GET-Zero: Graph Embodiment Transformer for Zero-shot Embodiment Generalization

TL;DR

GET-Zero introduces a transformer-based approach that leverages embodiment graphs to enable zero-shot generalization to new hardware configurations in robotic control tasks, without retraining.

Contribution

It presents the Graph Embodiment Transformer (GET) architecture and training method for embodiment-aware control that generalizes to unseen hardware variations without retraining.

Findings

Achieves 20% improvement over baselines in unseen configurations.

Enables zero-shot adaptation to hardware changes using embodiment graphs.

Demonstrates effectiveness on a dexterous robot hand task.

Abstract

This paper introduces GET-Zero, a model architecture and training procedure for learning an embodiment-aware control policy that can immediately adapt to new hardware changes without retraining. To do so, we present Graph Embodiment Transformer (GET), a transformer model that leverages the embodiment graph connectivity as a learned structural bias in the attention mechanism. We use behavior cloning to distill demonstration data from embodiment-specific expert policies into an embodiment-aware GET model that conditions on the hardware configuration of the robot to make control decisions. We conduct a case study on a dexterous in-hand object rotation task using different configurations of a four-fingered robot hand with joints removed and with link length extensions. Using the GET model along with a self-modeling loss enables GET-Zero to zero-shot generalize to unseen variation in graph structure and link length, yielding a 20% improvement over baseline methods. All code and qualitative video results are on https://get-zero-paper.github.io