EPIC: Exploring on Point Clouds

A Lightweight LiDAR-Based AAV Exploration Framework for Large-Scale Scenarios

Shuang Geng^*, Zelin Ning^*, Fu Zhang, and Boyu Zhou^†

^*Equal Contribution;  ^†Corresponding Author

💡 News

• [2025.03.28] EPIC is officially published by RAL !
• [2025.03.12] The source code of EPIC is released !
• [2025.03.08] EPIC is accepted by RAL 2025 🚀 !

📜 Introduction

EPIC (Exploring on PoInt Clouds) is a lightweight LiDAR-based AAV (Autonomous Aerial Vehicle) exploration framework that directly exploits point cloud data to explore large-scale environments. Experimental results demonstrate that our framework achieves faster exploration while significantly reducing memory consumption. (Click the image to view the video)

Please cite our paper if you use this project in your research:

@ARTICLE{10945408,
  author={Geng, Shuang and Ning, Zelin and Zhang, Fu and Zhou, Boyu},
  journal={IEEE Robotics and Automation Letters}, 
  title={EPIC: A Lightweight LiDAR-Based AAV Exploration Framework for Large-Scale Scenarios}, 
  year={2025},
  volume={10},
  number={5},
  pages={5090-5097},
  keywords={Point cloud compression;Autonomous aerial vehicles;Memory management;Laser radar;Trajectory;Surface treatment;Real-time systems;Planning;Navigation;Faces;Aerial systems: perception and autonomy;motion and path planning;aerial systems: applications},
  doi={10.1109/LRA.2025.3555878}}

Please kindly star ⭐️ this project if it helps you. We take great efforts to develop and maintain it 😁.

🛠️ Installation

Test Environment

• Ubuntu 20.04
• ROS Noetic
• C++17

🚀 Quick Start

Clone our repository and build

git clone https://github.com/SYSU-STAR/EPIC.git
cd EPIC 
catkin build

Download dataset

Download simulation maps from my Google cloud, create the folder MARSIM/map_generator/resource if it doesn't exist, and move the downloaded maps to this folder.

mkdir -p MARSIM/map_generator/resource
mv /path/to/downloaded/maps/*.pcd MARSIM/map_generator/resource/

Run program

source ./devel/setup.zsh && roslaunch epic_planner garage.launch

You can replace garage with other maps. We provide three test scenarios: garage, cave and factory.

Our simulation environment is developed based on the GPU version of MARSIM. So if you don't have a GPU, you may need to make some necessary modifications to the simulator.

⚠️ Known Issues

• After launching the program, if the terminal keeps displaying a no odom warning, this is likely due to graphics card compatibility issues. Please refer to this issue for solutions.

🤓 Acknowledgments

We would like to express our gratitude to the following projects, which have provided significant support and inspiration for our work:

• GCOPTER: A general-purpose trajectory optimizer for multicopters, our local planner is based on it.
• FUEL: An efficient framework for fast UAV exploration from which our global planner draws inspiration.
• MARSIM: A lightweight point-realistic simulator for LiDAR-based UAVs upon which our simulator is built.
• FALCON: An efficient framework for fast UAV exploration, from which our method for constructing topological maps draws inspiration.