A learning method is performed by a computers The method includes: receiving a first image that includes an object; generating a first rectangle in the first image, the first rectangle including therein a figure, that is set in advance to have a first inclination and that represents a gripping position of an object, and having a side parallel to a first direction; inputting the first image to a model, which outputs, from the input image, a rectangle parallel to the first direction and an inclination, to cause the model to output a second rectangle and a second inclination; and updating the model such that errors of the second rectangle and the second inclination with respect to the first rectangle and the first inclination respectively decrease.

The present disclosure is a robot system including a robot that performs work on a workpiece; a controller that controls the robot; a first camera that captures an image of the workpiece while being moved relative to the workpiece by means of the operation of the robot; and a second camera that is capable of acquiring, in synchronization with image capturing by the first camera, an image that represents the relative positional relationship between the first camera and the workpiece. The controller includes a correcting unit that corrects, on the basis of the image acquired by the second camera, the image-capturing timing of the first camera so that an image is captured at a position at which the workpiece is appropriately captured in the field of view of the first camera.

A non-transitory computer-readable storage medium storing a computer program controls a processor to execute (a) processing of recognizing a worker motion from an image of one or more worker motions captured by an imaging apparatus, (b) processing of recognizing hand and finger positions in a specific hand and finger motion when the worker motion contains the specific hand and finger motion, (c) processing of recognizing a position of a workpiece after work, and (d) processing of generating a control program for a robot using the worker motion, the hand and finger positions, and the position of the workpiece.

The present invention relates to pick planning for robotic picking applications to improve efficiency of automated picking operations and reduce robot down time. A pick plan is computed by obtaining data of a pick scene, processing the obtained data to identify objects and determine features associated with the objects, and determining an order and pick instructions based on the features. A computed pick plan may be periodically verified by reacquiring data of the pick scene and comparing the reacquired data with previous pick scene data in order to determine if a pick plan remains appropriate or should be updated or discarded and recomputed.

A medical safety-system for dynamic 3D healthcare environments, a medical examination system with motorized equipment, an image acquisition arrangement, and a method for providing safe movements in dynamic 3D healthcare environments. The medical safety-system for dynamic 3D healthcare environments includes a detection system, a processing unit, and an interface unit. The detection system includes at least one sensor arrangement to provide depth information of at least a part of an observed scene. The processing unit includes a correlation unit to assign the depth information and a generation unit to generate a 3D free space model to provide the 3D free space model.

A method for operating a robot includes providing target data for a target object; determining whether a pre-pick target for the target object is reachable by the robot; determining whether a pick target is reachable by the robot; and executing a pick routine directing the robot to pick up the target object and deposit the target object at a desired location responsive to a determination that the pre-pick target and the pick target are reachable by the robot.

A part manufacturing system and a method of manufacturing are provided. The system includes one or more part-moving robots, each having an end effector that grips a part. An operation robot performs an operation on the part while the part-moving robot holds the part. A fixed vision system is located apart from the robots and has at least one fixed vision sensor that senses an absolute location of the part and/or the end effector and generates a fixed vision signal representative of the absolute location. A controller collects the fixed vision signal and compares the absolute location with a predetermined desired location of the part and/or the end effector. The controller sends a repositioning signal to the part-moving robot if the absolute location varies from the predetermined desired location by at least a predetermined threshold, and the part-moving robot is configured to move the part upon receiving the repositioning signal.

The present invention relates to a robot teaching system based on image segmentation and surface electromyography and robot teaching method thereof, comprising a RGB-D camera, a surface electromyography sensor, a robot and a computer, wherein the RGB-D camera collects video information of robot teaching scenes and sends to the computer; the surface electromyography sensor acquires surface electromyography signals and inertial acceleration signals of the robot teacher, and sends to the computer; the computer recognizes a articulated arm and a human joint, detects a contact position between the articulated arm and the human joint, and further calculates strength and direction of forces rendered from a human contact position after the human joint contacts the articulated arm, and sends a signal controlling the contacted articulated arm to move along with such a strength and direction of forces and robot teaching is done.

A device and a method for controlling a robotic device, including a control model. The control model includes a robot trajectory model, which for the pickup includes a hidden semi-Markov model with one or multiple initial states, a precondition model, which for each initial state of the robot trajectory model includes a probability distribution of robot configurations before the pickup is carried out, and an object pickup model, which for a depth image outputs a plurality of pickup robot configurations having a respective associated probability of success.

An information processing device includes a connection device and a hardware processor connected to the connection device. The connection device connects to an imaging device. The imaging device images a region on which an object to be handled by using a load handling device is placed. The hardware processor acquires image information about the region via the connection unit. The hardware processor detects, from the image information, an identification region in which identification information attached to the object is included. The hardware processor recognizes the identification information in the identification region. The hardware processor outputs region information related to the identification region and also outputs the identification information.