ENVIRONMENT MAP GENERATING METHOD AND MOBILE ROBOT (AS AMENDED)

• US 20100324769A1
• Filed: 01/18/2008
• Published: 12/23/2010
• Est. Priority Date: 02/13/2007
• Status: Active Grant

• Current Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Original Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Inventors: Takaoka, Yutaka
• Granted Patent: US 7,987,021 B2
• Time in Patent Office: Days
• US Class Current: 701/25
• CPC Class Codes: B25J 9/1676 Avoiding collision or forbidden zones
• CPC Class Codes: G05D 1/024 in combination with a laser lasers per se H01S
• CPC Class Codes: G05D 1/0251 extracting 3D information from a plurality of images taken from different locations, e.g. stereo vision stereoscopic image analysis H04N13/00; depth recovery from images G06T7/593
• CPC Class Codes: G05D 1/0274 using mapping information stored in a memory device navigation using map-matching G01C21/30
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59

Abstract

To reduce a calculation time required for generating an environment map in a mobile robot. A mobile robot (100) generates a three-dimensional position data set representing an external environment using measurement information obtained from a range sensor (10). Next, based on an old environment map which is the environment map generated in the past and an amount of movement of the mobile robot (100), at least three-dimensional position data belonging to an area determined as the obstacle area in the old environment map, and three-dimensional position data belonging to an unobserved area which is not included in the old environment map are selected as plane-detection target data from among three-dimensional position data included in the three-dimensional position data set. Then, plane detection is executed using the plane-detection target data. Then, the movable area and the obstacle area which are included in the plane-detection target data are recognized using the results of the plane detection. Lastly, a new environment map is generated by integrating the movable area and obstacle area, which are recognized using the results of the plane detection, with an area determined as the movable area in the old environment map.

Claims

1. A method of generating an environment map for generating an environment map concerning an environment in which a mobile robot moves, the environment map specifying an obstacle area in which an obstacle exists and a movable area in which it is estimated that the mobile robot is movable, the method comprising:

generating a three-dimensional position data set indicating a position of a measurement object existing in the environment, by using measurement information obtained from a vision sensor;

selecting, based on an old environment map which is the environment map generated in the past and an amount of movement of the mobile robot, at least three-dimensional position data belonging to an area determined as the obstacle area in the old environment map, and three-dimensional position data belonging to an unobserved area which is not included in the old environment map, from among three-dimensional position data included in the three-dimensional position data set, as plane-detection target data;

performing plane detection using the plane-detection target data;

recognizing the movable area and the obstacle area which are included in the plane-detection target data, by using results of the plane detection; and

generating a new environment map which is an environment map newly generated by integrating the movable area and the obstacle area, which are recognized using the results of the plane detection, with an area determined as the movable area in the old environment map.

2. The method of generating an environment map according to claim 1, wherein selection of the plane-detection target data is performed by transforming a coordinate system of the old environment map into a coordinate system of the three-dimensional position data set in accordance with the amount of movement of the mobile robot, and by determining overlapping of the old environment map obtained after coordinate transformation and the three-dimensional position data set.

3. The method of generating an environment map according to claim 1, wherein, in the selection of the plane-detection target data, three-dimensional position data positioned near the area determined as the obstacle area in the old environment map is included in the plane-detection target data.

11. The method of generating an environment map according to claim 2, wherein, in the selection of the plane-detection target data, three-dimensional position data positioned near the area determined as the obstacle area in the old environment map is included in the plane-detection target data.

