SMARTNav: A Visual-Inertial Dataset for Reliable Robotic State Estimation

Alt text

On this page we present the SMARTNav dataset, containing raw sensor data recorded by aerial and ground robots. Collected across multiple projects targeting autonomous navigation in industrial settings, the dataset primarily supports evaluation of Visual-Inertial Odometry (VIO). More broadly, it is suitable for vision and LiDAR-based SLAM as well as multi-sensor fusion research.

Downloads

All sequences are provided as ROS 2 bag files. Because each sequence was captured with different hardware for different projects, the available topics vary; however, stereo images and IMU data are included in every sequence. Currently, all bag files are available exclusively via the Saxion Research Cloud Drive (SRCD).

Note: For those sequences of this dataset containing RTK data, px4-msgs (it must be release/1.15 branch) is needed. This dependency will be removed in future versions of dataset.

Seq. Name	Size[GB]	Descriptions	Link
greenhouse_flight_3	6.1	Topics: stereo images IMU RTK Sensors: ZEDX camera Holybro H-RTK F9P	SRCD
greenhouse_flight_5	3.1	Topics: stereo images IMU RTK Sensors: ZEDX camera Holybro H-RTK F9P	SRCD
greenhouse_ground_1	7.3	Topics: stereo images camera IMU LiDAR pointcloud LiDAR IMU GPS Sensors: ZEDX camera Holybro H-RTK F9P Ouster OS1-128	SRCD
greenhouse_ground_2	7.7	Topics: stereo images camera IMU LiDAR pointcloud LiDAR IMU GPS Sensors: ZEDX camera Holybro H-RTK F9P Ouster OS1-128	SRCD
greenhouse_handheld_1	2.4	Topics: stereo images IMU RTK Sensors: ZEDX camera Holybro H-RTK F9P	SRCD
optitrack_flight_1	0.5	Topics: stereo images IMU motion capture Sensors: ZED 2 camera OptiTrack Prime 13	SRCD
optitrack_flight_2	1.7	Topics: stereo images IMU motion capture Sensors: ZED 2 camera OptiTrack Prime 13	SRCD
optitrack_handheld_2	1.1	Topics: stereo images IMU motion capture Sensors: ZED 2 camera OptiTrack Prime 13	SRCD
optitrack_handheld_3	1.1	Topics: stereo images IMU motion capture Sensors: ZED 2 camera OptiTrack Prime 13	SRCD
outdoor_flight_3	2.5	Topics: stereo images IMU RTK Sensors: ZED 2 camera Holybro H-RTK F9P	SRCD
outdoor_flight_4	6.8	Topics: stereo images IMU RTK Sensors: ZED 2 camera Holybro H-RTK F9P	SRCD
outdoor_flight_5	8.7	Topics: stereo images IMU RTK Sensors: ZED 2 camera Holybro H-RTK F9P	SRCD
outdoor_flight_6	3.5	Topics: stereo images IMU RTK Sensors: ZED 2 camera Holybro H-RTK F9P	SRCD
sidewalk_ground_1	35.9	Topics: stereo images camera IMU LiDAR pointcloud LiDAR IMU GPS Sensors: ZEDX camera Holybro H-RTK F9P Ouster OS1-128	SRCD
corridor_ground_1	25.8	Topics: stereo images camera IMU LiDAR pointcloud LiDAR IMU Sensors: ZEDX camera Ouster OS1-128	SRCD

Ground Truth and Calibration

In the table below, for each sequence, a bag file containing ground-truth topic and well as the sensor calibration data are provided. The GT topic type and calibration for each sequence may differ. Our evaluation tool can handle and compare these different topic types.

Seq. Name	GT Source	GT Link	Calibration
greenhouse_flight_3	RTK	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml
greenhouse_flight_5	RTK	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml
greenhouse_ground_1	LiDAR SLAM (GLIM algorithm)	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml leftcam_rightcam_extrinsic.yaml lidar_imu_extrinsic.yaml
greenhouse_ground_2	LiDAR SLAM (LeGO-LOAM algorithm)	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml leftcam_rightcam_extrinsic.yaml lidar_imu_extrinsic.yaml
greenhouse_handheld_1	RTK	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml
optitrack_flight_1	Motion Capture System	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml
optitrack_flight_2	Motion Capture System	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml
optitrack_handheld_2	Motion Capture System	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml
optitrack_handheld_3	Motion Capture System	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml
outdoor_flight_3	RTK	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml
outdoor_flight_4	RTK	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml
outdoor_flight_5	RTK	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml
outdoor_flight_6	RTK	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml
sidewalk_ground_1	LiDAR SLAM (LeGO-LOAM algorithm)	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml leftcam_rightcam_extrinsic.yaml lidar_imu_extrinsic.yaml
corridor_ground_1	LiDAR SLAM (LeGO-LOAM algorithm)	download	left_camera_intrinsic.yaml right_camera_intrinsic.yaml imu_leftcam_extrinsic.yaml imu_rightcam_extrinsic.yaml leftcam_rightcam_extrinsic.yaml lidar_imu_extrinsic.yaml

Acknowledgements

This work was partly supported by the Netherlands Organization for Scientific Research (NWO) via SIA RAAK-Public project (Van bestrijden naar beheersen van de EPR, No.10.015) and SIA RAAK-MKB project (Smart Greenhouses, No.17.014).