- Introduction
- Installing WSL
- Sample ROS Autonomous Driving Project in Docker
- Kuksa
- Autoware
- Ankaios:
- Autowrx Gitlab Repository
- ThreadX
- Uprotocol
- Influxdb Connection
- TODOS
This document aims to provide an insight into the installation and usage of various tools and technologies for creating a simulation environment for autonomous systems and IoT projects. It covers the setup and integration of tools like Windows Subsystem for Linux (WSL), Nvidia Docker, ROS, Kuksa, Autoware, and Ankaios, allowing users to:
- Simulate and control autonomous vehicles using containerized environments.
- Process and communicate vehicle signal data through standards like VSS with Kuksa.
- Manage fleets using Ankaios and OCI-compatible tools.
In order to run a container based simulation environment there is a need to access to Nvidia GPU and setup Nvidia docker run time. It is easier to setup a WSL for Nvidia GPU configuration as it by default has direct access to sources of the host (compared to VM setup).
To install WSL the following tutorial can be followed: Installing WSL on Windows 10
After setting up Nvidia Docker run-time,
A sample autonomous driving project can be initialized with:
docker pull noshluk2/ros2-self-driving-car-ai-using-opencv:latest
docker run -it \
--name=ros2_sdc_container \
--env="DISPLAY=$DISPLAY" \
--env="QT_X11_NO_MITSHM=1" \
--volume="/tmp/.X11-unix:/tmp/.X11-unix:rw" \
--env="XAUTHORITY=$XAUTH" \
--volume="$XAUTH:$XAUTH" \
--net=host \
--privileged \
noshluk2/ros2-self-driving-car-ai-using-opencv \
bash
# Bridging the environment with Prius Car
ros2 launch self_driving_car_pkg world_gazebo.launch.py
# In another terminal run:
docker exec -it ros2_sdc_container bash
cd ~/ROS2-Self-Driving-Car-AI-using-OpenCV/
ros2 run self_driving_car_pkg computer_vision_node # Self driving mode
Important Note: Based on the documentation the original port to publish Kuksa.Val serveris 55555, however it doesn't work, using port 55556 is an alternative.
docker run -it --rm --publish 55556:55556 ghcr.io/eclipse/kuksa.val/databroker:master --port 55556 --insecure# To install Kuksa Python library:
pip install kuksa_client#!/usr/bin/env python3
# sample_kuksa.py
# The following project is executed using version 0.4.3
from kuksa_client.grpc import VSSClient
from kuksa_client.grpc import Datapoint
import time
with VSSClient('127.0.0.1', 55556) as client: # Attach to local host port 55556 to listen kuksa.val server
for speed in range(0, 100):
client.set_current_values({
'Vehicle.Speed': Datapoint(speed),
})
print(f"Feeding Vehicle.Speed to {speed}")
# Read back the value to confirm it was set correctly
response = client.get_current_values(['Vehicle.Speed'])
if 'Vehicle.Speed' in response:
read_back_speed = response['Vehicle.Speed'].value
print(f"Read back Vehicle.Speed: {read_back_speed}")
else:
print("Failed to read back Vehicle.Speed")
time.sleep(1)
print("Finished.")usr@usr:~$ chmod +x ./sample_kuksa.py
usr@usr:~$ python3 ./sample_kuksa.py
# Expected output:
Feeding Vehicle.Speed to 0
Read back Vehicle.Speed: 0.0
Feeding Vehicle.Speed to 1
Read back Vehicle.Speed: 1.0
Feeding Vehicle.Speed to 2
Read back Vehicle.Speed: 2.0
Feeding Vehicle.Speed to 3
Read back Vehicle.Speed: 3.0
...Kuksa supports VSS data type specification. For detailed documentation on VSS data types and definitions please refer to:
- Installation Process
# Clone autowarefoundation/autoware and move to the directory.
git clone https://github.com/autowarefoundation/autoware.git
cd autoware
# The setup script will install all required dependencies with Ansible:
./setup-dev-env.sh -y docker
# To install without NVIDIA GPU support:
./setup-dev-env.sh -y --no-nvidia dockerIn case you get such error "NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver." Please follow the video:
Important Note: Ankaios uses OCI image format. Hence it leverages Podman as container manager.
To install Podman on Ubunutu the following tutorial can be followed:
sudo sh -c "echo 'deb http://download.opensuse.org/repositories/devel:/kubic:/libcontainers:/stable/xUbuntu_${VERSION_ID}/ /' > /etc/apt/sources.list.d/devel:kubic:libcontainers:stable.list"
wget -nv https://download.opensuse.org/repositories/devel:kubic:libcontainers:stable/xUbuntu_${VERSION_ID}/Release.key -O- | sudo apt-key add -
sudo apt-get update -qq
sudo apt-get -qq --yes install podmanIt is possible to convert an existing Docker format image to OCI image format. Following tutorial shows the steps on how to convert a Docker image format into OCI image format which Ankaios supports.
- Docker2OCI
- Supports Docker2OCI conversion
- Supports validation of an Image based on OCI formatting
# state.yaml
apiVersion: v0.1
kind: Config
workloads:
nginx:
runtime: podman
agent: agent_A
restart: true
updateStrategy: AT_MOST_ONCE
accessRights: #
allow: []
deny: []
tags:
- key: owner
value: Ankaios team
runtimeConfig: |
image: docker.io/nginx:latest
ports:
- containerPort: 80
hostPort: 8081ank-server -k --startup-config state.yaml # Start ankaios server with provided configuration
ank -k apply state.yaml # Apply the workload
ank -k get workload # Get all workload states
ank -k get state # Get detailed state about all workload states- For IOT tracability demonstrate ROS/Autoware topic subscription and data visualization through Influxdb / Grafana