# AUV-RSControl-Code

**Repository Path**: linyunfeng201203/AUV-RSControl-Code

## Basic Information

- **Project Name**: AUV-RSControl-Code
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-01-06
- **Last Updated**: 2026-01-06

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# [IROS 2025] Never too Prim to Swim

[![arXiv](https://img.shields.io/badge/arXiv-2503.00527-b31b1b.svg)](https://arxiv.org/abs/2503.00527)  [![WebSite](https://img.shields.io/badge/Github_Page-Supp_Material-77DDFF.svg)](https://360zmem.github.io/AUV-RSControl/) [![WebSite](https://img.shields.io/github/last-commit/360ZMEM/AUV-RSControl-Code?color=green)](https://360zmem.github.io/AUV-RSControl-Code)

This repository contains the Simulator and Supplementary Materials for the paper "Never too Prim to Swim: An LLM-Enhanced RL-based Adaptive S-Surface Controller for AUVs under Extreme Sea Conditions".

:warning: Work-in-Progress Notice: This repository contains a code implementation for demonstration purposes.

<img src="attachments/intro.png" alt="intro" style="zoom:50%;" />


<img src="attachments/intro2.gif" alt="intro" style="zoom:100%;" />

## Get Started

Run this command to install the dependencies:

```bash
pip install -r requirements.txt
```

Important: The visualization module requires the H.264 codec. Download the [openh264](https://github.com/cisco/openh264) binary and place it in the project’s root directory.

Next, install our modified Python Vehicle Simulator (PVS), which includes our S-surface controller implementation:

```bash
cd PVS/
pip install -e .
```

We provide implementations for two high-level underwater tasks described in the paper. The code resides in `data_collection/`.

## 3D Data Collection Task Simulation

### Training

Execute training with:

```bash
python train_td3.py --wave 0.5 --usbl
```

- `--wave`: JONSWAP wave amplitude. (Default: 0, namely disable)
- `--usbl`: Enables USBL positioning noise (disable to use ground-truth positions).

Additional hyperparameters (e.g., for control, simulation, TD3 algorithm) can be adjusted via command-line arguments. Refer to the source code for details.

### Evaluation

Run inference with a trained model:

```bash
python eval_td3.py
```

Arguments follow the same definitions as above.

### Outputs  
Both phases generate visualization files and logs:  
- **Training**:  
  - Visualization records: `record_train/yyyy_mm_dd_hh_mm_ss.pkl` (timestamped).  
  - Logs: Saved to `logs/`.  
- **Evaluation**:  
  - Visualization records: `record_eval/yyyy_mm_dd_hh_mm_ss.pkl`.  


### Visualization Scripts  
- **2D Visualization**:  
  ```bash  
  python genvideo_2D.py --preview -1 --fname XXX --mode train --speed 5  
  ```
  - `--preview N`: Previews `N`-th frames (negative values skip preview).  
  - `--fname`: Input filename (without `.pkl`).  
  - `--speed`: Playback speed multiplier.  
  Output: `output_2D.mp4` (H.264/AVC, 10 Mbps). Use FFmpeg for further format conversion.  

- **3D Visualization**:  
  ```bash  
  python genvideo_3D.py --preview -1 --fname XXX --mode train --speed 5  
  ```
  Parameters match the 2D script.  

## Acknowledgement

The 6-DOF simulator is adapted from the [Python Vehicle Simulator](https://github.com/cybergalactic/PythonVehicleSimulator).

## Citation

If you find the code repository useful for your work, please cite:

```bibtex
@article{xie2025AUVRScontrol,
      title={Never too Prim to Swim: An LLM-Enhanced RL-based Adaptive S-Surface Controller for AUVs under Extre Sea Conditions},
      author={Xie, Guanwen and Xu, Jingzehua and Ding, Yimian and Zhang, Zhi and Zhang, Shuai and Li, Yi},
      journal={arXiv preprint arXiv:2503.00527},
      year={2025}}
```