raccoonlab-tools 0.0.18

RaccoonLab tools

Cyphal/DroneCAN nodes tools

tools is a collection of scripts for testing and configuration of Cyphal/CAN and DroneCAN nodes.

This package is under development.

1. INSTALLATION

1. Install the package from pypi, test.pypi or from sources

pip install raccoonlab-tools

pip install -i https://test.pypi.org/simple/ raccoonlab-tools

git clone https://github.com/PonomarevDA/tools.git
cd tools
pip install .

2. Install dependencies

If you already have a Cyphal in your system, you probably don't need to install anything additional and have your own way of handling the environment variables and DSDL. Just check scripts/ubuntu.sh script in case if you miss something.

But if you use Cyphal for the first time or deploy a project in a new system, consider to run scripts/ubuntu.sh script. This script:

1. Install all recommended dependencies,
2. Clone recommended DSDL to ~/.cyphal directory,
3. Create in ~/.cyphal directory a setup.sh script that configure cyphal related environment variables and append source $HOME/cyphal/setup.sh to the end of your .bashrc file, so your shell will automatically setup the environment variables.

# By default, it installs both Cyphal and DroneCAN dependencies:
./scripts/ubuntu.sh

# Try --help option to get usage details. It allows to perform more precise installation:
./scripts/ubuntu.sh --help

Click here for details about which environment variables are required for a Cyphal application

To start with Cyphal/CAN (pycyphal, yakut, yukon) the following environment variables should be configured:

Environment variable	Meaning
CYPHAL_PATH	Path to DSDL. Let's use the default:$HOME/.cyphal
UAVCAN__NODE__ID	The application node identifier
UAVCAN__CAN__IFACE	CAN iface name
UAVCAN__CAN__BITRATE	Arbitration/data segment bits per second
UAVCAN__CAN__MTU	Maximum transmission unit: 8 for classic CAN

Check pycyphal/yakut/yukon docs for additional details

2. LINUX PREPARATION (SOCKETCAN)

By default, DroneCAN and Cyphal/CAN uses cross-platform transport interface Python-CAN CAN over Serial / SLCAN.

On Linux, the socketcan interface is recommended. Unlike SLCAN, socketcan interface allows to share the same CAN interface with multiple processes, so you can run a few pycyphal scripts, yukon, yakut simultaniously.

You can run the following script:

./scripts/socketcan.sh

For a Cyphal application after creating socketcan interface, you need to update UAVCAN__CAN__IFACE environment variable. Just call source ~/.bashrc.

3. USAGE

1. Test cyphal specification

rl-test-cyphal-specification

2. Test dronecan specification

rl-test-dronecan-specification

rl-test-dronecan-gps-mag-baro

3. Get Node Info (Cyphal / DroneCAN)

rl-get-info

Return:

• Online CAN-sniffers
• Detect protocol if any CAN-node is avaliable: cyphal | dronecan | none
• Show node info of Cyphal/CAN or DroneCAN node if it is avaliable

4. Upload firmware with st-link linux / STM32CubeProgrammer Windows

rl-upload-firmware --config PATH_TO_YAML_CONFIG

rl-upload-firmware --binary PATH_TO_BIN_FILE

There are a few ways how you can specify the path to the binary:

1. (recommended) Using a GitHub Repository. It will always download the latest released firmware.

# config.yaml
metadata:
    link: RaccoonlabDev/mini_v2_node

2. Direct Link to the Firmware File

# config.yaml
metadata:
    link: https://github.com/RaccoonlabDev/docs/releases/download/v1.6.5/gnss_v2_cyphal_v1.6.5_c78d47c3.bin

3. Using a Local Path

# config.yaml
metadata:
    link: /user/home/firmwares/node.bin

4. Direct .bin path with --binary option

5. Upload config

rl-config --config PATH

Example of yaml config file:

# config.yaml
params:
    uavcan.node_id: 31
    uavcan.node.name: co.rl.mini

6. Monitor

rl-monitor

This script is used for automated node analysis. Within a single command:

• detect the protocol (Cyphal or DroneCAN),
• detect the node type,
• check the software version and highlight if it is not the latest,
• configure it if it has not been configured yet,
• subscribes on all possible topics,
• provide basic tests and diagnostics, highlight issues,
• publish some test commands if possible,
• print all data in real time.

An example of rl-monitor with gnss node:

7. U-center with Cyphal GNSS

rl-ublox-center

For RL GNSS it is possible to run u-center over Cyphal via gps.ubx_tx and gps.ubx_rx topics.

1. Download u-center (tested with u-center 23.08)
2. Run the u-center. On ubuntu you can use wine (wine64 u-centersetup_v23.08/u-center_v23.08.exe)
3. Configure the Cyphal environment
4. Run the script rl-ublox-center
5. Press Receiver - > Connection -> Network connection
6. Add new connection tcp://127.0.0.1:2001

An illustration:

4. DEVELOPER NOTES

Deploy to TestPyPI:

./scripts/deploy.sh

Deploy to PyPI:

./scripts/deploy.sh --pypi

5. USAGE TERMS

The scripts are distributed under MIT license. In general, you can do with them whatever you want. If you find a bug, please suggest a PR or an issue.