Stanford Interactive Perception and Robot Learning Lab

torchfilter

Bayesian filters in PyTorch

multimodal_representation

UniGrasp

Implementation of " UniGrasp: Learning a Unified Model to Grasp with Multifingered Robotic Hands"

GRAC

implementation of our self-guided and self-regularized actor-critic algorithm

sceneflownet

Motion-based Object Segmentation based on Dense RGB-D Scene Flow

Concept2Robot

simulations used in "Concept2Robot: Learning Manipulation Concepts from Instructions and Human Demonstrations"

nerf_shared

A object-oriented, minimalistic PyTorch implementation of NeRF (Neural Radiance Fields).

ao-grasp

franka-panda-iprl

IPRL's Franka Panda Driver and Operational Space Controller

ScaffoldLearning

Simulation Environments in "Learning to Scaffold the Development of Robotic Manipulation Skills"

coding-practices-tutorial

Tips for writing readable and maintainable code

baselines

OpenAI Baselines: high-quality implementations of reinforcement learning algorithms

FrankaPanda

model description and driver for the Panda arm

rrc_simulation

Simulation for the Real Robot Challenge (https://real-robot-challenge.com)

zeddy

contact_graspnet

Efficient 6-DoF Grasp Generation in Cluttered Scenes

ctrl-utils

dbot

Depth-Based Bayesian Object Tracking Library

fl

A Lightweight Bayesian Filtering Library with real-time support

mujoco-py

MuJoCo is a physics engine for detailed, efficient rigid body simulations with contacts. mujoco-py allows using MuJoCo from Python 3.

rrc_package

Example package for the Real Robot Challenge Submission System