02_installation.md

2. Installation

Setup for both x86-based and Jetson-based ECUs.

Note

Internet connection is required in this step.

2-1. x86-based ECU

Warning

Network settings are automatically updated.

During this procedure, IP addresses are assigned to some network interfaces (refer to the connection diagram on 1.Hardware setup for more detail) using netplan. This behavior may cause unexpected disconnection, if you are accessing the ECU remotely via those interfaces.

If you would like to change network interfaces or IP addresses to be assigned, edit edge-auto/ansible/playbooks/vars/edge_auto.yaml before running setup-dev-env.sh

Download the repository and setup your environment

As a first step, clone tier4/edge-auto and move to the directory.

git clone https://github.com/tier4/edge-auto.git
cd edge-auto

You can install the dependencies using the provided ansible script.

./setup-dev-env.sh

Finally, please reboot the system to make the installed dependencies and permission settings effective.

sudo reboot

Build edge-auto

Create your ROS workspace and clone repositories using vcstool.

cd edge-auto
mkdir src
vcs import src < autoware.repos

Install ros package dependencies and build your ROS workspace.

rosdep update
rosdep install -y --from-paths `colcon list --packages-up-to edge_auto_launch -p` --ignore-src --skip-keys autoware_launch

./build.sh

...
Finished <<< image_projection_based_fusion [0.25s]
Starting >>> edge_auto_launch
Finished <<< edge_auto_launch [0.22s]

Summary: 66 packages finished [6.85s]
  2 packages had stderr output: extrinsic_interactive_calibrator intrinsic_camera_calibrator

2-2. Jetson-based ECU

Note

This following steps can be performed from your x86-based ECU via ssh.

Download the repository and setup the environment

As a first step, clone tier4/edge-auto-jetson and move to the directory.

git clone https://github.com/tier4/edge-auto-jetson.git -b release/RQX-58G
cd edge-auto-jetson

Note

If you would like to contribute to the development, please clone the latest version of the source code from the main branch.

git clone https://github.com/tier4/edge-auto-jetson.git
cd edge-auto-jetson

You can install the dependencies using the provided ansible script. During the installation process, you will be asked if you want to install the TIER IV camera driver. If you already have the driver installed and want to skip this step, please type N to continue.

Note

setup-dev-env.sh script may take several hours.

./setup-dev-env.sh

[Warning] Do you want to install/update the TIER IV camera driver? [y/N]:
[Warning] Do you want to configure the network? This configuration may overwrite the IP address of the specific network interface [y/N]:

Finally, please reboot the system to make the installed dependencies and permission settings effective.

sudo reboot

If you chose y for the prompt of the network configuration, the IP address for the specified network interface is fixed (default: 192.168.2.2 for eth1). In that case, you can access the Jetson-based ECU and perform the following steps via remote access, such as:

ssh <IP_address> -l <username>

Build edge-auto-jetson workspace

Create your ROS workspace and clone repositories using vcstool.

cd edge-auto-jetson
mkdir src
vcs import src < autoware.repos

Build your ROS workspace.

./build.sh

(Optional) Update your workspace

If you want to update cloned repositories, use the following command.

vcs import src < autoware.repos
vcs pull src

(Optional) Modify camera exposure timing

Note

On the sample system introduced in 1.Hardware setup step, this doesn't need to be changed.

If you want to change the exposure time of cameras for sensor synchronization, please modify the following files.

edge-auto-jetson/src/individual_params/individual_params/config/
└── default
    ├── camera0
    │   ├── trigger.param.yaml
    ├── camera1
    │   ├── trigger.param.yaml

For more details, please refer to the tier4/sensor_trigger repository.