Highlights
This is a minor follow up to the model API generalizations already introduced in v19.7.0:
LidarModelParametersdeclaresget_vertical_fovandget_horizontal_fov. Every lidar has a vertical and a horizontal field of view, whatever its scanning design, the same way every camera has a resolution, but both were previously declared only on
RowOffsetStructuredSpinningLidarModelParameters. There was no way to ask an arbitrary lidar for its field of view, and a second model could have arrived with a different spelling or none at all.- The implementation stays on the row-offset model. It derives the vertical extent from
row_elevations_radand the horizontal one fromcolumn_azimuths_radandrow_azimuth_offsets_rad, all of which exist only there. A model with a different scanning design
derives the same two quantities from whatever it stores, which is the reason the contract belongs on the base and the derivation does not. - Concrete lidar parameter classes must now implement both, as they already must implement
type. Every model in NCore does, and no out-of-tree lidar model exists yet, so nothing outside this repository is affected.
v19.8.0 - 2026-09-15
➕ Added
- (data) declare the field-of-view contract on the lidar parameter base - (e7306c7) - Janick Martinez Esturo, Claude Opus 5