System and method for computing a probability that an object comprises a target using segment points

Patent metadata

Field	Value
Publication number	US-11087239-B2
Application number	US-201916679247-A
Country	US
Kind code	B2
Filing date	Nov 10, 2019
Priority date	Dec 16, 2016
Publication date	Aug 10, 2021
Grant date	Aug 10, 2021

Abstract

A method for computing a probability that an object comprises a target includes: performing a scan of an area comprising the object, generating points; creating a segment corresponding to the object using the points as segment points, the segment extending from a first segment point to a last segment point, the segment comprising a plurality of the segment points; and applying a metric, computing the probability that the object comprises the target.

First claim

What is claimed is:

1. A method for computing a probability that an object comprises a target, comprising:
performing a scan of an area comprising the object, generating points;
creating a segment corresponding to the object using the points as segment points, the segment extending from a first segment point to a last segment point, the segment comprising a plurality of the segment points;
applying a metric, computing the probability that the segment comprises the target, wherein the target comprises one or more of a human, a human appendage, a robot, a robot appendage, a forklift, a wall, a cart, a shelf, and a chair;
using the computed probability, identifying the segment as one or more of target and non-target;
removing a segment that corresponds to a non-moving object; and
filtering the segment to integrate the classification with knowledge about one or more of locations of humans and locations of non-humans.

2. The method of claim 1 , wherein the area is situated in an environment of a robot.

3. The method of claim 1 , wherein the metric comprises one or more of a distance between a point and the segment, a number of lines needed to cover points in the segment, a number of lines needed to cover all points in a segment to within a threshold distance, a best fit linear regression to the segment, a best fit circular approximation to the segment, and another metric.

4. The method of claim 1 , wherein the computing step is performed for each segment point.

5. A method for computing a probability that an object comprises a target, comprising:
performing a scan of an area, generating points;
creating a segment corresponding to the object using the points as segment points, the segment extending from a first segment point to a last segment point, the segment comprising a plurality of the segment points;
adding the segment to a candidate set of lines;
for a segment point, computing a point-segment distance from the point to the segment;
determining that the point-segment distance is not less than a threshold distance;
finding the farthest for a segment point, computing a point-segment distance from the point to the segment;
determining that the point-segment distance is not less than a threshold distance;
finding the farthest point that comprises the point that is farthest from the segment;
updating a metric usable to compute the probability that the object comprises the target, wherein the target comprises one or more of a human, a human appendage, a robot, a robot appendage, a forklift, a wall, a cart, a shelf, and a chair;
using the computed probability, identifying the segment as one or more of target and non-target;
removing a segment that corresponds to a non-moving object; and
filtering the segment to integrate the classification with knowledge about one or more of locations of humans and locations of non-humans.

6. A method for computing a probability that an object comprises a target, comprising:
performing a scan of an area, generating points;
creating a segment corresponding to the object using the points as segment points, the segment extending from a first segment point to a last segment point, the segment comprising a plurality of the segment points;
adding the segment to a candidate set of lines;
for at least one segment point, computing a point-segment distance from the point to the segment;
determining that the point-segment distance is less than a threshold distance;
updating a metric usable to compute the probability that the object comprises the target, wherein the target comprises one or more of a human, a human appendage, a robot, a robot appendage, a forklift, a wall, a cart, a shelf, and a chair;
using the computed probability, identifying the segment as one or more of target and non-target;
removing a segment that corresponds to a non-moving object; and
filtering the segment to integrate the classification with knowledge about one or more of locations of humans and locations of non-humans.

7. The method of claim 6 , wherein the area is situated in an environment of a robot.

8. The method of claim 6 , wherein the metric comprises one or more of a distance between a point and the segment, a number of lines needed to cover points in the segment, a number of lines needed to cover all points in a segment to within a threshold distance, a best fit linear regression to the segment, a best fit circular approximation to the segment, and another metric.

9. The method of claim 6 , wherein the computing step is performed for each segment point.

10. A method for computing a probability that an object comprises a target, comprising:
creating a segment corresponding to the object, the segment extending from a first segment point to a last segment point, the segment comprising a plurality of segment points, using points obtained in a scan of an area comprising the segment;
adding the segment to a candidate set of lines;
for at least one segment point, computing a point-segment distance from the point to the segment;
determining that the point-segment distance is less than a threshold distance;
updating a metric usable to compute the probability that the object comprises the target, wherein the target comprises one or more of a human, a human appendage, a robot, a robot appendage, a forklift, a wall, a cart, a shelf, and a chair;
using the computed probability, identifying the segment as one or more of target and non-target;
removing a segment that corresponds to a non-moving object; and
filtering the segment to integrate the classification with knowledge about one or more of locations of humans and locations of non-humans.

11. The method of claim 1 , wherein the filtering step comprises a sub-step of consulting a map that indicates where the non-moving objects of interest are located.

12. The method of claim 11 , wherein the consulting sub-step comprises performing one or more of reducing a weight accorded to the non-moving objects and completely removing the non-moving objects.

13. The method of claim 11 , wherein the filtering step comprises a sub-step, performed prior to the consulting sub-step, of generating the map using one or more of simultaneous localization and mapping (SLAM) and another mapping method.

14. The method of claim 11 , wherein the filtering step comprises using a costmap to reduce a weight of the non-moving objects of interest.

15. The method of claim 5 , wherein the filtering step comprises a sub-step of consulting a map that indicates where the non-moving objects of interest are located.

16. The method of claim 15 , wherein the consulting sub-step comprises performing one or more of reducing a weight accorded to the non-moving objects and completely removing the non-moving objects.

17. The method of claim 6 , wherein the filtering step comprises a sub-step of consulting a map that indicates where the non-moving objects of interest are located.

18. The method of claim 17 , wherein the consulting sub-step comprises performing one or more of reducing a weight accorded to the non-moving objects and completely removing the non-moving objects.

19. The method of claim 17 , wherein the filtering step comprises a sub-step, performed prior to the consulting sub-step, of generating the map using one or more of simultaneous localization and mapping (SLAM) and another mapping method.

20. The method of claim 10 , wherein the filtering step comprises a sub-step of consulting a map that indicates where the non-moving objects of interest are located.

Assignees

• Fetch Robotics Inc

Inventors

• Henning Alex
• Ferguson Michael
• Wise Melonee

Classifications

• G06N20/00 Primary

  Machine learning · CPC title

• G06N3/008

  based on physical entities controlled by simulated intelligence so as to replicate intelligent life forms, e.g. based on robots replicating pets or humans in their appearance or behaviour · CPC title

• Y10S901/01

  Mobile robot · CPC title

• Y10S901/46

  Sensing device · CPC title