ODrive Firmware Developer Guide

This guide is intended for developers who wish to modify the firmware of the ODrive. As such it assumes that you know things like how to use Git, what a compiler is, etc. If that sounds scary, turn around now.

The official releases are maintained on the master branch. However since you are a developer, you are encouraged to use the devel branch, as it contains the latest features.

The project is under active development, so make sure to check the Changelog to keep track of updates.


The recommended tools for ODrive development are:

  • make: Used to invoke tup

  • Tup: The build system used to invoke the compile commands

  • ARM GNU Compiler: For cross-compiling code

  • ARM GDB: For debugging the code and stepping through on the device

  • OpenOCD: For flashing the ODrive with the STLink/v2 programmer

  • Python 3, along with the packages PyYAML, Jinja2 and jsonschema: For running the Python tools (odrivetool). Also required for compiling firmware.

See below for specific installation instructions for your OS.

Depending on what you’re gonna do, you may not need all of the components.

Once you have everything, you can verify the correct installation by running:

arm-none-eabi-gcc --version
arm-none-eabi-gdb --version
openocd --version             # should be 0.10.0 or later
tup --version                 # should be 0.7.5 or later
python --version              # should be 3.7 or later

Installing Prerequisites

sudo add-apt-repository ppa:team-gcc-arm-embedded/ppa
sudo apt-get update
sudo apt-get install gcc-arm-embedded
sudo apt-get install openocd
sudo apt-get install git-lfs
sudo add-apt-repository ppa:jonathonf/tup && sudo apt-get update && sudo apt-get install tup
sudo apt-get install python3 python3-yaml python3-jinja2 python3-jsonschema

Configuring the Build

To customize the compile time parameters, copy or rename the file Firmware/tup.config.default to Firmware/tup.config and edit the following parameters in that file:

  • CONFIG_BOARD_VERSION The board version you’re using. Can be v3.1, v3.2, v3.3, v3.4-24V, v3.4-48V, v3.5-24V, v3.5-48V, etc. Check for a label on the upper side of the ODrive to find out which version you have. Some ODrive versions don’t specify the voltage: in that case you can read the value of the main capacitors: 120uF are 48V ODrives, 470uF are 24V ODrives.

  • CONFIG_DEBUG Defines whether debugging will be enabled when compiling the firmware; specifically the -g -gdwarf-2 flags. Note that printf debugging will only function if your tup.config specifies the USB_PROTOCOL or UART_PROTOCOL as stdout and DEBUG_PRINT is defined. See the IDE specific documentation for more information.

You can also modify the compile-time defaults for all .config parameters. You will find them if you search for AxisConfig, MotorConfig, etc.

Building and Flashing the Firmware

  1. Run make in the Firmware directory.

  2. Connect the ODrive via USB and power it up.

  3. Flash the firmware using odrivetool dfu.


Automated Testing

This section describes how to use the automated testing facilities. You don’t have to do this as an end user.

The testing facility consists of the following components:

  • Test rig: In the simplest case this can be a single ODrive optionally with a single motor and encoder pair. Can also be multiple ODrives with multiple axes, some of which may be mechanically coupled.

  • Test host: The PC on which the test script runs. All ODrives must be connected to the test host via USB.

  • test-rig.yaml: Describes your test rig. Make sure all values are correct. Incorrect values may physically break or fry your test setup.

  • test_runner.py: This is the main script that runs all the tests.

  • …_test.py The actual tests

The Tests

  • analog_input_test.py: Analog Input

  • calibration_test.py: Motor calibration, encoder offset calibration, encoder direction find, encoder index search

  • can_test.py: Partial coverage of the commands described in CAN Protocol.

  • closed_loop_test.py: Velocity control, position control (TODO: sensorless control), brake regen current hard limit, current control with velocity limiting

  • encoder_test.py: Incremental encoder, hall effect encoder, sin/cos encoder, SPI encoders (AMS, CUI)

  • fibre_test.py: General USB protocol tests

  • nvm_test.py: Configuration storage

  • pwm_input_test.py: PWM input

  • step_dir_test.py: Step/dir input

  • uart_ascii_test.py: Partial coverage of the commands described in ASCII Protocol

All tests in a file can be run with e.g.:

python3 uart_ascii_test.py --test-rig-yaml ../../test-rig-rpi.yaml

See the following sections for a more detailed test flow description.

