Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Official Repository for "DrEureka: Language Model Guided Sim-To-Real Transfer" (RSS 2024)
| Date | Stars |
|---|---|
| 2026-07-31 | 938 |
| 2026-08-04 | 938 |
| 2026-08-06 | 938 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# DrEureka: Language Model Guided Sim-to-Real Transfer
<div align="center">
[[Website]](https://eureka-research.github.io/dr-eureka/)
[[arXiv]](https://arxiv.org/abs/2406.01967)
[[PDF]](https://eureka-research.github.io/dr-eureka/assets/dreureka-paper.pdf)
[Yecheng Jason Ma<sup>1*</sup>](https://jasonma2016.github.io/), [William Liang<sup>1*</sup>](https://willjhliang.github.io), [Hung-Ju Wang<sup>1</sup>](https://www.linkedin.com/in/hungju-wang), [Sam Wang<sup>1</sup>](https://www.linkedin.com/in/sam-wang-penn),<br>
[Yuke Zhu<sup>2,3</sup>](https://www.cs.utexas.edu/~yukez/), [Linxi "Jim" Fan<sup>2</sup>](https://jimfan.me/), [Osbert Bastani<sup>1</sup>](https://obastani.github.io/), [Dinesh Jayaraman<sup>1</sup>](https://www.seas.upenn.edu/~dineshj/)
<sup>1</sup>University of Pennsylvania, <sup>2</sup>NVIDIA, <sup>3</sup>University of Texas, Austin
<sup>*</sup>Equal Contribution
[](https://github.com/eureka-research/Eureka)
[<img src="https://img.shields.io/badge/Framework-PyTorch-red.svg"/>](https://pytorch.org/)
[](https://github.com/eureka-research/Eureka/blob/main/LICENSE)
______________________________________________________________________
https://github.com/eureka-research/DrEureka/assets/21993118/d0fd772c-bfbd-4796-8f89-b0553ffb7b80
https://github.com/eureka-research/DrEureka/assets/21993118/0a9825b2-101b-4fb9-878d-4563c6a14090
</div>
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 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 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.
## Installation
This repository contains code for DrEureka's reward generation, RAPP, and domain randomization generation pipelines as well as the forward locomotion and globe walking environments. The two environments are modified from [Rapid Locomotion](https://github.com/Improbable-AI/rapid-locomotion-rl) and [Dribblebot](https://github.com/Improbable-AI/dribblebot), respectively.
The following instructions will install everything under one Conda environment. We have tested on Ubuntu 20.04.
1. Create a new Conda environment with:
```
conda create -n dr_eureka python=3.8
conda activate dr_eureka
```
2. Install Pytorch with CUDA:
```
pip3 install torch==1.10.0+cu113 torchvision==0.11.1+cu113 torchaudio==0.10.0+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html
```
3. Install IsaacGym, the simulator for forward locomotion and globe walking:
1. Download and install IsaacGym from NVIDIA: https://developer.nvidia.com/isaac-gym.
2. Unzip the file:
```
tar -xf IsaacGym_Preview_4_Package.tar.gz
```
3. Install the python package:
```
cd isaacgym/python
pip install -e .
```
4. Install DrEureka:
```
cd dr_eureka
pip install -e .
```
5. Install the forward locomotion and globe walking environments:
```
cd forward_locomotion
pip install -e .
cd ../globe_walking
pip install -e .
```
## Usage
We'llExcerpt of 11,216 characters
Read on GitHub
5
Jason Ma · Dyna Robotics · United States
4
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:7bd5b29f4c3b7e40, llm:Repository title and description: 'DrEureka: Language Model Guided Sim-To-Real Transfer' (RSS 2024) — suggests using language models to guide sim-to-real transfer for robotics; likely agents/robotics and multimodal or reinforcement-learning. No topics provided.
matched fp:7bd5b29f4c3b7e40, llm:Repository title and description: 'DrEureka: Language Model Guided Sim-To-Real Transfer' (RSS 2024) — suggests using language models to guide sim-to-real transfer for robotics; likely agents/robotics and multimodal or reinforcement-learning. No topics provided.
matched fp:7bd5b29f4c3b7e40, llm:Repository title and description: 'DrEureka: Language Model Guided Sim-To-Real Transfer' (RSS 2024) — suggests using language models to guide sim-to-real transfer for robotics; likely agents/robotics and multimodal or reinforcement-learning. No topics provided.
matched fp:7bd5b29f4c3b7e40, llm:Repository title and description: 'DrEureka: Language Model Guided Sim-To-Real Transfer' (RSS 2024) — suggests using language models to guide sim-to-real transfer for robotics; likely agents/robotics and multimodal or reinforcement-learning. No topics provided.