Abstract: Techniques for controlling a robotic picking arm using estimated seal quality metrics. A plurality of candidate contact points for holding an item using a suction device of the robotic picking arm, based on captured images of the item and an n-dimensional surface model of the item. An expected seal quality metric for a first one of the candidate contact points, by processing the n-dimensional surface model of the item and physical properties of the suction device of the robotic picking arm. Based on the expected seal quality metric, embodiments can determine whether to retrieve the item from the storage container by holding the item at the first candidate contact point using the suction device of the robotic picking arm.

Abstract: Aspects described herein include an end effector capable of prehending items using impactive and astrictive forces. The end effector includes an interface system having a deformable mounting plate and a pliable body member attached to the mounting plate. The end effector further includes a linkage system between a plurality of actuators and the interface system. The linkage system connects to lateral portions of the mounting plate.

Abstract: Aspects described herein include an end effector capable of prehending items using impactive and astrictive forces. The end effector includes an interface system having a deformable mounting plate and a pliable body member attached to the mounting plate. The end effector further includes a linkage system between a plurality of actuators and the interface system. The linkage system connects to lateral portions of the mounting plate.

Abstract: Method and apparatus for training a machine learning model for controlling a robotic picking device having a suction device end effector. A robotic control operation and a candidate contact point for holding a first item using a suction device of a robotic picking arm are determined, by processing information describing the first item as an input to a machine learning model. A seal quality metric is estimated for the candidate contact point, based on a predicted deformed n-dimensional shape of the suction device and a n-dimensional shape associated with the first item. One or more weights within the machine learning model are refined based on the estimated seal quality metric.

Abstract: Techniques for controlling a robotic picking arm using estimated seal quality metrics. A plurality of candidate contact points for holding an item using a suction device of the robotic picking arm, based on captured images of the item and an n-dimensional surface model of the item. An expected seal quality metric for a first one of the candidate contact points, by processing the n-dimensional surface model of the item and physical properties of the suction device of the robotic picking arm. Based on the expected seal quality metric, embodiments can determine whether to retrieve the item from the storage container by holding the item at the first candidate contact point using the suction device of the robotic picking arm.

Abstract: Techniques for controlling a robotic picking arm using estimated seal quality metrics. A plurality of candidate contact points for holding an item using a suction device of the robotic picking arm, based on captured images of the item and an n-dimensional surface model of the item. An expected seal quality metric for a first one of the candidate contact points, by processing the n-dimensional surface model of the item and physical properties of the suction device of the robotic picking arm. Based on the expected seal quality metric, embodiments can determine whether to retrieve the item from the storage container by holding the item at the first candidate contact point using the suction device of the robotic picking arm.

Abstract: Techniques for controlling a robotic picking arm using estimated seal quality metrics. A plurality of candidate contact points for holding an item using a suction device of the robotic picking arm, based on captured images of the item and an n-dimensional surface model of the item. An expected seal quality metric for a first one of the candidate contact points, by processing the n-dimensional surface model of the item and physical properties of the suction device of the robotic picking arm. Based on the expected seal quality metric, embodiments can determine whether to retrieve the item from the storage container by holding the item at the first candidate contact point using the suction device of the robotic picking arm.

Abstract: A transfer rack on one floor of an inventory system can receive multiple containers each carrying an inventory item designated for an operation on a different floor of the inventory system. A vertical lift can move the transfer rack carrying the multiple containers from the one floor to the different floor. On the different floor, the multiple containers may be swapped for different multiple containers each carrying an inventory item designated for an operation on the one floor. The vertical lift may return the transfer rack carrying the different multiple containers to the one floor.

Abstract: Techniques for providing an entity monitoring safety feature in an inventory management system may be described. For example, information that indicates respective locations for a plurality of mobile drive units within an inventory management system may be received. Tracking information may be received from a mobile location unit that is moving within a bounded area around a signal transmitter that is coupled to an entity. The tracking information may include an approximate location of the signal transmitter relative to the mobile location unit. Pathway information may be generated that indicates to the plurality of mobile drive units within the inventory management system to stay outside of a radius distance of the mobile location unit based on the received tracking information. The generated pathway information may be provided to the plurality of mobile drive units to enable safe passage for the entity within the inventory management system.