Our Test Rig

Our test rig essentially consists of the following components:

  • an ODrive as the test subject

  • a Teensy 4.0 to emulate external hardware such as encoders

  • a Motor + Encoder pair for closed loop control tests

  • a Raspberry Pi 4.0 as test host

  • a CAN hat for the Raspberry Pi for CAN tests

This document is therefore centered around this test rig layout. If your test rig differs, you may be able to run some but not all of the tests.

How to set up a Raspberry Pi as testing host

#. Install Raspbian Lite on a Raspberry Pi 4.0. This is easiest if you have a keyboard, mouse and screen (micro-HDMI!). I used the NOOBS Lite installer for this. Paste the ZIP-file’s contents onto a FAT32 formatted SD card (fs type 0b in fdisk) and boot it. Then follow the on-screen instructions.

  1. Prepare the installation:

    sudo systemctl enable ssh
    sudo systemctl start ssh
    # Transfer your public key for passwordless SSH. All subsequent steps can be done via SSH.
    sudo apt-get update
    sudo apt-get upgrade
    # Change /etc/hostname to something meaningful
  2. Add the following lines to /boot/config.txt:

    • enable_uart=1

    • dtparam=spi=on

    • dtoverlay=spi-bcm2835-overlay

    • dtoverlay=mcp2515-can0,oscillator=12000000,interrupt=25


      These oscillator and interrupt GPIO settings here are for the “RS485 CAN HAT” I have. There appear to be multiple versions, so they may be different from yours. Check the marking on the oscillator and the schematics.

  3. Remove the following arguments from /boot/cmdline.txt:
    • console=serial0,115200

  4. Append ODRIVE_TEST_RIG_NAME=[test-rig-name] to /etc/environment. The HWIL tests use this to look up the the file [test-rig-name].yaml which is supposed to describe your test rig.

  5. Reboot.

  6. Install the prerequisites:

    sudo apt-get install ipython3 python3-appdirs python3-yaml python3-jinja2 python3-usb python3-serial python3-can python3-scipy python3-matplotlib python3-ipdb git openocd
    # Optionally, to be able to compile the firmware:
    sudo apt-get install gcc-arm-none-eabi
  7. Install Teensyduino and teensy-loader-cli:

    sudo apt-get install libfontconfig libxft2 libusb-dev
    wget https://downloads.arduino.cc/arduino-1.8.13-linuxarm.tar.xz
    tar -xf arduino-1.8.13-linuxarm.tar.xz
    wget https://www.pjrc.com/teensy/td_153/TeensyduinoInstall.linuxarm
    chmod +x TeensyduinoInstall.linuxarm
    ./TeensyduinoInstall.linuxarm --dir=arduino-1.8.13
    sudo cp -R arduino-1.8.13 /usr/share/arduino
    sudo ln -s /usr/share/arduino/arduino /usr/bin/arduino
    git clone https://github.com/PaulStoffregen/teensy_loader_cli
    pushd teensy_loader_cli
    sudo cp teensy_loader_cli /usr/bin/
    sudo ln -s /usr/bin/teensy_loader_cli /usr/bin/teensy-loader-cli
    curl https://www.pjrc.com/teensy/49-teensy.rules | sudo tee /etc/udev/rules.d/49-teensy.rules
  8. Add the following lines to /etc/udev/rules.d/49-stlinkv2.rules:

    SUBSYSTEMS=="usb", ATTRS{idVendor}=="0483", ATTRS{idProduct}=="374b", MODE:="0666"
    SUBSYSTEMS=="usb", ATTRS{idVendor}=="0483", ATTRS{idProduct}=="3748", MODE:="0666"
  9. sudo mkdir /opt/odrivetest && sudo chown $USER /opt/odrivetest

  10. At this point you need the ODrive repository. See next section to sync it from your main PC. We assume now that you navigated to tools/odrive/tests/.

  11. sudo ../../odrivetool udev-setup

  12. sudo udevadm trigger

  13. Run once after every reboot: sudo -E ipython3 --pdb test_runner.py -- --setup-host

SSH Testing Flow

Here’s one possible workflow for developing on the local host and testing on a remote SSH host.

We assume that the ODrive repo is at /path/to/ODriveFirmware and your testing host is configured under the SSH name odrv.

