Abstract: A method of detecting a back of a shelf for supporting objects includes: obtaining an image depicting a shelf having a shelf edge and a support surface extending from the shelf edge to a shelf back; decomposing the image into a plurality of patches; for each patch: generating a feature descriptor; based on the feature descriptor, assigning one of a shelf back classification and a non-shelf back classification to the patch; generating a mask corresponding to the image, the mask containing an indication of the classification assigned to each of the patches; and presenting the mask.

Abstract: A method includes obtaining a point cloud captured by a depth sensor, and image data captured by an image sensor, the point cloud and the image data representing a support structure bearing a set of objects; obtaining an image boundary corresponding to an object from the set of objects; determining a portion of the point cloud corresponding to the image boundary; selecting, from the determined portion, a subset of points corresponding to a forward surface of the object; and generating a three-dimensional position of the object based on the forward surface.

Abstract: A method of determining a support structure depth of a support structure having a front and a back separated by the support structure depth includes: obtaining a point cloud of the support structure, and a mask indicating, for a plurality of portions of an image of the support structure captured from a capture pose, respective confidence levels that the portions depict the back of the support structure; selecting, from the point cloud, an initial set of points located within a field of view originating at the capture pose; selecting, from the initial set of points, an unoccluded subset of depth measurements, the depth measurements in the unoccluded subset corresponding to respective image coordinates; retrieving, from the mask, a confidence level for each of the depth measurements in the unoccluded subset; and based on the depth measurements in the unoccluded subset and the retrieved confidence levels, determining the support structure depth.

Abstract: A method includes obtaining a point cloud captured by a depth sensor, and image data captured by an image sensor, the point cloud and the image data representing a support structure bearing a set of objects; obtaining an image boundary corresponding to an object from the set of objects; determining a portion of the point cloud corresponding to the image boundary; selecting, from the determined portion, a subset of points corresponding to a forward surface of the object; and generating a three-dimensional position of the object based on the forward surface.

Abstract: A method of label detection includes: obtaining, by an imaging controller, an image depicting a shelf; increasing an intensity of a foreground subset of image pixels exceeding an upper intensity threshold, and decreasing an intensity of a background subset of pixels below a lower intensity threshold; responsive to the increasing and the decreasing, (i) determining gradients for each of the pixels and (ii) selecting a candidate set of the pixels based on the gradients; overlaying a plurality of shelf candidate lines on the image derived from the candidate set of pixels; identifying a pair of the shelf candidate lines satisfying a predetermined sequence of intensity transitions; and generating and storing a shelf edge bounding box corresponding to the pair of shelf candidate lines.

Abstract: A method in an imaging controller of detecting obstructions on a front of a support structure includes: obtaining (i) a point cloud of the support structure and an obstruction, and (ii) a support structure plane corresponding to the front of the support structure; for each of a plurality of selection depths: selecting a subset of points from the point cloud based on the selection depth; detecting obstruction candidates from the subset of points and, for each obstruction candidate: responsive to a dimensional criterion being met, determining whether the obstruction candidate meets a confirmation criterion; when the obstruction candidate meets the confirmation criterion, identifying the obstruction candidate as a confirmed obstruction; and presenting obstruction detection output data including the confirmed obstructions.

Abstract: A method of mobile automation apparatus localization in a navigation controller includes: controlling a depth sensor to capture a plurality of depth measurements corresponding to an area containing a navigational structure; selecting a primary subset of the depth measurements; selecting, from the primary subset, a corner candidate subset of the depth measurements; generating, from the corner candidate subset, a corner edge corresponding to the navigational structure; selecting an aisle subset of the depth measurements from the primary subset, according to the corner edge; selecting, from the aisle subset, a local minimum depth measurement for each of a plurality of sampling planes extending from the depth sensor; generating a shelf plane from the local minimum depth measurements; and updating a localization of the mobile automation apparatus based on the corner edge and the shelf plane.

Abstract: A method includes obtaining a point cloud captured by a depth sensor, and image data captured by an image sensor, the point cloud and the image data representing a support structure bearing a set of objects; obtaining an image boundary corresponding to an object from the set of objects; determining a portion of the point cloud corresponding to the image boundary; selecting, from the determined portion, a subset of points corresponding to a forward surface of the object; and generating a three-dimensional position of the object based on the forward surface.

Abstract: A method of detecting an edge of a support surface by an imaging controller includes: obtaining a plurality of depth measurements captured by a depth sensor and corresponding to an area containing the support surface; selecting, by the imaging controller, a candidate set of the depth measurements; fitting, by the imaging controller, a guide element to the candidate set of depth measurements; and detecting, by the imaging controller, an output set of the depth measurements corresponding to the edge from the candidate set of depth measurements according to a proximity between each candidate depth measurement and the guide element.

Abstract: A method in an imaging controller of detecting depth sensor artifacts includes: obtaining, from first and second sensors, first and second pluralities of points defined by respective (i) planar positions and depths in a common frame of reference, and (ii) scan angles relative to field of view centers of the first or second sensors; for each of a subset of candidate points from the first plurality of points: searching the second plurality of points for a validator point having (i) a planar position within a threshold distance of a planar position of the candidate point, and (ii) a scan angle smaller than a scan angle of the candidate point; responsive to identifying the validator point: when the depth of the validator point exceeds the depth of the candidate point, classifying the candidate point as an artifact.

