SePT_Stereo-Planetary-Tracks

This repo releases a dataset to challenge object SLAM algorithms in an unstructured, weakly textured, and lunar terrain environment. All the data are collected from the Unreal Engine 4 with AirSim.

1. Simulation Engine:

• Unreal Engine Version: 4.27.2
• AirSim Sensors: Stereo Cameras (150 mm baseline), Lidar (128 channels), IMU
• Setting Files: Settings for lunar rover exploration
• Offical Guidelines: AirSim

2. Dataset Download Link:

• SePT01 track: Google Drive
• SePT02 track: Google Drive
• SePT03 track: Google Drive
• SePT04 track: Google Drive

3. Structure:

Similar to the Euroc dataset, we structured our SePT dataset which can be tested in a typical SLAM framework (e.g., ORB-SLAMx, CubeSLAM).

4. Acknowledgement:

Thanks to work: Image-Matching-Toolbox, SAM, and EAO-SLAM.

• Q. Zhou, T. Sattler and L. Leal-Taixé, "Patch2Pix: Epipolar-Guided Pixel-Level Correspondences," CVPR 2021, Nashville, TN, USA, 2021, pp. 4667-4676. Paper.
• Kirillov, Alexander and Mintun, Eric and Ravi, Nikhila and et al., "Segment Anything," 2023, CoRR. paper.
• Y. Wu, Y. Zhang, D. Zhu, Y. Feng, S. Coleman and D. Kerr, "EAO-SLAM: Monocular Semi-Dense Object SLAM Based on Ensemble Data Association," IROS 2020, Las Vegas, NV, USA, 2020, pp. 4966-4973. Paper.

5. Contact:

• Author: Yaolin Tian (email: [email protected])
• The paper has been accepted by TGRS

@article{Yaolin2025,
  title={Lo-SLAM: Lunar Target-oriented SLAM Using Object Identification, Relative Navigation and Multi-level Mapping},
  journal={IEEE Transactions on Geoscience and Remote Sensing}, 
  author={Yaolin Tian, Xue Wan, Shengyang Zhang, Jianhong Zuo, Yadong Shao, Baichuan Liu, and Mengmeng Yang},
  year={2025},
  doi={10.1109/TGRS.2025.3547292}
}