle.
  • The Aurora Cloud
    • Data Services for Self‑Driving - provides a common interface tying together vehicles powered by the Aurora Driver with the people and goods it moves, the transportation networks it supports, and the fleet-management apparatus it requires.
  • Logistics Services
    • Communicate with the Aurora Cloud to dispatch and route vehicles to move goods between ports, distribution centers, homes, and businesses.
    • Vehicles for logistics services is custom-designed from larger platforms, such as cargo vans or large trucks.
  • Mobility Services -
    • It request rides from the Aurora Driver over a common API. The Aurora Cloud provides monitoring and support for the vehicles and their occupants. Vehicles used for these services tend to be smaller platforms, such as sedans, minivans, and SUVs.
  • Fleet Management Services
    • Coordinate with the Aurora Cloud to park, charge, warehouse, and maintain vehicles powered by the Aurora Driver.

3. Technology

  • Hardware
    • Designed to be easily installed on diverse vehicle platforms.
    • Crafted each element of hardware to be extensible, allowing quick adaptation and iteration when new technologies become available
    • Custom-built components that allow system to centrally manage power, communications, and networking.
    • Used silicon specifically designed to accelerate ML and facilitate camera signal processing.


  • Localization and Maps
    • Detailed maps that help Driver plan ahead.
    • As Maps are also critical for localization, which is the process vehicles use to determine where they are in the world.
    • Localization software can determine position within 10 centimeters and 0.1 degree of accuracy(even in environments that deny or deceive GPS)
  • Perception
    • Analyzing sensor data and generating a representation of all the actors, like pedestrians, that could affect our Driver
    • To handle inevitable edge cases, perception system is designed to “see” and track all perceived objects, even ones it doesn’t recognize.


  • Planning and Control
    • Planning and control software forecasts the intention and motion of actors, determines what Driver should do in response
    • Ultimately translates that plan into throttle, brake, and steering commands for the vehicle.


  • Infrastructure/Cloud
    • Allowing the same code to execute on laptops, vehicles, and in the cloud, rapid, scalable development and efficient testing is possible
    • Framework logs all vehicle events, from sensor data to intermediate results
    • Libraries, UI tools, and other creative solutions to help teams across the company efficiently store, sort, process, and visualize truly massive amounts of complex data.

4. etc.,

“Aurora is not building autonomous cars, it’s building safe drivers.” - PCMag

  • Aurora considers that safety is intrinsic to how Aurora is developing technology and how Aurora make every decision along the way.
  • Despite being company which builds self-driving car, Aurora is engaged with local, state, and federal governments to support effective regulatory frameworks for the introduction of self-driving vehicles.

  • Aurora writes Safety Report which let us know it put so much weights on the safety Safety Report by Aurora.tech

  • And also, team members’ addressing safety with blazing eyes tells us the philosophy of Aurora(Below Video)



The End.

2020

[Robot][EN]Hivemapper

2 minute read

Hivemapper,a decentralized mapping network that enables monitoring and autonomous navigation without the need for expensive sensors, aircraft, or satellites.

[CV][GAN][EN]Cycle GAN

8 minute read

Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

Back to top ↑