Machine learning device, robot controller, robot system, and machine learning method for learning action pattern of human

Abstract

A machine learning device for a robot that allows a human and the robot to work cooperatively, the machine learning device including a state observation unit that observes a state variable representing a state of the robot during a period in that the human and the robot work cooperatively; a determination data obtaining unit that obtains determination data for at least one of a level of burden on the human and a working efficiency; and a learning unit that learns a training data set for setting an action of the robot, based on the state variable and the determination data.

Description

BACKGROUND OF THE INVENTION[0001]1. Field of the Invention[0002]The present invention relates to a machine learning device, a robot controller, a robot system, and a machine learning method for learning the action pattern of a human.[0003]2. Description of the Related Art[0004]In conventional techniques, to ensure human safety, safety measures are taken to forbid humans from entering the working area of a robot when the robot is driven. For example, a robot is enclosed with a safety fence to forbid humans to step over the safety fence during the period in which the robot is driven. In recent years, a robot system which allows a human to work cooperatively with a robot is known. The robot system allows the robot and the human to simultaneously perform one task without enclosing the robot with a safety fence.[0005]Japanese Laid-open Patent Publication No. 2015-123505 discloses an industrial robot which works cooperatively with a human. The robot includes a movable unit, a protective m...

AI Technical Summary

Benefits of technology

The patent describes a training system for a robot that uses a learning unit to determine the best actions for the robot based on its current state. The learning unit calculates a reward based on the obstacles the robot avoids and the time it takes to reach its destination. The goal is to make the robot move faster and more efficiently. The technical effect of this invention is to provide a better learning system for robotics that can quickly determine the best actions based on its current state.

Problems solved by technology

Even when, for example, the same task is performed, the level of fatigue of a human varies and the level of burden also varies depending on the velocity or the distance from the human.

This makes it difficult to set an optimal method for controlling a robot in accordance with the working details.

Images