These docs are public and open source.Edit on GitHub

THE FILE SYSTEM

After logging in as racecar, open the terminal in the home directory ~. The file system installed during setup is now visible on the car.

THE HOME DIRECTORY

bash
racecar@neoracer:~$ ls
data        Downloads   logs                osracer_demo   Public      Videos
Desktop     home        Music               osracer_ws     ros2_ws
Documents   jupyter_ws  neoracer-installer  Pictures       Templates

The car comes with stock Ubuntu directories such as Desktop, Documents, Downloads, and more. There are six directories that are installed on the car.

Directory	Description
ros2_ws	Stores the latest version of the driver.
jupyter_ws	Stores the Python library, starter code, and any additional code you write.
osracer_ws	The vendor stack: robot description, SLAM and Nav2.
neoracer-installer	The installer that set the car up.
logs	One timestamped folder per run of the driver.
data	Object-detection datasets, written by the camlabel dashboard.

ros2_ws stores the neoracer_ros2_driver pulled from GitHub. You can view your dashboards, trained models, launch and config files, and all the setup scripts used to initialise the NeoRacer.

bash
~/ros2_ws$ ls
build  install  log  src

~/ros2_ws/src$ ls
lakibeam1  neoracer_ros2_driver

~/ros2_ws/src/neoracer_ros2_driver$ ls
CHANGELOG.md     docs      neoracer_ros2_driver  scripts
CITATION.cff     launch    package.xml           setup.cfg
config           LICENSE   README.md             setup.py
CONTRIBUTING.md  models    resource              test

jupyter_ws is the main workspace where code is written. JupyterLab is available in your browser as a service on port 8888, and its file system is this folder.

bash
~/jupyter_ws$ ls
neoracer-os  README.md

~/jupyter_ws/neoracer-os$ ls
labs  library

~/jupyter_ws/neoracer-os/labs$ ls
demo.py     lab_a  lab_c  lab_e  lab_g  template.py  utility
grand_prix  lab_b  lab_d  lab_f  lab_i  tests

~/jupyter_ws/neoracer-os/library$ ls
camera.py      drive.py  nav.py       racecar_core.py   simulation   vision.py
controller.py  led.py    physics.py   racecar_utils.py  slam.py
display.py     lidar.py  __pycache__  real              telemetry.py

In the workspace there is labs and library. labs holds the lab exercises and some template code. library has all the rc.* modules that we use in the Python API reference. The tests folder holds the test notebook used in Test the system.

//Opening JupyterLab

It is one of the four services that start at boot, so there is nothing to launch. Join the car's network and open its address on port 8888.

bash
http://192.168.10.100:8888     # cudy router
http://10.42.0.1:8888          # the car's access point

//Notebooks or scripts

	Notebook (.ipynb)	Script (.py)
Good for	Running small tests, checking values, teaching	A normal program run start to finish
How you run it	Cell by cell in the browser	python3 <file> from a terminal
State	Kept in the kernel between cells	Gone when the program exits

//Restart the kernel

A notebook holds the car's sensors for as long as its kernel is alive, even after the last cell has finished. If you open a second notebook, or run a script while a notebook is still loaded, the second one might fail.

When you are done, open the Kernel menu and choose Restart Kernel and Clear Outputs of All Cells.

A second ROS 2 workspace, preinstalled on the image. It holds the autonomy tier: the robot description, SLAM and Nav2.

bash
~/osracer_ws$ ls
build  install  log  src

~/osracer_ws/src$ ls
lakibeam1  osracer

~/osracer_ws/src/osracer$ ls
CHANGELOG.md     osracer_calib       osracer_description  osracer_slam
docs             osracer_debug       osracer_navigation   README.md
LICENSE          osracer_demo        osracer_race         ruff.toml
osracer_bringup  osracer_dependency  osracer_sim          tools

bash
racecar ws            # which one am I on?
racecar ws osracer    # this terminal now uses the vendor workspace
racecar ws neoracer   # back to the default

Coming soon

//neoracer-installer

The installer downloads the driver onto the car. It is cloned and run once during setup. Its logs folder records every install run.

bash
~/neoracer-installer$ ls
docs  logs  README.md  scripts  tests

//logs

This refers to the logs in the ~ directory, not the logs in neoracer-installer. Every run of the stack writes a timestamped folder here, and logs/latest points at the most recent one. The health dashboard reads it.

//data

data is created by the camlabel dashboard the first time you use it. It holds the captured images, their labels, and object classes, in YOLO format. This folder becomes a training dataset.

◀Previous

Remote desktop

Next▶

API reference