Defining boundary for detected object

Patent metadata

Field	Value
Publication number	US-11074464-B2
Application number	US-201916522349-A
Country	US
Kind code	B2
Filing date	Jul 25, 2019
Priority date	Jul 25, 2019
Publication date	Jul 27, 2021
Grant date	Jul 27, 2021

Abstract

A computer includes a processor and a memory storing instructions executable by the processor to receive data from a sensor specifying a plurality of points, the points including a plurality of first points that describe an object; define a boundary surrounding the first points while minimizing a volume of space that is both (i) contained by the boundary and (ii) identified as unoccupied; and actuate a component with respect to a vehicle based on the boundary.

First claim

What is claimed is:

1. A computer comprising a processor and a memory storing instructions executable by the processor to:
receive data from a sensor specifying a plurality of points, the points including a plurality of first points that describe an object;
define a boundary surrounding the first points while minimizing a volume of space that is both (i) contained by the boundary and (ii) identified as unoccupied, wherein defining the boundary includes calculating the volume; and
actuate a component with respect to a vehicle based on the boundary.

2. The computer of claim 1 , wherein the points are described in the sensor data as three-dimensional points, and the instructions further include to project the points into two-dimensional horizontal space before defining the boundary.

3. The computer of claim 1 , wherein the instructions further include to identify space as unoccupied upon determining that the sensor has an unobstructed view through the space to one of the points.

4. The computer of claim 1 , wherein the boundary is a rectangular bounding box.

5. The computer of claim 1 , wherein the instructions further include to generate a convex hull surrounding the first points, and to define the boundary as a rectangular bounding box based on the convex hull.

6. The computer of claim 5 , wherein
the instructions further include to generate a plurality of candidate rectangular bounding boxes based on the convex hull;
defining the boundary includes selecting a first candidate rectangular bounding box from the candidate rectangular bounding boxes; and
of the candidate rectangular bounding boxes, the first candidate rectangular bounding box contains a minimum volume of space that is both (i) contained by the respective candidate rectangular bounding box and (ii) identified as unoccupied.

7. The computer of claim 6 , wherein the convex hull includes a plurality of line segments, each candidate rectangular bounding box includes four sides, and each candidate rectangular bounding box has one side overlapping one line segment of the convex hull and has a minimum length and width to surround all the first points.

8. The computer of claim 1 , wherein the vehicle includes the computer, and the component is at least one of a steering system, a brake system, or a propulsion of the vehicle.

9. The computer of claim 1 , wherein the object is the vehicle, and the component is an infrastructure component in proximity to the vehicle.

10. A method comprising:
receiving data from a sensor specifying a plurality of points, the points including a plurality of first points that describe an object;
defining a boundary surrounding the first points while minimizing a volume of space that is both (i) contained by the boundary and (ii) identified as unoccupied, wherein defining the boundary includes calculating the volume; and
actuating a component with respect to a vehicle based on the boundary.

11. The method of claim 10 , wherein the points are described in the sensor data as three-dimensional points, the method further comprising projecting the points into two-dimensional horizontal space before defining the boundary.

12. The method of claim 10 , further comprising identifying space as unoccupied upon determining that the sensor has an unobstructed view through the space to one of the points.

13. The method of claim 10 , wherein the boundary is a rectangular bounding box.

14. The method of claim 10 , further comprising generating a convex hull surrounding the first points, and defining the boundary as a rectangular bounding box based on the convex hull.

15. The method of claim 14 , further comprising generating a plurality of candidate rectangular bounding boxes based on the convex hull;
wherein defining the boundary includes selecting a first candidate rectangular bounding box from the candidate rectangular bounding boxes; and
of the candidate rectangular bounding boxes, the first candidate rectangular bounding box contains a minimum volume of space that is both (i) contained by the respective candidate rectangular bounding box and (ii) identified as unoccupied.

16. The method of claim 15 , wherein the convex hull includes a plurality of line segments, each candidate rectangular bounding box includes four sides, and each candidate rectangular bounding box has one side overlapping one line segment of the convex hull and has a minimum length and width to surround all the first points.

17. The method of claim 10 , wherein the component is at least one of a steering system, a brake system, and a propulsion of the vehicle.

18. The method of claim 10 , wherein the object is the vehicle, and the component is an infrastructure component in proximity to the vehicle.

19. A system comprising:
a computer programmed to
receive data from a sensor specifying a plurality of points, the points including a plurality of first points that describe an object; and
define a boundary surrounding the first points while minimizing a volume of space that is both (i) contained by the boundary and (ii) identified as unoccupied, wherein defining the boundary includes calculating the volume; and

means for actuating a component with respect to a vehicle based on the boundary.

20. The computer of claim 1 , wherein
the instructions further include to generate a plurality of candidate boundaries,
defining the boundary includes selecting a first candidate boundary from the candidate boundaries; and
of the candidate boundaries, the first candidate boundary contains a minimum volume of space that is both (i) contained by the respective candidate boundary and (ii) identified as unoccupied.

Assignees

• Ford Global Tech Llc

Inventors

• Lee Sangjin
• Krishnan Krishanth
• Vora Ankit Girish
• Agarwal Siddharth

Classifications

• G06V10/82

  using neural networks · CPC title

• G06V20/64 Primary

  Three-dimensional [3D] objects · CPC title

• G06V20/582

  of traffic signs · CPC title

• G06V20/584 Primary

  of vehicle lights or traffic lights · CPC title

• G06V20/58

  Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads · CPC title