leggedrobotics

Isaac Gym Environments for Legged Robots

A fast and simple implementation of learning algorithms for robotics.

YOLO ROS: Real-Time Object Detection for ROS

Best practices, conventions, and tricks for ROS

Optimal Control for Switched Systems

Elevation Mapping on GPU.

PACE: A systematic approach for sim-to-real transfer of legged robots, identifying actuator and joint dynamics with standard joint encoders.

Repository for our papers: Robotic World Model: A Neural Network Simulator for Robust Policy Optimization in Robotics and Uncertainty-Aware Robotic World Model Makes Offline Model-Based...

ViPlanner: Visual Semantic Imperative Learning for Local Navigation

Pointcloud-based graph SLAM written in C++ using open3D library.

Pure Pursuit Control and SE(2) Planning

Traversability mapping for mobile rough terrain navigation

An Architecture for the Versatile Control of Legged Robots

iPlanner: Imperative Path Planning. An end-to-end learning planning framework using a novel unsupervised imperative learning approach

Learned Perceptive Forward Dynamics Model

This package provides localization in a pre-built map using ICP and odometry (or the IMU measurements).

Visualization of Motions for Legged Robots in ros-rviz

Wild Visual Navigation: A system for fast traversability learning via pre-trained models and online self-supervision

RAISIM, A PHYSICS ENGINE FOR ROBOTICS AND AI RESEARCH

A graph-based multi-sensor fusion framework. It can be used to fuse various relative or absolute measurments with IMU readings in real-time.

Spatially-Enhanced Recurrent Units (SRU) - PyTorch Implementation

source code and trained models for DeFM (Depth Foundation Model)

Official repository for holistic fusion.

Local Navigation Planner for Legged Robots

Supplementary materials for "Learning Locomotion over Challenging Terrain"

Lightweight implementation of Robotic World Model (RWM) and Uncertainty-Aware Robotic World Model (RWM-U)

Self-supervised Deep LiDAR Odometry for Robotic Applications

Code for "Perceptive Model Predictive Control for Continuous Mobile Manipulation"

Physics engine benchmark for robotics applications: RaiSim vs Bullet vs ODE vs MuJoCo vs DartSim

HOI-Retarget: Contact-Centric Retargeting for Human-Object Interaction

Unity project for OpenXR control of ROS systems

official reimplementation of AME-2

The GrandTour Dataset: A Legged Robotics Dataset in the Wild

Official implementations from the paper "Informed, Constrained, Aligned: A Field Analysis on Degeneracy-aware Point Cloud Registration in the Wild"

Code for Watch Your Step: Anticipated Perception for Humanoid Locomotion, training a humanoid with a foothold prediction model.

SRU Navigation Learning - RL Training Framework

IsaacLab SRU RL Navigation Extension - SRU Project

The official implementation of the DARPA SubT winning LiDAR mapping and odometry solution.

Pre-built TensorFlow for C/C++ and CMake.

Navigation Suite for IsaacLab

Vitruvio is a framework for rapid leg design analysis and optimization for legged robots. The purpose of the simulation framework is to guide the early stages of legged robot design. The end...

Bridging the Gap: Enabling Soft Actor-Critic for High Performance Legged Locomotion

Kleinkram: Open Robotic Data Management

Learning-based localizability estimation for robust LiDAR localization.

Imposition of Hard Convex Constraints on Neural Networks

Robot Deployment - SRU Navigation on Unitree B2W

Accompanying code for the publication "MPC-Net: A First Principles Guided Policy Search"

This package implements a simple tree detector from point cloud data. It makes no assumptions about the ground plane and can handle arbitrary terrains.

Terrain Grounded Imagination: a recipe for training humanoid locomotion on sparse and rough terrain through a world model with differentiable heightmap lookups.

Datasets collected by Team CERBERUS during the DARPA Subterranean Challenge

https://rsl.ethz.ch/partnership/spinoff/raisim.html

ros2 version of the robot_self_filter package.

Depth Perception Pretraining - SRU Project

Reinforcement learning framework from RSL for policy training with RaiSim.

Implementation of Radiance Fields for Robotic Teleoperation

Fast, orientation-aware trajectory planning using a novel Gaussian overlap-based collision formulation, modeling both robot and environment as Gaussian Splat.

Image processing pipeline for cameras that provide raw data

A Reinforcement Learning Software Toolbox for Robotics

Guidelines on how to hardware synchronize the time of multiple sensors, e.g., IMU, cameras, etc.