An Accessible Platform for Mobile, Multi-Agent Robotics.
The Multi-agent System for non-Holonomic Racing (MuSHR) is an open-source robotic hardware and software platform for learning and researching AI in a the setting of autonomous vehicles and mobile robotics.
Operate your own vehicle in simulation
Quickstart
Run the MuSHR platform on your machine!
Intro to ROS
Fundamental Robot Operating System (ROS) concepts using MuSHR.
MuSHR System Overview
Basic components of the car and their associated software counterparts.
Car Pose Detection and Yolo Learning
Train a model to detect MuSHR cars.
MuJoCo Simulation
Run a MuSHR simulation with the MuJoCo physics engine.
Applying Deep Learning to Autonomous Driving
A basic tutorial on using Imitation learning and Reinforcement learning for driving the MuSHR car
Rally Chassis
A powerful 1/10 scale rally car chassis provides four wheel drive, brushed DC motor, and ESC suitable for most speeds (0-30 mph).
Budget Friendly
The base MuSHR car with no sensing has a bill-of-materials (BOM) cost of $610, while a fully equipped car costs around $930.
Simulation
Our ROS-based simulator let's you prototype your system without needing to turn on the robot, saving you valuable time when you do.
LiDAR
Integration with the YDLiDAR X4 provides a 360 degree FOV at 10hz, all in a radically economic package.
IMU
Learn the principles of sensor fusion or test your latest SLAM algorithm using the Razor 9dof IMU or any other of your choice.
RGBD Camera
MuSHR uses the Intel Realsense line of RGBD cameras to provide point clouds for tasks such as obstacle avoidance or object tracking.
Classroom Tested
We have refined the MuSHR platform through two undergraduate and two graduate robotics courses, distilling lessons learned into the V3 roboracer. From these endeavors, we offer ready-to-go labs and assignments, as well as sample curricula and slides that were well-received in the recent undergraduate level robotics course "Mobile Robots".
| Winter, 2020 | Undergraduate | Autonomous Robots |
| Spring, 2019 | Undergraduate | Mobile Robots |
| Winter, 2019 | Graduate | Robotics: Algorithms and Applications |
| Autumn, 2018 | Graduate | Introduction to AI for Mobile Robots |
| Winter, 2018 | Undergraduate | Robotics |
Citation
If you plan to use any part of the the MuSHR platform (including tutorials, codebase, or hardware instructions) for a project or paper, please cite MuSHR: A Low-Cost, Open-Source Robotic Racecar for Education and Research.
@article{srinivasa2019mushr,
title={{MuSHR}: A Low-Cost, Open-Source Robotic Racecar for Education and Research},
author={Srinivasa, Siddhartha S. and Lancaster, Patrick and Michalove, Johan and Schmittle, Matt and Summers, Colin and Rockett, Matthew and Smith, Joshua R. and Chouhury, Sanjiban and Mavrogiannis, Christoforos and Sadeghi, Fereshteh},
journal={CoRR},
volume={abs/1908.08031},
year={2019}
}