This repository contains datasets collected using AgileX mobile robots in different scenarios of the University of Udine (Italy).
Data acquisition is performed with the robot moving autonomously following a predefined path (unless otherwise stated). More details on the navigation method used are reported on the related paper.
These datasets can be exploited to test SLAM algorithms based on LiDAR, IMU, and GNSS data.
LiDAR sensor: Velodyne VLP16 (Puck) datasheet
9-Axis IMU: Xsens MTi 630 datasheet
GNSS receiver: Ardusimple simpleRTK2B Budget datasheet
Info: The following information applies to everything below unless otherwise stated! The Velodyne LiDAR sensor range is set to 100 m and LiDAR point clouds are published at 10 Hz in the topic velodyne_points. The IMU sensor reference frame is considered coincident with the LiDAR sensor reference frame, i.e., the
Mobile robot: AgileX Scout Mini
Mobile platform specifications: https://docs.trossenrobotics.com/agilex_scout_mini_docs/specifications.html
Data acquired in a single corridor of the west wing of the Rizzi building (40 m long, 8 m wide, and 4 m high).
Download dataset at: download_link
The TLS ground truth is available at: download_link
Data acquired in the whole squared plant of the west wing of the Rizzi building (80 × 80 m, measured along the center line of the corridors). During the survey new and unknown portions of the building are explored.
Download dataset at: download_link
Relevant topics list:
/imu/data
/velodyne_points
Thank you for citing Tiozzo Fasiolo, D et al. (2023) if you use any of this data.
@article{fasiolo2023comparing,
title={Comparing LiDAR and IMU-based SLAM approaches for 3D robotic mapping},
author={Fasiolo, Diego Tiozzo and Scalera, Lorenzo and Maset, Eleonora},
journal={Robotica},
pages={1--17},
year={2023},
publisher={Cambridge University Press}
}
Mobile robot: AgileX Scout 2.0
Mobile platform specifications: https://docs.trossenrobotics.com/agilex_scout_20_docs/specifications.html
Data acquired in one of the courtyards inside the building (45 × 45 m, measured along the center line of the external corridors). The robot moves on grass and tiles cousing vibrations.
Download dataset at: download_link
Relevant topics list:
/imu/data
/velodyne_points
The TLS ground truth is available at: download_link
Thank you for citing Tiozzo Fasiolo, D et al. (2023) if you use any of this data.
@article{fasiolo2023comparing,
title={Comparing LiDAR and IMU-based SLAM approaches for 3D robotic mapping},
author={Fasiolo, Diego Tiozzo and Scalera, Lorenzo and Maset, Eleonora},
journal={Robotica},
pages={1--17},
year={2023},
publisher={Cambridge University Press}
}
Mobile robot: AgileX Bunker
Mobile platform specifications: https://docs.trossenrobotics.com/agilex_bunker_docs/specifications.html
The robot performs a complete scan of the rectangular Lab Village building (dimension 160 × 60 m), in the presence of walking workers and a few moving vehicles, covering a total distance of 496 m and returns to its initial position at the end of the survey.
Download dataset at: data will be available as soon as the paper is published :(
The robot performs two laps along a rectangular path for a total distance of 142 m, while two people walk in and out of the building through two entrance doors and another person walks back and forth in the parking lot.
Download dataset at: data will be available as soon as the paper is published :(
Topics list:
/imu/data
/tf
/ublox_position_receiver/fix
/velodyne_points
Info: GNSS data are corrected by means of a NTRIP client. Be careful! Poses published in the tf topic are also available in this dataset. Do not play that topic if you are using another localization algorithm. Use for instance:
rosbag play [bagfilename] --topic [topicname1] [topicname2]
Mobile robot: AgileX Scout 2.0
Mobile platform specifications: https://docs.trossenrobotics.com/agilex_scout_20_docs/specifications.html
Download June dataset at: download_link
Download July dataset at: download_link
Download December dataset at: download_link
/imu/data
/velodyne_points
Info: During this data acquisition campaign the robot was teleoperated. Be careful! IMU data may be degradated since the port was not set at low latency.
Other dataset acquired in vineyard, with reliable IMU data and additional GNSS data, will be available soon!