RT-1-X is a Transformer-based foundation model for end-to-end robotic control, pretrained on demonstration data from over 20 different robots. While the code is open sourced, no examples of running inference with an actual robot is given. This repository contains a ROS2 package that makes using the model much more accesible, as all required components are included and ready-to-run.
This package was developed in the course of my Master's Thesis with the title "Bringing the RT-1-X Foundation Model for Robotic Control to New Embodiments" where I explored the potential of adapting the model to a UMI robotic arm at the Intelligent Robotics Lab of University of Amsterdam. Along with this, a fine tuning pipeline for the RT-1-X model was developed.