Robot control device, robot control method, and robot control program

• US 20060173577A1
• Filed: 12/03/2003
• Published: 08/03/2006
• Est. Priority Date: 12/10/2002
• Status: Active Grant

• Current Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Original Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Inventors: Yokoyama, Taro, Takeda, Masanori
• Granted Patent: US 7,873,448 B2
• Time in Patent Office: Days
• US Class Current: 700/245
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: G06T 7/73 using feature-based methods
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

The content of the voice designation of a designator is recognized by a voice recognition part 130 at the time of controlling a robot, and the content of the indication of gesture or the like is recognized by an image recognition part 120. A movement destination and a map around the specific position designated are referred from a map data base 150 registering the position of an obstacle and the current position estimated by a self-position estimation part 140. After movement ease is decided by a movement ease decision part 112, the behavior is decided. When movement designation is given to the robot, correspondence according to a situation can be performed.

Claims

1. A robot control device for controlling a robot having a microphone, an imaging device and a self-position detection device, comprising:

a voice recognition part for recognizing the designation content of a designator based on sounds collected by the microphone;

an image recognition part for recognizing the designation content of the designator based on an image imaged by the imaging device;

a self-position estimation part for estimating the current position of the robot based on an output from the self-position detection device;

a map data base for retaining map data registering at least the position of an obstacle;

a decision part for deciding whether the movement to a specific position is required based on the recognition result of the voice recognition part and image recognition part;

a movement ease decision part for deciding movement ease to the specific position based on the current position estimated by the self-position estimation part and the map data when the decision part decides that the movement to the specific position is required;

a behavior decision part for deciding the behavior according to the movement ease decided by the movement ease decision part; and

a behavior control part for executing the behavior according to the decision of the behavior decision part.

2. The robot control device according to claim 1, wherein the movement ease decision part reads the position of the obstacle surrounding the movement route to the specific position from the map data base, and sets at least two or more areas based on the distance from the obstacle, and the behavior decision part decides the behavior according to an area containing the specific position and an area where the robot exists.

3. The robot control device according to claim 1, wherein the movement ease decision part including:

an obstacle recognition part for recognizing the obstacle surrounding the movement route to the specific position from the current position of the robot estimated by the self-position estimation part and the map data;

a warning area set part for setting an area having a possibility to interfere with an obstacle as a warning area when the robot exists, based on the position of the obstacle recognized by the obstacle recognition part;

a margin area set part for setting an area with a predetermined distance from the warning area as a margin area; and

a safety area set part for setting an area distant from the margin area from the obstacle as a safety area, and wherein

the movement ease to the specific position is decided based on the area including the specific position and the area where the robot exists, respectively, applicable to either one of the warning area, the margin area and the safety area.

4. The robot control device according to claim 3, wherein the warning area set part decides the position of a circle where a distance between representative points of the surface of the obstacle is set as a diameter, and sets the warning area using the position of the circle.

5. The robot control device according to claim 1, wherein the behavior decision part decides at least any one behavior of movement, the response of movement refusal, the reconfirmation of designation, stop, deceleration and acceleration.

6. The robot control device according to claim 1, wherein the voice recognition part has a designating range specification part for narrowing a designating area using a reference term, and the behavior decision part recognizes a specific position from the area of the logical product of the designating area narrowed by the designating range specification part and designating area recognized by the image recognition part.

7. The robot control device according to claim 1, further comprising a behavior schedule transmission part for making the behavior control part output a behavior schedule.

8. A robot control method for controlling a robot having a microphone, an imaging device and a self-position detection device, comprising the steps of:

a designation content discriminating step of recognizing the designation content of a designator based on sounds collected by the microphone, recognizing the designation content of the designator based on an image imaged by the imaging device, estimating the current position of the robot based on an output from the self-position detection device, and deciding whether the designation of the movement to a specific position is required from the designation content recognized by the sounds and the designation content recognized from the image;

a movement ease deciding step of deciding the movement ease to the specific position based on the current position estimated based on the output from the self-position detection device and map data provided as a map data base and registering at least the position of an obstacle when the designation content discriminating step decides that the movement to the specific position is required;

a behavior deciding step of deciding the behavior according to the movement ease decided by the movement ease deciding step; and

a behavior controlling step of executing the behavior according to the decision of the behavior deciding step.

9. The robot control method according to claim 8, wherein the movement ease deciding step reads the position of the obstacle surrounding the movement route to the specific position from the map data base, and sets at least two or more areas based on the distance from the obstacle, and

the behavior deciding step decides the behavior according to an area including the specific position and an area where the robot exists.

10. The robot control method according to claim 8, wherein the movement ease deciding step includes:

an obstacle recognizing step of recognizing the obstacle surrounding the movement route to the specific position from the current position of the robot estimated based on the self-position detection device and the map data;

a warning area setting step of setting an area having a possibility to interfere with an obstacle when the robot exists based on the position of the obstacle recognized by the obstacle recognizing step as a warning area;

a margin area setting step of setting an area with a predetermined distance from the warning area as a margin area;

a safety area setting step of setting an area distant from the margin area from the obstacle as a safety area; and

wherein

a step of deciding the movement ease to the specific position based on the area including the specific position and the area where the robot exists, respectively applicable to either one of the warning area, the margin area and the safety area.

11. The robot control method according to claim 10, wherein the warning area setting step decides the position of a circle where a distance between representative points of the surface of the obstacle is set as a diameter, and sets the warning area using the position of the circle.

12. The robot control method according to claim 8, wherein the behavior deciding step decides at least any one behavior of movement, the response of movement refusal, the reconfirmation of designation, stop, deceleration and acceleration.

13. The robot control method according to claim 8, wherein the designating content recognizing step includes a designating range specifying step of narrowing a designating area using a reference term contained in the sound, and an image recognizing step of narrowing the designating area from the image, and the behavior deciding step recognizes the specific position from the area of the logical product of the designating area narrowed by the designating range specifying step and designating area recognized by the image recognizing step.

14. The robot control method according to claim 8, wherein the behavior controlling step has a behavior schedule transmitting step for outputting the behavior schedule decided by the behavior deciding step.

15. A robot control program for making a computer mounted on a robot execute a method according to any one of claims 8 to 14 so as to control the robot having a microphone, an imaging device and a self-position detection device.

16. A robot control program for making a computer mounted on a robot function as a voice recognition means for recognizing the designation content of a designator based on sounds collected by a microphone,

an image recognition means for recognizing the designation content of the designator based on the image imaged by an imaging device,

a self-position estimation means for estimating the current position of the robot based on the output from the self-position detection device,

a decision means for deciding whether the movement to a specific position is required based on the recognition result of the voice recognition means and image recognition means, and the current position of the robot estimated by the self-position estimation part,

a movement ease decision means for deciding the movement ease to the specific position based on the current position estimated by the self-position estimation means and map data provided as a map data base and registering at least the position of an obstacle when the decision means decides that the movement to the specific position is required, and

a behavior decision means for deciding the behavior according to the movement ease decided by the movement ease decision means so as to control the robot having the microphone, the imaging device and the self-position detection device.