Orchard multisensor dataset for weed and trunk mapping – Campinas A (session CA-10)

Session CA-10 of the orchard multisensor dataset: one 22-minute recording of 3D LiDAR, IMU, RGB-D camera and GNSS data under the canopy of Campinas A, an older orange orchard with larger trees and uneven terrain in southern Portugal, recorded on 27 June 2026 with the second version of the purpose-built ground platform. The recording is a ROS 2 bag. To protect the location of the private orchard, every GNSS fix was moved to a synthetic origin at latitude 0, longitude 0: distances, headings and the shape of the track are preserved, but not the absolute position. The same track is given in local East–North–Up coordinates in gnss_enu.csv. The session contains a gap of about 20 s in all sensor streams, described in the README. The main record of the dataset, with the sensor calibration (valid for this session), the trunk survey and the labelled images, is https://doi.org/10.5281/zenodo.23805668. This session is used in Appendix B of the MSc thesis “Mobile Robot Navigation for Vegetation Mapping in Agricultural Applications” (Instituto Superior Técnico, Universidade de Lisboa, 2026).