Granted Patent B2

US 12,352,592 · App. 18/545,801 · Granted Jul 8, 2025

Inventors: Mingdong Wang (San Diego, CA); Chenzhe Qian (San Diego, CA); Xue Mei (San Diego, CA)

Assignee: TUSIMPLE, INC.

G01C21/3602G01C21/32G01S13/89G01S17/89G01S19/13

Abstract

Claims (62)

1. A method implemented by a processor, comprising:

generating a response map from a response image generated by an image device of one or more imaging devices of a vehicle, the response map comprising a plurality of response points indicating a location of the vehicle;

comparing the response map to a region map which corresponds to an approximated region of the vehicle and includes a plurality of region points to determine (a) differences between the plurality of response points and the plurality of region points and (b) a confidence score that comprises a variance that is computed using a logistic function and that represents a matching level between the response map and the region map; and

predicting a position of the vehicle based on a result of the comparing.

2. The method of claim 1 , wherein the plurality of region points is obtained using a Global Positioning System (GPS) device or an inertial measurement unit (IMU) device.

3. The method of claim 1 , wherein the generating the response map further comprises:

detecting lane markers in the response image, the lane markers pertaining to physical aspects contained in the response image; and

plotting the plurality of response points on the response map, the plurality of response points indicating locations of the lane markers.

4. The method of claim 1 , further comprising:

generating, based on the differences, the confidence score.

5. The method of claim 1 , wherein the response image is generated from radar sensing equipment, light detection and ranging (LIDAR) sensing equipment, Global Positioning System (GPS) sensing information, and/or images.

6. The method of claim 1 , wherein the region map and the response map are compared at a selected frequency.

7. The method of claim 6 , wherein the selected frequency is at least 20 cycles per second.

8. An apparatus, comprising:

a processor configured to:

generate a response map from a response image generated by an image device of one or more imaging devices of a vehicle, the response map comprising a plurality of response points indicating a location of the vehicle;

compare the response map to a region map which corresponds to an approximated region of the vehicle and includes a plurality of region points to determine (a) differences between the plurality of response points and the plurality of region points and (b) a confidence score that comprises a variance that is computed using a logistic function and that represents a matching level between the response map and the region map; and

predict a position of the vehicle based on a result of the comparing.

9. The apparatus of claim 8 , wherein the processor is further configured to:

output a predicted location of the vehicle to an advanced driver-assistance system (ADAS).

10. The apparatus of claim 8 , wherein each of the one or more imaging devices is adapted to perceive a different aspect of an environment of the vehicle.

11. The apparatus of claim 8 , wherein the variance is computed as:

variance

=

x

min

+

(

x

max

-

x

min

)

1

+

e

S

⁡

(

G

⁢

x

-

m

)

,

wherein x is a matching score of the response map, x min is a minimum value of the response map, x max is a maximum value of the response map, S is a steepness parameter, G is a growth rate parameter, and m is a midpoint of the response map.

12. The apparatus of claim 8 , wherein the imaging device comprises a light detection and ranging (LIDAR) device.

13. The apparatus of claim 8 , wherein the imaging device comprises a camera that captures images at a rate of at least 10 images per second.

14. A non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform a method comprising:

generating a response map from a response image generated by an image device of one or more imaging devices of a vehicle, the response map comprising a plurality of response points indicating a location of the vehicle;

comparing the response map to a region map which corresponds to an approximated region of the vehicle and includes a plurality of region points to determine (a) differences between the plurality of response points and the plurality of region points, and (b) a confidence score that comprises a variance that is computed using a logistic function and that represents a matching level between the response map and the region map; and

predicting a position of the vehicle based on a result of the comparing.

15. The non-transitory computer-readable storage medium of claim 14 , wherein the method further comprises:

comparing the confidence score to a threshold; and

outputting a predicted position of the vehicle based on the confidence score being lower than the threshold.

16. The non-transitory computer-readable storage medium of claim 14 , wherein the response map is generated using a camera pointing in a direction opposite to a direction of travel of the vehicle.

17. The non-transitory computer-readable storage medium of claim 14 , wherein lane markers in the response map are based on one or more physical structures.

18. The non-transitory computer-readable storage medium of claim 17 , wherein the one or more physical structures comprise a bridge, a barrier, a sign, or a building.

19. The non-transitory computer-readable storage medium of claim 17 , wherein the lane markers comprise at least one of solid lines, striped lines, or reflectors.

20. The non-transitory computer-readable storage medium of claim 14 , wherein the region map further comprises information related to a speed of a collection vehicle, a direction of travel of the collection vehicle, or a velocity of the collection vehicle relative to an expected roadway lane.

Assignments (3)

Continuity (3)

References Cited (400)