# Overview

[Deoxys](https://github.com/UT-Austin-RPL/deoxys_control) is a modularized, real-time controller library for Franka Emika Panda to facilitate robot learning research. `Deoxys` is developed by [Yifeng Zhu](https://zhuyifengzju.github.io/) and aims to democratize basic knowledge of robot manipulation to the robot learning community through open-sourcing the controller implementation.

<img src="../_images/infra_overview.svg"/>


Here are a list of features that we identified as strengths of our
library.

&#10004; A user-friendly [Python Interface](https://github.com/UT-Austin-RPL/deoxys_control/tree/main/deoxys/deoxys) and [real-time controller implementation](https://github.com/UT-Austin-RPL/deoxys_control/tree/main/deoxys/franka-interface/src/controllers) in C++<br>
&#10004; Specialized for research on closed-loop [visuomotor skill learning](https://ut-austin-rpl.github.io/deoxys-docs/html/project.html)<br>
&#10004; Easy [configuration](https://github.com/UT-Austin-RPL/deoxys_control/tree/main/deoxys/config) of robot controllers<br>
&#10004; Seamless transfer from [robosuite](https://robosuite.ai/) for real-robot control<br>

<!-- This is a real robot control system that is designed to be amenable to people whose research directions lie at the intersection between robot learning and visuomotor manipulation control. Before beginning with this repo, here are the take-aways for you to decide whether to use the repo or not. -->

<!-- (TODO) -->


<!-- Now if you are here, then probably you want to use our codebase! That -->
<!-- is great. Now you will go through some hacky things for setting up the -->
<!-- hardware such as Desktops, NUCs, and most important of all, the -->
<!-- robot. But first, let's take a look at the overall workflow. -->


## Past research

<div id="p1">

A number of [projects](../project.html) have been boosted by
`Deoxys`. Here is a list of publications based on `Deoxys` so far.

</div>

- [VIOLA: Imitation Learning for Vision-Based Manipulation with Object
Proposals Priors](https://arxiv.org/abs/2210.11339) Yifeng Zhu,
Abhishek Joshi, Peter Stone, Yuke Zhu
- [Bottom-Up Skill Discovery from Unsegmented Demonstrations for
  Long-Horizon Robot Manipulation](https://arxiv.org/abs/2109.13841)
  Yifeng Zhu, Peter Stone, Yuke Zhu
- [Robot Learning on the Job: Human-in-the-Loop Manipulation and
  Learning During Deployment](https://arxiv.org/abs/2211.08416)  Huihan Liu, Soroush Nasiriany, Lance Zhang, Zhiyao Bao, Yuke Zhu 
- [Augmenting Reinforcement Learning with Behavior Primitives for
  Diverse Manipulation Tasks ](https://arxiv.org/abs/2110.03655)
  Soroush Nasiriany, Huihan Liu, Yuke Zhu 
- [Learning and Retrieval from Prior Data for Skill-based Imitation
  Learning](https://arxiv.org/abs/2210.11435)  Soroush Nasiriany, Tian
  Gao, Ajay Mandlekar, Yuke Zhu 
- [Synergies Between Affordance and Geometry: 6-DoF Grasp Detection
  via Implicit Representations](https://arxiv.org/pdf/2104.01542.pdf)
  Zhenyu Jiang, Yifeng Zhu, Maxwell Svetlik, Kuan Fang, Yuke Zhu. 
- [Ditto: Building Digital Twins of Articulated Objects from
  Interaction](https://arxiv.org/abs/2202.08227)  Zhenyu Jiang,
  Cheng-Chun Hsu, Yuke Zhu 
- [ACID: Action-Conditional Implicit Visual Dynamics for Deformable
  Object Manipulation](https://arxiv.org/abs/2203.06856) Bokui Shen,
  Zhenyu Jiang, Christopher Choy, Leonidas J. Guibas, Silvio Savarese,
  Anima Anandkumar, Yuke Zhu

## Reference

If you find this repo useful for your research, please cite us through
the following work:

```bibtex
@article{zhu2022viola,
  title={VIOLA: Imitation Learning for Vision-Based Manipulation with Object Proposal Priors},
  author={Zhu, Yifeng and Joshi, Abhishek and Stone, Peter and Zhu, Yuke},
  journal={6th Annual Conference on Robot Learning},
  year={2022}
}
```