Shinpei Kato edited this page Aug 20, 2018 · 39 revisions

Autoware.AI

Welcome to the Autoware Wiki

Autoware is ROS-based open-source software, enabling self-driving mobility to be deployed in open city areas. It provides, but not limited to, the following modules.

Localization is achieved by 3D maps and SLAM algorithms in combination with GNSS and IMU sensors. Detection uses cameras and LiDARs with sensor fusion algorithms and deep neural networks. Prediction and Planning are based on probabilistic robotics and rule-based systems, partly using deep neural networks as well. The output of Autoware to the vehicle is a twist of velocity and angular velocity (also curvature). This is a part of Control, though the major part of Control is supposed to reside in the by-wire controller of the vehicle, where PID and MPC algorithms are often adopted.

All in all, Autoware provides a complete software stack for self-driving technology. Join Autoware now, and your contribution will be loved by the world.

Getting started

You can easily install Autoware using Docker and run the demo using ROSBAG.

  1. Installation
  2. Demo

How to contribute

  • Coding

Autoware is managed by Github at https://github.com/cpfl/Autoware. You are always welcome to fork it and send your contribution as a pull request to the repository. Whenever you contribute to coding of Autoware modules, however, please respect and follow the Contribution Rules so that the repository can keep organized. To install Autoware, we strongly recommend you using Autoware Docker. Otherwise, you can follow the instruction provided by Source Build.

  • Field testing

Autoware is widely used in research and development on self-driving technology. Many Autoware-based self-driving cars have shown amazing demonstrations of on public roads. A remarkable story in early days is that Udacity has adopted Autoware (see the story) for their Self-Driving Car Nanodegree Program, and one spinout team has made a successful field demonstration on El Camino Real, CA (see the story). Other examples include: [Palo Alto, USA][Aichi, Japan][Tokyo, Japan]. It is impressive that a few Japanese teams have made Autoware-based driverless cars on public roads: [Aichi, Japan][Shimane, Japan][Tokushima, Japan]. The first media coverage of Autoware-based driverless car is here. We appreciate many successors of field testing. Please also upload your ROSBAG files recorded during field testing to ROSBAG STORE so that other Autoware and ROS users who do not own cars can reproduce the scenes and simulate the functions.

  • Production

Autoware can also be used for products and services. The following are some examples. Tier IV offers a compact self-driving development kit, called AI Pilot, where cameras, LiDAR, GPS/IMU, and computers (DRIVE PX2, R-Cars, etc.) are all integrated in a package. AutonomouStuff distributes by-wire vehicles, where Autoware is available. ZMP also distributes Autoware-preinstalled vehicles. Aisan Technology provides high-definition/accuracy/resolution 3D maps that use the mapping format supported by Autoware (see videos). We appreciate these companies choosing Autoware as a commercial solution, and hope to see more and more companies adopting Autoware in production.

You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.
Press h to open a hovercard with more details.