DrEureka: Language Model Guided Sim-To-Real Transfer
This addresses the problem of slow and human-intensive sim-to-real transfer for robotics researchers and practitioners, offering an incremental improvement by automating design steps.
The paper tackles the labor-intensive process of manually designing reward functions and simulation parameters for sim-to-real transfer in robotics by introducing DrEureka, an LLM-guided approach that automates this design. The result shows it can achieve competitive performance on existing tasks and solve novel tasks like quadruped balancing on a yoga ball without manual iteration.
Transferring policies learned in simulation to the real world is a promising strategy for acquiring robot skills at scale. However, sim-to-real approaches typically rely on manual design and tuning of the task reward function as well as the simulation physics parameters, rendering the process slow and human-labor intensive. In this paper, we investigate using Large Language Models (LLMs) to automate and accelerate sim-to-real design. Our LLM-guided sim-to-real approach, DrEureka, requires only the physics simulation for the target task and automatically constructs suitable reward functions and domain randomization distributions to support real-world transfer. We first demonstrate that our approach can discover sim-to-real configurations that are competitive with existing human-designed ones on quadruped locomotion and dexterous manipulation tasks. Then, we showcase that our approach is capable of solving novel robot tasks, such as quadruped balancing and walking atop a yoga ball, without iterative manual design.