Official repo for GraspGen: A Diffusion-based Framework for 6-DOF Grasping (Forked Repo to Be Tested at IITGN Robotics Lab x Tohoku Univeristy)

A digital data-generation pipeline that synthesizes humanoid loco-manipulation data from 3D assets and video priors.

Model Predictive Path Integral (MPPI) with approximate dynamics implemented in pytorch

EllipseLIO is an Adaptive LiDAR Inertial Odometry Approach with an Ellipsoid Representation

GPU-accelerated robotics (CUDA + C++)

GPU-accelerated 6-DOF robot arm kinematics with ROS 2 with batch FK/IK on CUDA kernels

OMPL workshop code

Python implementation of MPPI (Model Predictive Path-Integral) controller to understand the basic idea. Mandatory dependencies are numpy and matplotlib only.

This repository contains the preprint of the book "Riemannian Manifolds in Robot Learning, Optimization, and Control"

Multi-Purpose MPC for Reference Path Tracking, Time-Optimal Driving and Obstacle Avoidance

SDPG: Self-Distilled Policy Gradient

Web-based 3D visualization in Python

Dexbotic: Open-Source Vision-Language-Action Toolbox

Various retargeting optimizers to translate human hand motion to robot hand motion.

multi-agent deep reinforcement learning for networked system control.

Official implementation for paper "EquiBot: SIM(3)-Equivariant Diffusion Policy for Generalizable and Data Efficient Learning".

[CoRL 2024 Outstanding Paper Award Finalist] Equivariant Diffusion Policy

[ICLR 2023] Equivariant Descriptor Fields: SE(3)-Equivariant Energy-Based Models for End-to-End Visual Robotic Manipulation Learning

This repository contains the code of the paper SO(2)-Equivariant Reinforcement Learning (ICLR 2022) and On-Robot Learning With Equivariant Models (CoRL 2022).

The code will soon be open source.

using OSQP to solve LMPC problem

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

Fast Robot Kinematics and Dynamics in JAX

Pedestrian simulator powered by the social force model

Sumo: Dynamic and Generalizable Whole-Body Loco-Manipulation

Reinforcement learning on general 2D physics environments in JAX. ICLR 2025 Oral.