Dataset for the paper "A framework for robotic excavation and dry stone construction using on-site materials"

Stone data from the Science Robotics paper "A framework for robotic excavation and dry stone construction using on-site materials" containing:

Mesh files of 1,100 stones (quarried boulders, erratics, and concrete debris) that were digitized by the autonomous excavator HEAP

1100 Unprocessed Stone Meshes.zip: Raw mesh files directly from the poisson reconstruction of accumulated LiDAR points, containing some artifacts and floating geometries
1100 Closed Stone Meshes.zip: Clean, closed, downsampled meshes
Stone_Shape_Properties.csv: Properties file with a list of the stone IDs (IDs in the 1xxx and 3xxx range typically correspond to concrete elements) and select shape properties
A dataset of candidate placements from automatically generated stone walls.  The candidate placement data zip files contain:

Candidate_Placement_Data-npy.zip: SDF (.npy) representation of each candidate, with three channels of 32x32x32 for distances to the stone, the already-placed stones, and the target wall
Candidate_Placement_Data-pcd.zip: Point cloud (.pcd) representations of each candidate, with separate files for the placed stone, target wall (search volume), and already-placed stones (where they exist)
sdf_classifier.zip: Python examples:

Rendering the three channel SDF data to mesh geometry using marching cubes and libigl
Candidate SDF classification using the pretrained model
Candidate attributes and labels

candidate_attributes_labels.csv: CSV file containing a list of UUID's corresponding to each candidate placement in the dataset.  For each candidate, additional information is included about the dimensions and location of the solution, together with the (subjectively) hand-labelled binary value for placement viability. 
README.md: An additional readme with some details on the attributes file
If you use this data in your research, please cite the journal article.