Detection system, processing apparatus, movement object, detection method, and non-transitory computer-readable storage medium

1 . A detection system, comprising:

processing circuitry configured to:

acquire point-cloud information corresponding to a plurality of positions of a detection target object using a light scanning device;

estimate a location and an attitude of the detection target object based on the point-cloud information using consistency with an outer shape model of the detection target object; and

output information relating to a movement target location based on an estimation result,

wherein the processing circuitry configured to estimate fits an outer shape model indicating an outer shape of the detection target object to a point cloud according to the point-cloud information, and uses only point-cloud information existing outside the outer shape model to estimate the location and the attitude of the detection target object,

wherein the detection system identifies the outside of the outer shape model based on a number of intersections between a half straight line connecting points as targets of inside-outside determination of the detection target object and an arbitrary point of the detection target object and a line segment group forming an outline of the outer shape model.

2 . A detection system, comprising:

processing circuitry configured to:

acquire point-cloud information corresponding to a plurality of positions of a detection target object using a light scanning device;

estimate a location and an attitude of the detection target object based on the point-cloud information using consistency with an outer shape model of the detection target object; and

output information relating to a movement target location based on an estimation result,

wherein the processing circuitry is configured to estimate fits an outer shape model indicating an outer shape of the detection target object to a point cloud according to the point-cloud information, and uses only point-cloud information existing outside the outer shape model to estimate the location and the attitude of the detection target object,

wherein the processing circuitry configured to estimate estimates the location and attitude by minimizing a distance between points, in the point cloud information outside of the outer shape model, and a line segment group forming an outline of the outer shape model.

3 . The detection system according to claim 2 , wherein the processing circuitry configured to estimate estimates the location and attitude of the detection target object by adjusting an arrangement of the outer shape model so as to make points determined to be at the outside of the outer shape model whose location and attitude are estimated based on the point-cloud information to become less.

4 . The detection system according to claim 2 , wherein the processing circuitry configured to estimate evaluates the estimation result of the location and attitude of the outer shape model by using the point cloud existing outside the outer shape of the outer shape model whose location and attitude are estimated.

5 . The detection system according to claim 2 ,

wherein a lateral surface of the detection target object includes a member for reflecting the light, and

a cross section of the member receiving the light for scanning is discretely arranged in a direction along the lateral surface of the detection target object.

6 . The detection system according to claim 5 , wherein the point cloud corresponding to the location of the member reflecting the light is discretely arranged in a direction along the outer shape of the outer shape model.

7 . The detection system according to claim 1 , wherein

the processing circuitry configured to estimate estimates the location and attitude by minimizing a distance between points, in the point cloud information outside of the outer shape model, and the line segment group forming an outline of the outer shape model.

8 . A processing apparatus, comprising:

processing circuitry configured to:

acquire point-cloud information corresponding to a plurality of positions of a detection target object using a light scanning device;

estimate a location and an attitude of the detection target object based on the point-cloud information using consistency with an outer shape model of the detection target object; and

output information relating to a movement target location based on an estimation result,

wherein the processing circuitry configured to estimate fits an outer shape model indicating an outer shape of the detection target object to a point cloud according to the point-cloud information, and uses only point-cloud information existing outside the outer shape model to estimate the location and the attitude of the detection target object, and

wherein the processing circuitry configured to estimate estimates the location and attitude by minimizing a distance between points, in the point cloud information outside of the outer shape model, and a line segment group forming an outline of the outer shape model.

9 . A movement object, comprising:

a detection system according to claim 1 ; and

a movement mechanism driving the movement object based on an estimation result of the location and attitude of the detection target object.

10 . The movement object according to claim 9 , further includes a distance sensor generating the point-cloud information,

wherein the estimation result in which the location and attitude of the detection target object is estimated is acquired by a detection result of the distance sensor.

11 . The movement object according to claim 10 , wherein the processing circuitry is further configured to:

extract point-cloud information referring to a table in which an extraction target region of the detection target object is defined to extract the point-cloud information in the extraction target region; and

control the movement mechanism based on the information of the location and attitude of the detection target object, and

wherein the processing circuitry configured to estimate estimates the location and attitude of the detection target object using the extracted point-cloud information as the point-cloud information.

12 . The movement object according to claim 9 , wherein an extraction target region of the detection target object is designated as a relative location with respect to a self-moving object or a relative location with respect to a surrounding environment of the self-moving object.

13 . The movement object according to claim 12 , wherein

the detection target object is a transportation target coupled with the self-moving object and transported, and

the processing circuitry configured to control determines a location where the self-moving object is coupled with the transportation target as a movement target of the self-moving object based on the estimation result of the detection target object and controls moving the self-moving object to the movement target of the self-moving object.

14 . The movement object according to claim 13 , wherein the processing circuitry configured to output refers to a table in which the location and attitude of the transportation target is defined to determine the movement object of the self-moving object in a case in which the transportation target is coupled with the self-moving object.

15 . The movement object according to claim 13 , wherein the processing circuitry configured to output refers to a table in which the location and attitude of the transportation target is defined to determine the movement object of a self-moving object in a case of arranging the transportation targets having the same shape to be aligned.

16 . The movement object according to claim 11 , further comprising a movement mechanism moving a location of the distance sensor, wherein the processing circuitry configured to control refers to the outer shape model to move the distance sensor to a location suitable for the detection of the detection target object.

17 . The movement object according to claim 9 ,

wherein the processing circuitry configured to estimate uses the point cloud existing outside the outer shape of the outer shape model whose location and attitude are estimated to evaluate the estimation result of the location and attitude of the outer shape model,

the processing circuitry configured to output refers to a table in which a relative location and attitude with respect to a self-moving object is defined to drive the movement mechanism based on the information of the relative location and attitude that are defined in the table in a case in which an evaluation as the estimation result equal to or better than a predetermined threshold value is not achieved, and

the processing circuitry configured to estimate estimates the location and attitude of the detection target object again based on a result of re-observing the same detection target object as a previous observation from a location and an attitude different from the location and attitude in the previous observation of the detection target object.

18 . The movement object according to claim 9 , wherein the detection target object is loaded such that a contour of the outer shape of the detection target object is divided at a height scanned by irradiating the light, a load can be loaded inside the lateral surface of the detection target object, and the load is in an observable state by a distance sensor provided at outside of the lateral surface of the detection target object.

19 . A detection method performed by a computer of a detection system, comprising:

scanning light to acquire point-cloud information corresponding to a plurality of positions of a detection target object;

using consistency with an outer shape model of the detection target object to estimate a location and an attitude of the detection target object based on the point-cloud information;

outputting information relating to a movement target location based on an estimation result; and

fitting an outer shape model indicating an outer shape of the detection target object to a point cloud according to the point-cloud information, and using only point-cloud information existing outside the outer shape model to estimate the location and the attitude of the detection target object,

wherein using consistency to estimate the location and attitude comprises minimizing a distance between points, in the point cloud information outside of the outer shape model, and a line segment group forming an outline of the outer shape model.

20 . The detection system according to claim 2 , wherein the processing circuitry configured to estimate estimates the location and attitude of the outer shape model so as to make points determined to be outside the outer shape model whose location and attitude are estimated based on the point-cloud information to become less.