YOPO-Rally - GitLite

KO | EN

gitlite — search

#javascript #python #react #ai #llm #go #android #deep-learning #nodejs #windows #awesome #cli #pytorch #open-source

YOPO-Rally

(★ 63)

A Sim-to-Real Single-Stage Planner for Off-Road Terrain

파일 탐색기

최종 버전 다운로드 (.zip)

- cost-map-generation.mp4
- data-acquisition.jpg
- inference.mp4
- system-overview.svg
- trajectory-optimization.mp4
- yopo-sim.jpg
- 01-system-overview.md
- 01-系统概述.md
- 02-installation-and-environment.md
- 02-安装与环境.md
- 03-simulation-and-inference.md
- 03-仿真推理.md
- 04-data-training-pipeline.md
- 04-数据训练流水线.md
- 05-real-world-deployment.md
- 05-实车部署.md
- 06-faq.md
- 06-常见问题.md
- INDEX.md
- INDEX_en.md
- - pyproject.toml
  - README.md
- record_all.sh
- record_stereo.sh
- - __init__.py
  - configs.py
  - eval_network.py
  - generate_data.py
  - train_network.py
  - visualize_data.py
- - diablo_ros2
  - ROS-TCP-Endpoint
.gitignore
.gitmodules
.python-version
pyproject.toml
README.md
README_zh.md
requirements.txt
uv.lock

# 설치 가이드

1. 코드 내려받기

git clone https://github.com/TJU-Aerial-Robotics/YOPO-Rally

깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.

cd YOPO-Rally

방금 내려받은 프로젝트 폴더 안으로 이동합니다.

2. Python

쉬움 추천

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Python 3 설치 시 'Add Python to PATH' 옵션을 꼭 체크하세요 (Windows).

pip install -r requirements.txt

requirements.txt 등에 명시된 파이썬 라이브러리를 설치합니다.

python <실행할 파일명>.py # README에서 정확한 실행 파일명을 확인하세요

파이썬 스크립트(또는 모듈)를 실행합니다.

✅ 에러 메시지 없이 실행되고 터미널에 안내 문구가 출력되면 정상입니다.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# YOPO-Rally

A single-stage off-road terrain planner based on imitation learning, supporting Zero-Shot Sim-to-Real transfer.

Video Demos: [YouTube](https://youtu.be/dyoufaKgVa0) · [Bilibili](https://www.bilibili.com/video/BV1D1V7zQEWr)

[English](README.md) | [中文](README_zh.md)

## System Overview

![System Overview](https://raw.githubusercontent.com/TJU-Aerial-Robotics/YOPO-Rally/main/.media/system-overview.svg)

The planner takes the depth map and vehicle state as inputs, and directly outputs multiple candidate trajectories along with their cost scores. The MPC local planner then tracks the optimal trajectory. Training is conducted entirely in the simulation environment (YOPO-Sim), where expert trajectory labels are generated through terrain cost maps, achieving zero-shot transfer from simulation to real world.

For detailed architecture, see [System Overview](docs/01-system-overview.md).

## Simulator

We use [YOPO-Sim](https://github.com/TJU-Aerial-Robotics/YOPO-Sim.git) as the simulation and data collection environment.

![YOPO-Sim](https://raw.githubusercontent.com/TJU-Aerial-Robotics/YOPO-Rally/main/.media/yopo-sim.jpg)

## Imitation Learning

### Cost Map Generation

The terrain point cloud is analyzed for Terrain Type Analysis (TTA) using CloudCompare, extracting slope and roughness features to generate a terrain traversability cost map in GeoTIFF format.

https://github.com/user-attachments/assets/9114d64a-cec4-42c8-93c3-7d9c15ee0e10

### Data Acquisition

Record depth maps, vehicle position, and velocity data in the simulator.

![Data Acquisition](https://raw.githubusercontent.com/TJU-Aerial-Robotics/YOPO-Rally/main/.media/data-acquisition.jpg)

### Trajectory Optimization

For each reference starting point, CasADi is used to solve multiple candidate Hermite curve trajectories under sector constraints. Terrain costs are queried from the cost map to generate expert trajectory labels.

https://github.com/user-attachments/assets/d9902d38-5c5b-4b8a-aec8-42a6decb11fb

## Inference

The planner takes the depth map, velocity, and goal vector as inputs, outputting a set of candidate trajectories with cost scores. The MPC node tracks the optimal trajectory.

https://github.com/user-attachments/assets/a4583249-a9d8-4ef8-8d7d-87dd9a78e298

## Quick Start

**Requirements:** ROS2 Humble · Python 3.10 · `uv` · CloudCompare 2.13+

```bash
# 1. Clone the repository
git clone https://github.com/TJU-Aerial-Robotics/YOPO-Rally.git --recursive

# 2. Create virtual environment and install dependencies
uv venv --system-site-packages
uv sync

# 3. Build ROS2 packages (DO NOT use --symlink-install)
source .venv/bin/activate
colcon build && source install/setup.bash

# 4. Launch inference node (requires YOPO-Sim running)
ros2 launch diablo_ros_wrapper diablo_sim.launch.py
```

After launching, load the `Assets\Scenes\EvaluationScene.unity` scene in the simulator, click **Connect** in the simulator, and then trigger planning by publishing `/initialpose (geometry_msgs/msg/PoseWithCovarianceStamped)` topic.

*Note: Please download the pre-trained weights from the **Release** page and place them in the `models/` directory before running.*

For full installation steps, see [Installation and Environment](docs/02-installation-and-environment.md).

## Training Pipeline

1. Collect depth maps and pose data in YOPO-Sim
2. Process the point cloud into a terrain cost map (GeoTIFF) using CloudCompare
3. Run `generate_data.py` to generate expert trajectory labels
4. Run `train_network.py` to train the network

For detailed steps, see [Data Training Pipeline](docs/04-data-training-pipeline.md).

## Real-World Deployment

For detailed instructions, see [Real-World Deployment](docs/05-real-world-deployment.md).

## Documentation

| Document | Content |
|------|------|
| [System Overview](docs/01-system-overview.md) | Architecture, directory structure, network I/O formats |
| [Installation & Environment](docs/02-installation-and-environment.md) | Complete installation steps |
| [Simulation & Inference](docs/03-simulation-and-inference.md) | Running jointly with YOPO-Sim |
| [Data Training Pipeline](docs/04-data-training-pipeline.md) | Data acquisition → Cost map → Trajectory optimization → Training |
| [Real-World Deployment](docs/05-real-world-deployment.md) | ROS nodes, sensor configuration |
| [FAQ](docs/06-faq.md) | Frequently Asked Questions |