Mobile robot

• US 8,090,193 B2
• Filed: 12/03/2008
• Issued: 01/03/2012
• Est. Priority Date: 12/06/2007
• Status: Active Grant

• Current Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Original Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Inventors: Higaki, Nobuo, Watanabe, Ryujin
• Primary Examiner(s): Chawan, Sheela
• Application Number:
  US12/315,411
• Publication Number:
  US 20090148034A1
• Time in Patent Office: 1,126 Days
• Field of Search: 382/106, 382/153, 382117-118, 382/168, 382171-173, 382/190, 382/199, 382/203, 382/266, 382/274, 382/289, 382/296, 382/263, 382/264, 382/299, 382/254, 382/275, 382/155, 382/156, 382/181, 901/46, 901/47, 348/61, 348/143, 348/159, 348/113, 348/E7.086, 348/148, 348/E7.08, 348/154, 250/559.33, 345/8, 345/7, 356/27, 356/615, 700/245
• US Class Current: 382/153
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 30/2504 Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches

Abstract

There is disclosed a mobile robot including an image processor that generates recognition information regarding a target object included in a taken image, and a main controller integrally controlling the robot based on this recognition information. The image processor executes steps of: generating a low-resolution image and at least one high-resolution image whose resolution higher than that of the low-resolution image; generating first target object information regarding the target object from the low-resolution image; determining which high-resolution image should be processed if two or more high-resolution images are generated, and then defining a resolution process region in the low-resolution image; processing a region in the high-resolution region corresponding to the resolution process region in the low-resolution image, so as to generate second target object information in the high-resolution image; and determining whether or not the first and the second target object information are matched; and based on this determination, using at least either of the first and the second target object information, thereby to generate the recognition information.

Claims

1. A mobile robot comprising an image processor that processes an image taken by a vision sensor and generates recognition information regarding a target object to be recognized included in the taken image, and a controller that integrally controls the mobile robot based on the generated recognition information,

the image processor comprising:

a multiple-resolution image generation unit that down-sizes the taken image to generate a low-resolution image having a resolution lower than that of the taken image and at least one high-resolution image having a resolution higher than that of the low-resolution image;

a low-resolution image process unit that processes the low-resolution image generated by the multiple-resolution image generation unit, and generates first target object information regarding the target object included in the low-resolution image;

an image-process determination unit that determines whether or not to process the high-resolution image in accordance with predetermined input information or a predetermined rule, also determines which high-resolution image having what resolution should be processed if the multiple-resolution image generation unit generates two or more high-resolution images, and then defines a resolution process region at part of the low-resolution image;

a high-resolution image process unit that processes a region in the high-resolution region that corresponds to the resolution process region at part of the low-resolution image defined by the image-process determination unit so as to generate second target object information regarding the target object included in the high-resolution image; and

an image-process result generation unit that;

determines whether or not the first target object information generated by the low-resolution image process unit and the second target object information generated by the high-resolution image process unit are met, and

based on the determination of the first and second target object information, uses at least either of the first and the second target object information, first and the second target object information, thereby to generate the recognition information.

2. The mobile robot according to claim 1, wherein, in accordance with the predetermined rule, the image-process determination unit defines the resolution process region at a center of the low-resolution image.

3. The mobile robot according to claim 1, wherein, in accordance with the predetermined rule, the image-process determination unit repeatedly defines the resolution process region in the low-resolution image, shifting the resolution process region in the horizontal direction in the low-resolution image.

4. The mobile robot according to claim 1, wherein

the low-resolution image process unit extracts a face region of a person included in the low-resolution image as input information, and

the image-process determination unit defines the resolution process region on this extracted face region extracted by the low-resolution image process unit.

5. The mobile robot according to claim 1, wherein, if input traveling direction or a future traveling direction of the mobile robot is input to the image-process determination unit as the input information, the image-process determination unit defines the resolution process region in the low-resolution image depending on the input traveling direction or the future traveling direction of the mobile robot.

6. The mobile robot according to claim 1, wherein, if input traveling speed or a future traveling speed of the mobile robot is input to the image-process determination unit as the input information, the image-process determination unit is more likely to determine to process a higher-resolution image and narrows the resolution process region in the low-resolution image as the input traveling speed or the future traveling speed becomes faster.

7. The mobile robot according to claim 1, wherein, if a direction and a sound pressure of a sound source of the target object are input as the input information from a sound-source position determination unit that identifies the direction and the sound pressure of the sound source, the image-process determination unit defines the resolution process region in the low-resolution image depending on the direction of the sound source.

8. The mobile robot according to claim 7, wherein, if the input sound pressure becomes smaller, the image-process determination unit is more likely to determine to process a higher-resolution image and narrows the resolution process region in the low-resolution image.

9. The mobile robot according to claim 1, wherein, if a direction and a distance to a tag attached on an object to be detected are input from a object detector for detecting the direction and the distance to the tag, the image-process determination unit defines the resolution process region in the low-resolution image depending on the input direction of the tag.

10. The mobile robot according to claim 9, wherein the image-process determination unit is more likely to determine to process a higher-resolution image and narrows the resolution process region in the low-resolution image as the input distance of the tag becomes greater.

11. The mobile robot according to claim 1, wherein,

when the mobile robot executes a transport task to transport an article, the low-resolution image process unit extracts a predetermined transport container on which the article are placed, which is included in the low-resolution image, as input information,

the image-process determination unit determines whether or not a task being currently executed by the mobile robot is the transport task, and defines the resolution process region on the transport container extracted by the low-resolution image process unit.

12. The mobile robot according to claim 1, wherein,

when the mobile robot executes a guide task to guide a person, a guide direction in which the person should be guided is input to the image-process determination unit as input information,

the image-process determination unit determines whether or not a task being currently executed by the mobile robot is the guide task, and defines the resolution process region in the low-resolution image depending on the input guide direction.

13. The mobile robot according to claim 1, wherein the image-process result generation unit uses both the first and the second target object information if the first and the second target object information are matched;

otherwise uses only the second target object information if not matched, thereby to generate the recognition information.