To flash and start remote debugging:

  1. Start OpenOCD remotely, along with a tunnel to localhost:

    ssh -t odrv -L3333:localhost:3333 bash -c "\"openocd '-f' 'interface/stlink-v2.cfg' '-f' 'target/stm32f4x_stlink.cfg'\""

    You can keep this open for multiple debug sessions. Press Ctrl + C to quit.

  2. Compile the firmware.

  3. In VSCode, select the run configuration “Debug ODrive v3.x/v4.x - Remote” and press Run. This will flash the new firmware before dropping you into the debugger.

To run a test:

rsync -avh -e ssh /path/to/ODriveFirmware/ odrv:/opt/odrivetest --exclude="Firmware/build" --exclude="Firmware/.tup" --exclude=".git" --exclude="GUI" --delete
ssh odrv
cd /opt/odrivetest/tools/odrive/tests/
ipython3 --pdb uart_ascii_test.py


If you’re using VSCode, make sure you have the Cortex Debug extension, OpenOCD, and the STLink. You can verify that OpenOCD and STLink are working by ensuring you can flash code. Open the ODrive_Workspace.code-workspace file, and start a debugging session (F5). VSCode will pick up the correct settings from the workspace and automatically connect. Breakpoints can be added graphically in VSCode.

  • Run make gdb. This will reset and halt at program start. Now you can set breakpoints and run the program. If you know how to use gdb, you are good to go.

Setting up an IDE

For working with the ODrive code you don’t need an IDE, but the open-source IDE VSCode is recommended. It is also possible to use Eclipse. If you’d like to go that route, please see the respective configuration document:


This project uses the STM32CubeMX tool to generate startup code and to ease the configuration of the peripherals. You can download it from here <http://www2.st.com/content/st_com/en/products/development-tools/software-development-tools/stm32-software-development-tools/stm32-configurators-and-code-generators/stm32cubemx.html?icmp=stm32cubemx_pron_pr-stm32cubef2_apr2014&sc=stm32cube-pr2>`___. All CubeMX related files are in :code:`Firmware/Board/v3.

You will likely want the pinout for this process. It is available here.

Maintaining Modified Generated Code

When generating the code, STM32CubeMX will nuke everything except some special sections that they provide. These sections are marked like USER CODE BEGIN…`USER CODE END`. We used to try to make sure all edits we made to the generated code would only go in these sections, so some code structrure may reflect that. However over time we realized this will not be tenable, so instead we use git to rebase all changes of the generated code whenever we need to regenerate it. We use two special branches that will help us to do this, they are STM32CubeMX-start and STM32CubeMX-end. How to use these is shown in the following example.


Due to how this rebasing is done, all development that changes the generated code should be done directly on STM32CubeMX-end, and not based on devel, then follow step 4 below to carry them over to your feature branch. If you did some changes to the generated code based from devel, you need to cherry pick just those changes over to STM32CubeMX-end.

  1. Ensuring a clean slate

  • We do all changes to the STM32CubeMX config and regenerate the code on top of STM32CubeMX-start. * git checkout STM32CubeMX-start

  • Run stm32cubeMX and load the Firmware/Board/v3/Odrive.ioc project file. * If the tool asks if you wish to migrate to a new version, choose to download the old firmware package (unless you want to use the latest libraries)

  • Without changing any settings, press Project -> Generate code.

  • You may need to let it download some drivers and such.

  • STM32CubeMX may now have a newer version of some of the libraries, so there may be changes to the generated code even though we didn’t change any settings. We need to check that everything is still working, and hence check in the changes:

  • git config --local core.autocrlf input - This will tell git that all files should be checked in with LF endings (CubeMX generates CRLF endings).

  • git diff - Ignore the pile of line ending warnings.

  • If you feel qualified: you can now ispect if CubeMX introduced something stupid. If there were any changes, and they look acceptable, we should commit them:

    • git commit -am "Run STM32CubeMX v1.21" - Replace with actual version of CubeMX

  1. Making Changes to the STM32CubeMX Config

  • After completing the above steps, make sure the working directory is clean: * git status should include “nothing to commit, working tree clean”

  • Make your changes in STM32CubeMX, save the project and generate the code. (Project -> Generate code)

  • git diff - Check that the introduced changes are as expected

  • If everything looks ok, you can commit your changes.