Abstract: A method in an imaging controller of detecting depth sensor artifacts includes: obtaining, from first and second sensors, first and second pluralities of points defined by respective (i) planar positions and depths in a common frame of reference, and (ii) scan angles relative to field of view centers of the first or second sensors; for each of a subset of candidate points from the first plurality of points: searching the second plurality of points for a validator point having (i) a planar position within a threshold distance of a planar position of the candidate point, and (ii) a scan angle smaller than a scan angle of the candidate point; responsive to identifying the validator point: when the depth of the validator point exceeds the depth of the candidate point, classifying the candidate point as an artifact.

Abstract: A method in an imaging controller of detecting obstructions on a front of a support structure includes: obtaining (i) a point cloud of the support structure and an obstruction, and (ii) a support structure plane corresponding to the front of the support structure; for each of a plurality of selection depths: selecting a subset of points from the point cloud based on the selection depth; detecting obstruction candidates from the subset of points and, for each obstruction candidate: responsive to a dimensional criterion being met, determining whether the obstruction candidate meets a confirmation criterion; when the obstruction candidate meets the confirmation criterion, identifying the obstruction candidate as a confirmed obstruction; and presenting obstruction detection output data including the confirmed obstructions.

Abstract: A method of detecting a back of a shelf for supporting objects includes: obtaining an image depicting a shelf having a shelf edge and a support surface extending from the shelf edge to a shelf back; decomposing the image into a plurality of patches; for each patch: generating a feature descriptor; based on the feature descriptor, assigning one of a shelf back classification and a non-shelf back classification to the patch; generating a mask corresponding to the image, the mask containing an indication of the classification assigned to each of the patches; and presenting the mask.

Abstract: A method of recovering label positions in an imaging controller includes: receiving (i) a definition of a plane containing edges of the shelf, and (ii) a plurality of initial label indicators having locations on the plane; assigning the initial label indicators among a plurality of candidate subsets each representing a single physical label; for each candidate subset of initial indicators: generating, for each of a plurality of projection depths relative to the plane, a set of projections of the initial indicators in the candidate subset; determining an aggregate surface area for each set of projections; selecting, as a recovered depth for the candidate subset, one of the projection depths corresponding to the minimum aggregate surface area; and generating a recovered position for the candidate subset based on the recovered depth; and storing the recovered positions.

Abstract: A method in an imaging controller of correcting translucency artifacts in data representing one or more objects disposed on a shelf includes: obtaining a plurality of depth measurements captured by a depth sensor and corresponding to an area containing the shelf; obtaining (i) a definition of a plane containing edges of the shelf, (ii) a location in the plane of an upper shelf edge, and (iii) a location in the plane of a lower shelf edge adjacent to the upper shelf edge; generating a depth map containing, for each of a plurality of positions in the plane, a nearest object depth; detecting an upper object boundary in the depth map between the upper and lower support surface edges; updating each nearest object depth between the upper object boundary and the lower shelf edge to contain a depth of the upper object boundary; and storing the corrected depth map.

Abstract: A method of detecting a back of a shelf for supporting objects includes: obtaining an image depicting a shelf having a shelf edge and a support surface extending from the shelf edge to a shelf back; decomposing the image into a plurality of patches; for each patch: generating a feature descriptor; based on the feature descriptor, assigning one of a shelf back classification and a non-shelf back classification to the patch; generating a mask corresponding to the image, the mask containing an indication of the classification assigned to each of the patches; and presenting the mask.

Abstract: A method of detecting an edge of a support surface by an imaging controller includes: obtaining a plurality of depth measurements captured by a depth sensor and corresponding to an area containing the support surface; selecting, by the imaging controller, a candidate set of the depth measurements; fitting, by the imaging controller, a guide element to the candidate set of depth measurements; and detecting, by the imaging controller, an output set of the depth measurements corresponding to the edge from the candidate set of depth measurements according to a proximity between each candidate depth measurement and the guide element.

Abstract: A device and method for merging lidar data is provided. Point cloud data is combined, via a lidar imaging controller, into a common point cloud data set, each set of point cloud data representing respective angular lidar scans of a region as at least two lidar devices are moved relative to the region of a shelf. The respective angular lidar scans from each lidar device occur at a non-zero and non-perpendicular angle to a movement direction. Common point cloud data set points are binned into a plane perpendicular to the movement direction of a mobile automation apparatus and extending from a virtual lidar position. The lidar imaging controller combines points among multiple planes.

Abstract: A method of object status detection for objects supported by a shelf, from shelf image data, includes: obtaining a plurality of images of a shelf, each image including an indication of a gap on the shelf between the objects; registering the images to a common frame of reference; identifying a subset of the gaps having overlapping locations in the common frame of reference; generating a consolidated gap indication from the subset; obtaining reference data including (i) identifiers for the objects and (ii) prescribed locations for the objects within the common frame of reference; based on a comparison of the consolidated gap indication with the reference data, selecting a target object identifier from the reference data; and generating and presenting a status notification for the target product identifier.

Abstract: A method of determining a support structure depth of a support structure having a front and a back separated by the support structure depth includes: obtaining a point cloud of the support structure, and a mask indicating, for a plurality of portions of an image of the support structure captured from a capture pose, respective confidence levels that the portions depict the back of the support structure; selecting, from the point cloud, an initial set of points located within a field of view originating at the capture pose; selecting, from the initial set of points, an unoccluded subset of depth measurements, the depth measurements in the unoccluded subset corresponding to respective image coordinates; retrieving, from the mask, a confidence level for each of the depth measurements in the unoccluded subset; and based on the depth measurements in the unoccluded subset and the retrieved confidence levels, determining the support structure depth.