Robot

• US 20150321346A1
• Filed: 04/22/2015
• Published: 11/12/2015
• Est. Priority Date: 05/08/2014
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: MORIYA, Toshio, FUJIMOTO, Keisuke, KIMURA, Nobutaka, FUJI, Taiki
• Granted Patent: US 9,802,310 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: B25J 9/10 characterised by positioning means for manipulator elements
• CPC Class Codes: G05D 1/024 in combination with a laser lasers per se H01S
• CPC Class Codes: G05D 1/0274 using mapping information stored in a memory device navigation using map-matching G01C21/30
• CPC Class Codes: Y10S 901/01 Mobile robot
• CPC Class Codes: Y10S 901/47 Optical

Abstract

A robot is provided. A measurement unit measures an object in a space. An own-position estimation unit calculates an own position of a mobile unit moving through the space by matching measurement data acquired by the measurement unit with map data representing information about the object in the space. A classification unit acquires class-based measurement data obtained by classifying the measurement data into at least two predetermined classes. A class-based own-position estimation unit estimates an own position of the mobile unit for each of the at least two predetermined classes by matching a class-based map having a position of the objects recorded therein for each class with the class-based measurement data. An integration unit configured to integrate the own positions estimated for the respective at least two predetermined classes. A control unit controls the mobile unit based on information from the integration unit.

Claims

1. A robot comprising:

a measurement unit configured to measure an object in a space;

a storage unit configured to store map data representing information about the object in the space determined in advance;

an own-position estimation unit configured to calculate an own position of a mobile unit moving through the space by matching measurement data acquired by the measurement unit with the map data;

a classification unit configured to acquire class-based measurement data obtained by classifying the measurement data into at least two predetermined classes;

a class-based own-position estimation unit configured to estimate an own position of the mobile unit for each of the at least two predetermined classes by matching a class-based map having a position of the objects recorded therein for each class with the class-based measurement data;

an integration unit configured to integrate the own positions estimated for the respective at least two predetermined classes; and

a control unit configured to control the mobile unit based on information from the integration unit.

2. The robot according to claim 1,

wherein the class-based map is recorded for each class in a coordinate system determined in advance, the class-based map defining a comparative positional relationship between the coordinate systems as a coordinate conversion amount, and

wherein the integration unit is configured to integrate the own positions estimated for the respective at least two predetermined classes based on the coordinate conversion amount.

3. The robot according to claim 1, wherein the integration unit comprises means for changing an integration method of the own positions for the respective at least two predetermined classes based on a movement mode set in advance.

4. The robot according to claim 1, wherein the integration unit is configured to integrate the own positions based on a predetermined weighting set for each class.

5. The robot according to claim 1, wherein the integration unit is configured to use the own position estimated by a class having a high priority based on a predetermined priority set for each class.

6. The robot according to claim 3, further comprising:

a movement mode determination map having a predetermined movement mode for each location in the space set therein in advance; and

a movement mode determination unit configured to select the movement mode shown in the movement mode determination map corresponding to the own position.

7. The robot according to claim 6, wherein the movement mode determination unit is configured to determine the own position of a class for which most measurement data is obtained of the class-based measurement data.

8. The robot according to claim 1, further comprising at least one articulated robot,

wherein the at least one articulated robot is controlled based on the own positions.

9. The robot according to claim 1, further comprising an update unit configured to update, based on a class calculated by the classification unit from information about an object newly measured by the measurement unit, a map having an object classified in the class recorded therein.

10. The robot according to claim 9, wherein the update unit is configured to update the map based on an update degree conferred in advance based on the class.

11. The robot according to claim 1,

wherein the measurement unit is capable of measuring a predetermined attribute of the object in the space,

wherein the at least two predetermined classes are classified based on the predetermined attribute to be measured by the measurement unit, and

wherein the classification unit is configured to divide the measurement data into classes based on the predetermined attribute.

12. The robot according to claim 1, wherein the classification unit is configured to divide the measurement data into classes by geometrically matching the measurement data with a predetermined shape, and extracting a portion similar to the predetermined shape.

13. The robot according to claim 11, wherein the classification unit is configured to extract characteristic points relating to a corner and an edge from the measurement data, and perform classification based on a size and a planarity of a shape existing between the characteristic points.

14. The robot according to claim 1,

wherein the at least two predetermined classes are further divided into a still object and a columnar object each having a fixed predetermined shape in the space, and

wherein the classification unit is configured to determine a class by geometrically comparing each portion of the measurement data with each of the still object and the columnar object.

15. The robot according to claim 1,

wherein the at least two predetermined classes are further classified into a moving object that is still moving, and

wherein the classification unit is configured to compare the measurement data measured immediately before with measurement data measured previously, and set a portion that has a difference as the moving object.