A machine learning device according to the present invention learns an operation condition of a robot that stores a plurality of objects disposed on a carrier device in a container using a hand for grasping the objects. The machine learning device includes a state observation unit for observing the positions and postures of the objects and a state variable including at least one of cycle time to store the objects in the container and torque and vibration occurring when the robot grasps the objects during operation of the robot; a determination data obtaining unit for obtaining determination data for determining a margin of each of the cycle time, the torque, and the vibration against an allowance value; and a learning unit for learning the operation condition of the robot in accordance with a training data set constituted of a combination of the state variable and the determination data.

A carrier device includes a conveyor configured to carry objects supplied continuously; an object detection unit configured to detect the positions and orientations of the objects disposed in a predetermined area on the conveyor; a combination calculation unit, when a robot grasps multiple objects, out of the objects, with a hand and places the grasped objects in a container, configured to calculate combinations of sequences to grasp the objects by the hand; an index calculation unit configured to calculate an index for each combination using the distances and rotation amounts between the objects to be grasped by the hand based on the positions and orientations of the objects; and a robot control unit configured to determine the sequences to grasp the objects by the hand based on the indexes, and grasping the objects and placing the objects in the containers in accordance with the determined sequences.

A controller of a material handling system performs a method of creating a multidrop pattern of articles for robotic placement in layers on a pallet. A pattern is presented on a user interface of any currently positioned representations of articles on a pallet. A control affordance for inputting drag'n'drop and numeric inputs is presented on the user interface for robotic control operations to perform a multidrop of the more than one article in an end effector of a robotic arm for placement of the more than one article. User inputs are received that indicate placement position of a first subset of the more than one article. User inputs are received that indicate placement position of a second subset, which is mutually exclusive of the first subset, of the more than one article. The user inputs are converted into a place sequence of robotic control operations to perform a multidrop of the articles by the robotic arm.

An article conveying device having a temporary placement section and capable of conveying articles. The article conveying device has: a supplying section configured to sequentially convey or supply plural articles; a discharging section configured to sequentially convey plural containers for containing the articles; a temporary placement section on which at least one article can be temporarily placed; a first detecting section configured to detect a position/posture of the article on the supplying section and successively detect an amount of movement of the supplying section; a second detecting section configured to detect a position/posture of the container on the discharging section and successively detect an amount of movement of the discharging section; a working machine configured to convey the article between the supplying section, the temporary placement section and the discharging section; and a controlling section configured to control the working machine based on a predetermined condition.

A robot system includes a detector for detecting the position and posture of a workpiece; a robot for performing a predetermined operation on the workpiece; and a robot control device. The robot control device includes an area divider for dividing an operation area into a plurality of areas; an area determiner for determining in which area the workpiece is present; a learning controller for learning an operation speedup ratio to speed up an operation by varying speed or acceleration on an area-by-area basis in which the workpiece is present; a memory for storing the position of the workpiece and the operation speedup ratio; and a controller that performs the operation on a new workpiece using the learned operation speedup ratio when the operation has been learned in the area having the new workpiece, and makes the learning controller learn the operation speedup ratio when the operation has not been learned.

A method for optimizing an automated process to select and grip an object by a robot in an arrangement that includes a plurality of robots with regard to a specifiable optimization criterion, wherein the objects to be potentially gripped irregularly occur with respect to their spatial position and a time of their arrival, where detection of objects to be potentially gripped by robots is performed, detection of a priority characteristic as well as an assignment to one of the robots for the objects to be potentially gripped via an automated learning algorithm, taking the optimization criterion into account, and where selection and gripping depending on the assignment and the priority characteristic is implemented.