  1. Rebasing the Modifications to the Generated Code

  • git checkout STM32CubeMX-end

  • git rebase STM32CubeMX-start

  • Make sure the rebase finishes, fixing any conflicts that may arise

  1. Merge New STM32CubeMX Code to your Feature Branch

Simply merge the new state at:code: STM32CubeMX-end into your feature branch. * git checkout your-feature * git merge STM32CubeMX-end

  1. Pushing back Upstream

  • Generate a PR like normal for your feature.

  • Make sure youhave pushed to the STM32CubeMX-start and STM32CubeMX-end branches on your fork.

  • Make a note in your PR to the maintainer that they need to update the STM32CubeMX branches when they merge the PR.


Cannot identify target as a STM32 family when flashing using openocd

Problem: When I try to flash ODrive v4.1 with make flash then I get:

** Programming Started **
auto erase enabled
Info : device id = 0x10006452
Warn : Cannot identify target as a STM32 family.
Error: auto_probe failed
embedded:startup.tcl:487: Error: ** Programming Failed **
in procedure 'program'
in procedure 'program_error' called at file "embedded:startup.tcl", line 543
at file "embedded:startup.tcl", line 487

Solution: Compile and install a recent version of openocd from source. The latest official release (0.10.0 as of Nov 2020) doesn’t support the STM32F722 yet.

sudo apt-get install libtool libusb-1.0
git clone https://git.code.sf.net/p/openocd/code openocd
cd openocd/
./configure --enable-stlink
sudo make install


In Progress

ODrive documentation is now built using Sphinx.

  • Prerequisites:

    • Install Sphinx: pip install -U sphinx

    • Install packages: pip install sphinx-copybutton sphinx-panels sphinx-rtd-theme

  • Run make html within the ./docs/reStructuredText folder

  • Open ./docs/reStructuredText/_build/index.html to view

Modifying libfibre

If you need to modify libfibre add CONFIG_BUILD_LIBFIBRE=true to your tup.config and rerun make. After this you can start odrivetool (on your local PC) and it will use the updated libfibre.

To cross-compile libfibre for the Raspberry Pi, run make libfibre-linux-armhf or make libfibre-all. This will require a docker container. See ;:ref:fibre-cpp readme <../Firmware/fibre-cpp/README.md> for details.

docker run -it -v "$(pwd)":/build -v /tmp/build:/build/build -w /build fibre-compiler configs/linux-armhf.config

If you’re satisfied with the changes don’t forget to generate binaries for all supported systems using make libfibre-all.


We use GitHub Releases to provide firmware releases.

  1. Cut off the changelog to reflect the new release

  2. Merge the release candidate into master.

  3. Push a (lightweight) tag to the master branch. Follow the existing naming convention.

  4. If you changed something in libfibre, regenerate the binaries using make libfibre-all. See Modifying libfibre for details.

  5. Push the python tools to PyPI (see setup.py for details).

  6. Edit the release on GitHub to add a title and description (copy&paste from changelog).

Code Maintenance Notes

The cortex M4F processor has hardware single precision float unit. However double precision operations are not accelerated, and hence should be avoided. The following regex is helpful for cleaning out double constants:





Notes for Contributors

In general the project uses the Google C++ Style Guide with a few exceptions:

  • The default indentation is 4 spaces.

  • The 80 character limit is not very strictly enforced, merely encouraged.

  • The file extensions *.cpp and *.hpp are used instead of *.cc and *.h.

Your help is welcome! However before you start working on a feature/change that will take you a non-negligible amount of time and that you plan to upstream please discuss your plans with us on GitHub or Discord. This will ensure that your implementation is in line with the direction that ODrive is going.

When filing a PR please go through this checklist:

  • Make sure you adhere to the same coding style that we use (see note above).

  • Update CHANGELOG.md.

  • If you removed/moved/renamed things in odrive-interface.yaml make sure to add corresponding bullet points tp the “API migration notes” section in the changelog. Use git to compare against the devel branch.

  • Also, for each removed/moved/renamed API item use your IDE’s search feature to search for occurrences of this name. Update the places you found (this will usually be documentation and test scripts).

  • If you added things to odrive-interface.yaml make sure the new things have decent documentation in the YAML file. We don’t expect 100% coverage but use good sense of what to document.

  • Make sure your PR doesn’t contain spurious changes that unnecessarily add or remove whitespace. These add noise and make the reviewer’s lifes harder.

  • If you changed any enums in odrive-interface.yaml, make sure you update enums.py and ODriveEnums.h. The file includes instructions on how to do this. Check the diff to verify that none of the existing enumerators changed their value.