Donkeycar
- Software
- Cars
- Framework
- Runs on hardware you can build
Donkeycar is a minimalist Python library for building and experimenting with self-driving robot cars, aimed at hobbyists, students and researchers.
Source on GitHub: github.com/autorope/donkeycar (opens in a new tab). Facts read at commit ba25266 (opens in a new tab).
Licenses
- Code
- MIT (opens in a new tab)
Read by Robotheca from LICENSE (opens in a new tab) (MIT) at commit ba25266 (opens in a new tab), on 16 Oct 2026.
- Commercial use
- Yes
What it does
- Lets you build a small robot car and teach it to drive itself, starting from configurable templates.
- Organizes a car's behavior as a pipeline of software parts that run in order on each pass of the vehicle loop.
- Supports many cameras (including 3D cameras and lidar), GPS receivers, game and RC controllers, and motor drivetrains.
- Records telemetry such as camera images and steering and throttle inputs for training.
- Provides three autopilot types: deep learning (TensorFlow, TensorFlow Lite, PyTorch), GPS path following, and computer vision.
- Includes a web UI, an onscreen touch controller, and a simulator with online virtual race league support.
Written by Robotheca from the project's README (opens in a new tab) at commit ba25266.
Getting started
- Difficulty
- Beginner
- Hardware it needs, as the project states it
- It requires a built car with a camera and motors, though a simulator can be used before building a robot.
- How to start
- The project's README (opens in a new tab)
Maintenance
- Status
- Active: a commit in the last 90 days
- Newest commit
- 17 Aug 2026 on main
- Latest release
- Donkey Car 5.4.0 (22 Aug 2026)
Checked on GitHub on 16 Oct 2026.
Links
Robots on Robotheca that use it
No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.
Robotheca lists software as a reference: nothing is hosted or run here, and every fact links to where it was read.