Robot, robot system, control device, and control method

Patent metadata

Field	Value
Publication number	US-9764467-B2
Application number	US-201514612701-A
Country	US
Kind code	B2
Filing date	Feb 3, 2015
Priority date	Feb 4, 2014
Publication date	Sep 19, 2017
Grant date	Sep 19, 2017

Abstract

A robot includes: a hand; and a control unit that operates the hand, in which the control unit generates three-dimensional point group information for a partial image forming a captured image obtained by an imaging unit, and causes the hand to hold an object included in the partial image.

First claim

What is claimed is:

1. A robot comprising:
a hand; and
a control unit that operates the hand,
wherein an imaging device captures an image of an object and generates three-dimensional image data, and
the control unit generates three-dimensional point group information for a partial image of a portion of the generated three-dimensional image data, calculates a position and a posture of the object based on the three-dimensional point group information, and causes the hand to hold the object included in the partial image.

2. The robot according to claim 1 ,
wherein the control unit
defines a first region indicating the object, and a second region including the first region and having a predetermined shape, from the captured image;
generates first three-dimensional point group information corresponding to the first region and second three-dimensional point group information corresponding to the second region on the basis of the captured image;
selects a set in which a length of a line segment between two points is within a predetermined length range from among a plurality of sets each of which is formed by two points included in the first three-dimensional point group information;
extracts points which are within a predefined range from the line segment between the two points forming the selected set, from the second three-dimensional point group information; and
selects a central axis in which a difference between a distance to each extracted point and a predetermined threshold value is within a predefined range, from the second three-dimensional point group information.

3. The robot according to claim 2 ,
wherein the control unit detects a region in which luminance is greater than a predefined threshold value from the captured image, and determines a region in which an index value indicating a shape of the detected region is within a predetermined range as the first region.

4. The robot according to claim 2 ,
wherein a shape of the second region is a rectangular shape, and
wherein the control unit
defines a third region whose length is larger than a length of the second region in a longitudinal direction from the captured image,
extracts third three-dimensional point group information corresponding to the third region from the captured image, and

calculates a length of the object in a direction of the central axis on the basis of a distribution of points included in the third three-dimensional point group information in the direction of the central axis.

5. A control device comprising:
an image reception unit that receives a captured image obtained by an imaging unit capturing an image of a plurality of objects and generates three-dimensional image data; and
a processing unit that operates a robot,
wherein the processing unit generates three-dimensional point group information for a partial image of a portion of the generated three-dimensional image data, calculates a position and a posture of at least one of the plurality of objects based on the three-dimensional point group information, and causes the robot to hold at least one of the plurality of objects included in the partial image.

6. A robot system comprising:
an imaging unit that captures an image of a plurality of objects and generates three-dimensional image data;
a robot; and
a control unit that operates the robot,
wherein the control unit generates three-dimensional point group information for a partial image of a portion of the generated three-dimensional image data, calculates a position and a posture of at least one of the plurality of objects based on the three-dimensional point group information, and causes the robot to hold at least one of the plurality of objects included in the partial image.

7. A control method comprising:
capturing an image of a plurality of objects and generating three-dimensional image data;
generating three-dimensional point group information for a partial image of a portion of the generated three-dimensional image data;
calculating a position and a posture of at least one of the plurality of objects based on the three-dimensional point group information; and
causing a robot to hold at least one of the objects included in the partial image by using the generated three-dimensional point group information.

Assignees

• Seiko Epson Corp

Inventors

• Harada Tomoki
• Hashimoto Koichi
• Arai Shogo

Classifications

• B25J9/1612 Primary

  characterised by the hand, wrist, grip control · CPC title

• G05B2219/40053

  Pick 3-D object from pile of objects · CPC title

• B25J9/1697

  Vision controlled systems · CPC title

• Y10S901/47

  Optical · CPC title

• G05B2219/39543

  Recognize object and plan hand shapes in grasping movements · CPC title