Abstract: Payload properties may be determined utilizing actuator parameter data. A mobile drive unit may transport a payload around a workspace in accordance with one or more safety. To comply with safety policies, a motion controller of the mobile drive unit may limit acceleration and deceleration to safe maximums for given payload characteristics such as payload mass and payload center of gravity. The motion controller may utilize actuators to cause various motions of the mobile drive unit, including motion across a surface, vertical movements of the payload and motion rotating the payload. To ameliorate worst case assumptions about payload characteristics, in accordance with at least one embodiment of the invention, a payload characterization module may obtain information about payload characteristics based on measurements of operational parameters of the actuators.

Abstract: An inventory system includes an inventory holder and an actively balanced mobile drive unit. A control module of the mobile drive unit can receive sensing information about the inventory holder and/or the mobile drive unit and use the sensing information to control a drive module of the mobile drive unit so as to maintain the inventory holder and/or the mobile drive unit within a predetermined deviation amount from an equilibrium state.

Abstract: Robotic arms or manipulators can be utilized to grasp inventory items within an inventory system. Information can be obtained about constraints relative to relevant elements of a process of transferring the item from place to place. Examples of such elements may include a grasping location from which an item is to be grasped, a receiving location in which a grasped item is to be placed, or a space between the grasping location and the receiving location. The information about the constraints can be used to select from multiple possible grasping options, such as by eliminating options that conflict with the constraints or preferring options that outperform others given the constraints.

Abstract: Item information such as an identity or arrangement of a selected inventory item of an inventory system can be determined. A classification of the selected inventory item can be obtained based at least in part on the item information. The selected inventory item can be routed based at least in part on the classification of the selected inventory item. For example, the selected inventory item may be routed for manual handling if the classification of the selected inventory item indicates that the selected inventory item is designated for an operation to be performed relative to the selected inventory item by a human operator. Alternatively, the selected inventory item may be routed for automated handling if the classification of the selected inventory item indicates that the selected inventory item is designated for an operation to be performed relative to the selected inventory item by a robotic manipulator.

Abstract: Robotic arms or manipulators can be utilized to grasp inventory items within an inventory system. Information can be obtained about constraints relative to relevant elements of a process of transferring the item from place to place. Examples of such elements may include a grasping location from which an item is to be grasped, a receiving location in which a grasped item is to be placed, or a space between the grasping location and the receiving location. The information about the constraints can be used to select from multiple possible grasping options, such as by eliminating options that conflict with the constraints or preferring options that outperform others given the constraints.

Abstract: Inventory handlers can include one or more gripping elements connected with a gripping assembly. The gripping element or elements can be subjected to shear force when in contact with inventory items such that the anisotropically adhesive surface engages an inventory item with an adhesive force. Inventory systems can employ robotic inventory handlers as described to lift and displace inventory items using the adhesive force.

Abstract: A transfer rack on one floor of an inventory system can receive multiple containers each carrying an inventory item designated for an operation on a different floor of the inventory system. A vertical lift can move the transfer rack carrying the multiple containers from the one floor to the different floor. On the different floor, the multiple containers may be swapped for different multiple containers each carrying an inventory item designated for an operation on the one floor. The vertical lift may return the transfer rack carrying the different multiple containers to the one floor.

Abstract: Inventory handlers can include one or more gripping elements connected with a gripping assembly. The gripping element or elements can be subjected to shear force when in contact with inventory items such that the anisotropically adhesive surface engages an inventory item with an adhesive force. Inventory systems can employ robotic inventory handlers as described to lift and displace inventory items using the adhesive force.

Abstract: Robotic arms or manipulators can be utilized to grasp inventory items within an inventory system. Information can be obtained about constraints relative to relevant elements of a process of transferring the item from place to place. Examples of such elements may include a grasping location from which an item is to be grasped, a receiving location in which a grasped item is to be placed, or a space between the grasping location and the receiving location. The information about the constraints can be used to select from multiple possible grasping options, such as by eliminating options that conflict with the constraints or preferring options that outperform others given the constraints.

Abstract: Robotic arms may be utilized to grasp inventory items within an inventory system. Information about an inventory item to be grasped can be detected and used to determine a grasping strategy in conjunction with information from a database. Instructions for grasping an inventory item can be generated based on the detected information and the database.