Autoware (Architecture Proposal)

What's this

This is the source code of the feasibility study for Autoware architecture proposal.

WARNING: This source is solely for demonstrating an architecture proposal. It should not be used to drive cars.

How to setup

Requirements

Hardware

• x86 CPU (8 or more cores)
• 16 GB or more of memory
• Nvidia GPU (4GB or more of memory) :

Software

• Ubuntu 18.04
• Nvidia driver

If cuda or tensorRT is already installed, it is recommended to remove it.

Autoware setup

1. Clone this repository

git clone https://github.com/tier4/AutowareArchitectureProposal.git
cd AutowareArchitectureProposal/

2. Run the setup script

./setup_ubuntu18.04.sh

In this step, the following software are installed. Please confirm their licenses before using them.

• osqp
• ROS Melodic
• CUDA 10.2
• cuDNN 7
• TensorRT 7
• geographiclib-tools

3. Build the source

catkin build --cmake-args -DCMAKE_BUILD_TYPE=Release

Note that the computer need to be connected to Internet to download neural network weight files.

How to run

Simulator

Quick Start

Rosbag

1. Download sample map from here.

2. Download sample rosbag from here.

3. Launch Autoware

roslaunch autoware_launch autoware.launch map_path:=[path] rosbag:=true

4. Play rosbag

rosbag play --clock [rosbag file] -r 0.2

Note

• sample map : © 2020 TierIV inc.
• rosbag : © 2020 TierIV inc.
  • Image data are removed due to privacy concerns.
    • Cannot run traffic light recognition
    • Decreased accuracy of object detection

Planning Simulator

1. Download sample map from here.

2. Launch Autoware

roslaunch autoware_launch planning_simulator.launch map_path:=[path]

3. Set initial pose
4. Set goal pose
5. Push engage button. autoware_web_controller

Note

• sample map : © 2020 TierIV inc.

Tutorial in detail

See here. for more information.

References

Videos

• Scenario demo
• Obstacle avoidance in the same lane
• Obstacle avoidance by lane change
• Object recognition
• Auto parking
• 360° FOV perception(Camera Lidar Fuison)
• Robustness of localization

Credits

• Neural Network Weight Files