Abstract: In a method for subpixel disparity calculation, image data for various images each representing a field of view of an input device is received by a processor, and the image data is applied to a machine learning model. The machine learning module uses the image data to compute an output representing calculated subpixel disparity between the various images. In an example of the method, the machine learning model is a neural network that produces accurate and reliable subpixel disparity estimation in real-time using synthetically generated data.

Abstract: A system and method are described that provide for mapping features of a warehouse environment having improved workflow. In one example of the system/method of the present invention, a mapping robot is navigated through a warehouse environment, and sensors of the mapping robot collect geospatial data as part of a mapping mode. A Frontend N block of a map framework may be responsible for reading and processing the geospatial data from the sensors of the mapping robot, as well as various other functions. The data may be stored in a keyframe object at a keyframe database. A Backend block of the map framework may be useful for detecting loop constraints, building submaps, optimizing a pose graph using keyframe data from one or more trajectory blocks, and/or various other functions.

Abstract: To load and unload a trailer, an autonomous mobile robot determines its location and the location of objects within the trailer relative to the trailer itself, rather than relative to a warehouse. The autonomous mobile robot determines its location the location of objects within the trailer relative to the trailer. The autonomous mobile robot navigates within the trailer and manipulates objects within the trailer from the trailer's reference frame. Additionally, the autonomous mobile robot uses a centerline heuristic to compute a path for itself within the trailer. A centerline heuristic evaluates nodes within the trailer based on how far away those nodes are from the centerline. If the nodes are further away from the centerline, they are assigned a higher cost. Thus, when the autonomous mobile robot computes a path, the path is more likely to stay near the centerline of the trailer rather than get closer to the sides.

Abstract: Loading last few rows of pallets onto a trailer using an autonomous mobile robot. The robot determines that the next pallet to be loaded is a first in a row where the trailer does not have sufficient space to accommodate the robot. The robot identifies a pose of a previous pallet in an immediately prior row and determines a front plane of the immediately prior row. The robot navigates to a first goal position at least partially inside the trailer, determined by the pose of the previous pallet and the trailer. The robot then side-shifts the fork toward the trailer's side wall until contact is detected and subsequently adjust the fork back to prevent scraping. The robot proceeds to a second goal position, which is within a predetermined threshold distance of the drop position, before lowering and releasing the pallet.

Abstract: Unloading first few rows of pallets from a trailer using an autonomous mobile robot. The robot determines that a pallet is a first in a row where the trailer does not have sufficient space to accommodate the robot. The robot determines a pose of each observable pallet in the trailer and determines a front plane for the pallets in the same row as the target pallet. The robot navigates to a first goal position inside the trailer, based on the pallet and trailer poses, then picks up the pallet. The robot side-shifts toward an adjacent pallet until detecting contact, then adjusts back by a predetermined distance to maximize clearance between the pallet and a side wall of the trailer. The robot navigates backward in a straight line to a second goal position on a ramp, and then proceeds to drop off the pallet in the staging area.

Abstract: Loading a pallet into a trailer using an autonomous mobile robot. The robot determines a pose of the trailer based on sensor data and navigates to a first goal position inside the trailer, determined based on the pose of the trailer. The robot then side-shifts the fork toward the trailer's side wall until sensors detect contact between the pallet and the side wall, with the pallet positioned above a lip on the trailer wall. The robot retracts the fork by a distance corresponding to the lip's width to prevent the pallet and the side wall of the trailer from scraping each other. The robot then navigates in a straight line forward to a second goal position, which is within a predetermined threshold distance from the drop position. The robot releases the pallet at the second goal position.

Abstract: Unloading pallets from a trailer using an autonomous mobile robot. The robot determines a pose of the trailer and a pose of each observable pallet inside. The robot identifies a target pallet for retrieval based on the observed poses and determines a front plane for the pallets in the same row as the target pallet. The robot navigates to a first goal position, side-shifts its fork to align the fork with the target pallet's pockets, inserts the fork, and lifts the pallet. The robot then navigates in reverse to a second goal position, determined based on the front plane and trailer pose. From the second position, the robot proceeds to a drop-off point in a staging area, side-shifting the fork towards the center during transit.

Abstract: A method implemented at an autonomous mobile robot equipped with a fork to carry a pallet. The robot transitions between a first and second piecewise flat floor segments with differing geometries. The robot uses sensor data from sensors such as LIDAR, stereo cameras, GPS, and ultrasound sensors to determine the transition between the first and second piecewise flat floor segments. The fork operates based on parameters that meet reference constraints. When the robot detects that a second floor geometry would cause these parameters to no longer meet the reference constraints, the robot determines new parameters that will satisfy the reference constraints. Control signals are then sent to adjust the fork's operation as the robot transitions to the second floor segment.

Abstract: A system and method are described that provide for mapping features of a warehouse environment having improved workflow. In one example of the system/method of the present invention, a mapping robot is navigated through a warehouse environment, and sensors of the mapping robot collect geospatial data as part of a mapping mode. A Frontend N block of a map framework may be responsible for reading and processing the geospatial data from the sensors of the mapping robot, as well as various other functions. The data may be stored in a keyframe object at a keyframe database. A Backend block of the map framework may be useful for detecting loop constraints, building submaps, optimizing a pose graph using keyframe data from one or more trajectory blocks, and/or various other functions.

Abstract: A system and a method are disclosed that generate for display to a remote operator a user interface comprising a map, the map comprising visual representations of a source area, a plurality of candidate robots, and a plurality of candidate destination areas. The system receives, via the user interface, a selection of a visual representation of a candidate robot of the plurality of candidate robots, and detects a drag-and-drop gesture within the user interface of the visual representation of the candidate robot being dragged-and-dropped to a visual representation of a candidate destination area of the plurality of candidate destination areas. Responsive to detecting the drag-and-drop gesture, the system generates a mission, where the mission causes the candidate robot to autonomously transport an object from the source area to the candidate destination area.

Abstract: In a method for subpixel disparity calculation, image data for various images each representing a field of view of an input device is received by a processor, and the image data is applied to a machine learning model. The machine learning module uses the image data to compute an output representing calculated subpixel disparity between the various images. In an example of the method, the machine learning model is a neural network that produces accurate and reliable subpixel disparity estimation in real-time using synthetically generated data.

Abstract: To load and unload a trailer, an autonomous mobile robot determines its location and the location of objects within the trailer relative to the trailer itself, rather than relative to a warehouse. The autonomous mobile robot determines its location the location of objects within the trailer relative to the trailer. The autonomous mobile robot navigates within the trailer and manipulates objects within the trailer from the trailer's reference frame. Additionally, the autonomous mobile robot uses a centerline heuristic to compute a path for itself within the trailer. A centerline heuristic evaluates nodes within the trailer based on how far away those nodes are from the centerline. If the nodes are further away from the centerline, they are assigned a higher cost. Thus, when the autonomous mobile robot computes a path, the path is more likely to stay near the centerline of the trailer rather than get closer to the sides.

Abstract: A system and a method are disclosed that generate for display to a remote operator a user interface comprising a map, the map comprising visual representations of a source area, a plurality of candidate robots, and a plurality of candidate destination areas. The system receives, via the user interface, a selection of a visual representation of a candidate robot of the plurality of candidate robots, and detects a drag-and-drop gesture within the user interface of the visual representation of the candidate robot being dragged-and-dropped to a visual representation of a candidate destination area of the plurality of candidate destination areas. Responsive to detecting the drag-and-drop gesture, the system generates a mission, where the mission causes the candidate robot to autonomously transport an object from the source area to the candidate destination area.