Disclosed is a deep reinforcement learning apparatus and method for a pick-and-place system. According to the present disclosure, a simulation learning framework is configured to apply reinforcement learning to make pick-and-place decisions using a robot operating system (ROS) in a real-time environment, thereby generating stable path motion that meets various hardware and real-time constraints.

A method for picking objects is specified, in which at least one object is removed from a source loading aid and placed into a target loading aid using a robot. After the operation of removing the object, a first sensor system of the robot is used to check whether at least one object is held by the robot. A number and/or a type of the at least one removed object is ascertained using the second sensor system. The operation of placing the at least one object into the target loading aid is aborted or modified if no object is held by the robot or the number and/or the type of the at least one removed object does not contribute to completing the picking order, which defines a desired number and/or desired type of objects in the target loading aid. Furthermore, a device and a computer program product for performing the presented method is specified.

An information processing apparatus includes an acquisition unit acquiring a first image and a second image, the first image being an image of a target area in an initial state, the second image being an image of the target area where a first object conveyed from a supply area is placed, an estimation unit estimating one or more second areas in the target area, based on a feature of a first area estimated using the first image and the second image, the first area being where the first object is placed, the one or more second areas each being an area where an object in the supply area can be placed and being different from the first area. A control unit controls a robot to convey a second object different from the first object from the supply area to any of the one or more second areas.

A portable device repair machine may be configured to receive portable devices, identify the portable device, diagnose, repair and/or upgrade the portable device, collect payment for the service and return the portable device to the payee. The portable device repair machine may include an audio/video interface for interaction with a portable device user as well as a mechanized connection, inspection, diagnostic and upgrade system. The mechanized connection, inspection, diagnostic and upgrade system may comprise a conveyance mechanism for conveying the portable device among multiple stations such as a physical inspection station, device connection mechanism station, diagnostic, repair and upgrade station, and a physical upgrade station. Various mechanical devices, such as robotic arms may interact with the portable device, based in part on feedback obtained via various image sensors within the portable device repair machine.

A method for picking and placing items includes the steps of: providing a picking conveyor transporting items to be picked; providing a placing conveyor to which the items are to be placed; and providing a plurality of robots configured to move the items from pick positions on the picking conveyor to place positions on the placing conveyor. For at least one of the plurality of robots there is defined an actual work area A.sub.ac that fulfils the condition A.sub.ac<A.sub.th−(A.sub.ol+A.sub.ex), wherein A.sub.th is a theoretical work area, A.sub.ol is an overlapping work area and A.sub.ex is an excessive work area of the respective robot. By limiting the actual work area A.sub.ac of the robots more than what is done conventionally, the total workload between the robots in pick and place systems may be balanced.

A dynamic movement analytics system is disclosed that includes a programmable motion device including an end-effector, and a perception transfer system for receiving an object from the end-effector and for moving the object toward a distribution conveyance system moving in a first direction at a first speed. The perception transfer system includes at least one perception system for providing perception data regarding any of weight, shape, pose authority, position authority or identity information regarding the object and for moving the object along the first direction at the first speed for transferring the object to the distribution conveyance system.

Feedback information is an important aspect in all machine learning systems. In robot systems that are picking objects from a plurality of objects this has been arranged by acquiring images of objects that have been picked. When images are acquired after picking they can be imaged accurately and the information about picking and also dropping success can be improved by using acquired images as a feedback in the machine learning arrangement being used for controlling the picking and dropping of objects.

An article picking system includes a detection device which detects a position of a plurality of articles which are moved, and a control unit, and the control unit performs work data creation processing which creates work data having position data of each of the articles, work data storage processing which stores the plurality of the created work data, region determination processing which determines determination region on the periphery of watching-target work data, which should be paid attention to, among the stored work data, and order determination processing which determines picking order of the articles by using the watching-target work data and the peripheral work data, which are within the determination region.

The invention provides a controllable gripper for a robot. A plurality of gripping members, e.g. suction cups, are arranged to engage with a surface of an object to be gripped. The gripping members are arranged in a controllable gripping configuration. A base part is arranged to be moved by a robotic actuator, e.g. a gantry type of robotic actuator. Two or more arms connected to the base part are slidably arranged along their lengths relative to the base part. Each arm has a gripping member at or near its distal end. A controllable actuator system is arranged to control position of the arms in different directions, relative to the base part, thus allowing various gripping configurations to be formed with respect to at least size. Especially, four arms of fixed length, each with a suction cup, may be actuated by one or two electric motors, so a to allow the four suction cups to form various gripping quadrangle sizes. This allows the gripper to be capable of gripping small and large objects in random order, since the gripping configuration can be rapidly changed. Still, the gripper can be formed rather simple with few elements, and yet it is flexible and can be very compact in a compressed state of the arms to allow navigation into narrow spaces, e.g. to grip an object in a 3D bulk of objects. The gripper is preferably mounted to a robotic actuator by means of a controllable rotation and tilting arrangement.

A box handling system is disclosed for use in an object processing system. The box handling system includes a box tray including a recessed area for receiving a box, and the recessed area includes a plurality of floor and edge portions for receiving the box that contains objects to be processed.