Train quadruped locomotion using reinforcement learning in Mujoco

A ROS2 package to control the Unitree GO1 robot

This is the official implementation of the ICRA 2025 paper TANGO: Traversability-Aware Navigation with Local Metric Control for Topological Goals

A training and deployment pipeline for visuomotor RL policy used to enhance the locomotion of Unitree Go1 robot dog.

A Fleet of Unitree GOs (GO1 and GO2) capable of Heterogeneous Collaborative Exploration.

go1pylib is a Python library designed to control the Go1 robot by Unitree Robotics. It provides an easy-to-use interface for robot movement, state management, collision avoidance, battery monitoring, and MQTT communication. Ideal for research and robotics development. https://pypi.org/project/go1pylib/

Reinforcement learning with Flax NNX

Near-complete Unitree Go1 Pro backup (Pi + both Jetson Nanos, sanitized — no secrets). MISSING the head Nano .13, which is dead on my unit so my front cameras don't work. 🙏 PLEASE — if you have a working Go1 with a live .13 Nano, send me a backup/image of it (issue or PR). Not even sure it's recoverable, but a reference would mean a lot.

PPO-based blind locomotion for the Unitree Go1 quadruped using proprioceptive sensing. Dual MuJoCo/Genesis backend with ~85x GPU training speedup and 1.41 m/s learned gait.

📅 Scheduling Appointments for the Go-1 🤖 Web App.

Explore an innovative undergraduate research project featuring a Unitree robot dog that communicates through speech, dances, and navigates using Python and the Google Text-to-Speech (gTTS) library.

AI-powered surveillance system built on the Unitree Go1 robot dog

This project showcases various functionalities for the Go-1 🤖, including voice interaction, Wi-Fi setup, movement, and navigation.

This project aims to enhance the robot's ability to associate its surroundings with objects, thereby imbuing them with meaning, and subsequently applying SLAM (Simultaneous Localization and Mapping) techniques.

A Ros2 interface to the Unitree Go1 robot and its SDK's

The repo for the Master's Thesis "Verbal robot dog training"

Official repository for the IEEE paper "Distributed Hybrid Control Architecture for Quadrupeds in Human-Robot Interaction Tasks". Features ROS Noetic, YOLOv8 segmentation, Gazebo simulation, and reactive obstacle avoidance for the Unitree Go1.

A three-tier learning stack that turns a simulated Unitree Go1 quadruped into something that acts like a cat.