ROAM@CRAS - A harbor multidomain mapping dataset
Authors: Daniel Filipe Campos, Pedro Nuno Leite, Renato Silva, Maria Inês Pereira, Rafael Marques, Andry Maykol Pinto.
Marina de Leça and INESC TEC (CRAS), Porto, Portugal. 2020.


The ROAM@CRAS dataset was acquired using an Autonomous Surface Vehicle (ASV), the SENSE, with a set of incorporated sensors for perceiving the surface and underwater domains (Velodyne VLP-16, Mynt Eye D and Imagenex ”Delta T” 837B) as well as navigation sensors such as L1/L2 RTK, GPS and IMU. This dataset was acquired during a campaign of the project DIIUS (Distributed perceptIon for inspectIon of aqUatic Structures) [1] on the harbor of Marina de Leça, Porto, Portugal (https://goo.gl/maps/nHz9mA9Thqii6sy99). The ASV was remotely operated at all time performing a trajectory with distinct challenges for 3D mapping and sensor fusion applications, such as data imperfecteness, inconsistency and heterogeneity. All data was recorded using ROS bags each with ~14s duration to be played in sequence all together taking up to the total mission time (~1196s).

For the SENSE tfs, launches and more details check the github repository - https://github.com/danielfbcampos/ROAM-CRAS


This dataset includes:

        - 83 ROS bag files containing the captured data with filename as follows:
	harbor_xx.bag

Each bag is composed by the following topics:

- /cmd_vel: geometry_msgs/Twist - Velocity command from the control station                     
- /gps/basestation_ecef: nav_msgs/Odometry - Swift Navigation Piksi Multi RTK base station position in ECEF                     
- /gps/fix: sensor_msgs/NavSatFix - Swift Navigation Piksi Multi GPS position              
- /gps/rtkfix: nav_msgs/Odometry - SENSE Swift Navigation Piksi Multi GPS RTK odometry                     
- /gps/time: sensor_msgs/TimeReference -Swift Navigation Piksi Multi GPS Time reference           
- /imu/data: sensor_msgs/Imu - Xsens MTi-30 IMU data                   
- /imu_nav/data: sensor_msgs/Imu - Sparkfun Razor IMU 9DOF data                        
- /left/cmd: roboteq_msgs/Command - Left thruster command for Roboteq                 
- /mbes/scan: sensor_msgs/LaserScan - Imagenex ”Delta T” 837B scan                  
- /mynteye/depth/camera_info: sensor_msgs/CameraInfo - Mynt Eye D camera parameters                 
- /mynteye/depth/image_raw: sensor_msgs/Image - Mynt Eye D depth image                       
- /mynteye/left/image_color/compressed: sensor_msgs/CompressedImage - Mynt Eye D left image           
- /mynteye/right/image_color/compressed: sensor_msgs/CompressedImage - Mynt Eye D right image             
- /right/cmd: roboteq_msgs/Command - Right thruster command for Roboteq                   
- /roboteq_driver/status: roboteq_msgs/Status - Roboteq status feedback                     
- /safety_stop: std_msgs/Bool - Emergency failsafe status                            
- /ublox_nav/fix: sensor_msgs/NavSatFix - Vk-162 Glonass Navigation USB GPS position data                    
- /ublox_nav/fix_velocity: geometry_msgs/TwistWithCovarianceStamped - Vk-162 Glonass Navigation USB GPS velocity data
- /velocity: geometry_msgs/TwistStamped - Angular velocity published by the Xsens MTi-30 IMU            
- /velodyne_points: sensor_msgs/PointCloud2 - Velodyne VLP-16 Point cloud


References

[1] INESC TEC, DIIUS - Distributed perceptIon for inspectIon of aqUatic Structures, http://diius.inesctec.pt/, accessed in September 2020