4. (canceled)

5. A mobile robot which moves in an environment in which an obstacle exists, by utilizing an environment map specifying an obstacle area in which the obstacle exists and a movable area in which it is estimated that no obstacle exists and the mobile robot is movable, the mobile robot comprising:

a vision sensor that visually recognizes the environment;

a three-dimensional position data generation unit that generates a three-dimensional position data set indicating a position of a measurement object existing in the environment, by using measurement information obtained from the vision sensor;

a plane-detection target data selection unit that selects plane-detection target data to be subjected to plane detection from among the three-dimensional position data set representing the environment, based on an old environment map which is the environment map generated in the past and an amount of movement of the mobile robot;

a plane detection unit that performs plane detection using the plane-detection target data;

an area recognition unit that recognizes the movable area and the obstacle area which are included in the plane-detection target data, by using results of the plane detection; and

an environment map update generation unit that generates a new environment map which is an environment map newly generated by integrating the movable area and the obstacle area, which are recognized by the area recognition unit, with an area determined as the obstacle area in the environment map generated in the past,

wherein the plane-detection target data selection unit selects as the plane-detection target data at least three-dimensional position data belonging to the area determined as the obstacle area in the old environment map, and three-dimensional position data belonging to an unobserved area which is not included in the old environment map, from among three-dimensional position data included in the three-dimensional position data set.

8. The mobile robot according to claim 5, wherein the vision sensor is a range sensor that obtains as the measurement information a range image having a range value as a pixel value of each pixel of a two-dimensional image, or a stereo camera that obtains a parallax image as the measurement information.

6. A mobile robot which moves in an environment in which an obstacle exists, by utilizing an environment map specifying an obstacle area in which the obstacle exists and a movable area in which it is estimated that no obstacle exists and the mobile robot is movable, the mobile robot comprising:

a vision sensor that visually recognizes the environment;

a three-dimensional position data generation unit that generates a three-dimensional position data set indicating a position of a measurement object existing in the environment, by using measurement information obtained from the vision sensor;

a plane-detection target data selection unit that selects plane-detection target data to be subjected to plane detection from among the three-dimensional position data set representing the environment, based on an old environment map which is the environment map generated in the past and an amount of movement of the mobile robot;

a plane detection unit that performs plane detection using the plane-detection target data;

an area recognition unit that recognizes the movable area and the obstacle area which are included in the plane-detection target data, by using results of the plane detection; and

an environment map update generation unit that generates a new environment map which is an environment map newly generated by integrating the movable area and the obstacle area, which are recognized by the area recognition unit, with an area determined as the obstacle area in the environment map generated in the past,

wherein the plane-detection target data selection unit performs selection of the plane-detection target data by transforming a coordinate system of the old environment map into a coordinate system of the three-dimensional position data set in accordance with the amount of movement of the mobile robot, and by determining overlapping of the old environment map obtained after coordinate transformation and the three-dimensional position data set.

9. The mobile robot according to claim 6, wherein the vision sensor is a range sensor that obtains as the measurement information a range image having a range value as a pixel value of each pixel of a two-dimensional image, or a stereo camera that obtains a parallax image as the measurement information.

7. A mobile robot which moves in an environment in which an obstacle exists, by utilizing an environment map specifying an obstacle area in which the obstacle exists and a movable area in which it is estimated that no obstacle exists and the mobile robot is movable, the mobile robot comprising:

a vision sensor that visually recognizes the environment;

a three-dimensional position data generation unit that generates a three-dimensional position data set indicating a position of a measurement object existing in the environment, by using measurement information obtained from the vision sensor;

a plane-detection target data selection unit that selects plane-detection target data to be subjected to plane detection from among the three-dimensional position data set representing the environment, based on an old environment map which is the environment map generated in the past and an amount of movement of the mobile robot;

a plane detection unit that performs plane detection using the plane-detection target data;

an area recognition unit that recognizes the movable area and the obstacle area which are included in the plane-detection target data, by using results of the plane detection; and

an environment map update generation unit that generates a new environment map which is an environment map newly generated by integrating the movable area and the obstacle area, which are recognized by the area recognition unit, with an area determined as the obstacle area in the environment map generated in the past,

wherein the plane-detection target data selection unit allows three-dimensional position data positioned near the area determined as the obstacle area in the old environment map to be included in the plane-detection target data.

10. The mobile robot according to claim 7, wherein the vision sensor is a range sensor that obtains as the measurement information a range image having a range value as a pixel value of each pixel of a two-dimensional image, or a stereo camera that obtains a parallax image as the measurement information.