IP Library Granted Patent US 12699179
Granted Patent B2
US 12699179 · App. 17/731,021 · Granted Aug 4, 2026

Method for generating radar projections to represent angular uncertainty

Inventors: Charles Christopher Ochoa (Los Altos, CA); Arjun Bhargava (San Francisco, CA); Chao Fang (Los Altos, CA); Kun-Hsin Chen (Los Altos, CA); Kuan-Hui Lee (San Jose, CA)
Assignees: TOYOTA RESEARCH INSTITUTE, INC.; TOYOTA JIDOSHA KABUSHIKI KAISHA
G01S13/867G01S13/89G01S13/931
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12699179
App. No.
17/731,021
Filed
Apr 27, 2022
Granted
Aug 4, 2026
Kind
B2
Art Unit
3646
USPC
342/52
Abstract

The disclosure generally relates to methods for gathering radar measurements, wherein the radar measurements includes one or more angular uncertainties, generating a two dimensional radar uncertainty cloud, wherein the radar uncertainty cloud includes one or more shaded regions that each represent an angular uncertainty, capturing image data, wherein the image data includes one or more targets within a region of interest, and fusing the two dimensional radar uncertainty cloud with the image data to overlay the one or more regions of uncertainty over a target.

Claims (41)

1 . A method of fusing radar measurements and image data comprising:

transmitting a radar signal towards a target within a region of interest;

gathering radar measurements, wherein the radar measurements include one or more angular uncertainties;

generating a radar uncertainty cloud representing uncertainty of the radar signal in relation to the target using probabilistic data correlated to an angle of signal delivery, wherein

the radar uncertainty cloud includes one or more shaded regions that represent the one or more angular uncertainties;

capturing image data, wherein the image data includes the target; and

fusing the radar uncertainty cloud with the image data by resampling the radar measurements and normalizing overlapping radar measurements to overlay the one or more shaded regions of the radar angular uncertainty cloud over the target.

2 . The method of claim 1 , wherein the one or more shaded regions include an outer region of uncertainty and an inner region of uncertainty.

3 . The method of claim 2 , wherein the inner region of uncertainty includes radar measurements indicative of a center of a radar point cloud.

4 . The method of claim 3 , wherein the outer region of uncertainty includes radar measurements less indicative of the center of the radar point cloud compared to the inner region of uncertainty.

5 . The method of claim 2 , wherein the inner region of uncertainty is represented in a two dimensional cloud as a first shaded region, and the outer region of uncertainty is represented in the two dimensional cloud as a second shaded region.

6 . The method of claim 5 , wherein the first shaded region is darker than the second shaded region.

7 . The method of claim 5 , wherein the radar uncertainty cloud extends outwardly in a first direction for a first length and outwardly in a second direction for a second length, wherein the first direction and second direction are generally perpendicular.

8 . The method of claim 7 , wherein the first length is greater than the second length.

9 . A method of fusing radar measurements and image data comprising:

transmitting a radar signal towards a target within a region of interest;

gathering radar measurements, wherein the radar measurements include one or more angular uncertainties;

generating a two dimensional radar uncertainty cloud representing uncertainty of the radar signal in relation to the target using probabilistic data correlated to an angle of signal delivery, wherein

the two-dimensional radar uncertainty cloud includes one or more shaded regions of uncertainty that each represent one of the one or more angular uncertainties;

capturing image data, wherein the image data includes one or more targets within a region of interest; and

fusing the two-dimensional radar uncertainty cloud with the image data by resampling the radar measurements and normalizing overlapping radar measurements to overlay the one or more shaded regions of angular uncertainty over one or more targets.

10 . The method of claim 9 , wherein the two dimensional radar uncertainty cloud includes an outer region of uncertainty and an inner region of uncertainty.

11 . The method of claim 10 , wherein the inner region of uncertainty includes radar signal data indicative of a center of a vertical portion of a radar point cloud along a y-axis.

12 . The method of claim 11 , wherein the outer region of uncertainty includes radar measurements less indicative of the center of the vertical portion of the radar point cloud along the y-axis compared to the inner region of uncertainty.

13 . The method of claim 10 , wherein the inner region of uncertainty is represented in the radar uncertainty cloud as a first shaded region, and the outer region of uncertainty is represented in the radar uncertainty cloud as a second shaded region.

14 . The method of claim 13 , wherein the first shaded region is darker than the second shaded region.

15 . The method of claim 9 , wherein the two-dimensional radar uncertainty cloud extends outwardly in a first direction for a first length and outwardly in a second direction for a second length, wherein the first direction and second direction are generally perpendicular.

16 . The method of claim 15 , wherein the first length is greater than the second length.

17 . A method of fusing radar measurements and image data comprising:

transmitting one or more radar signals toward one or more targets within a region of interest;

gathering radar measurements regarding an approximate location of each electromagnetic wave, wherein the radar measurements include one or more angular uncertainties for each radar signal;

generating a radar uncertainty cloud for each radar signal representing uncertainty of the radar signal in relation to the target using probabilistic data correlated to an angle of signal delivery, wherein

each radar uncertainty cloud includes one or more shaded regions that each represent one or more angular uncertainties of each radar signal;

capturing image data, wherein the image data includes the one or more targets; and

fusing each radar uncertainty cloud with image data from each target by resampling the radar measurements and normalizing overlapping radar measurements to display one or more regions of angular uncertainty for each radar signal over each target.

18 . The method of claim 17 , wherein the one or more angular uncertainties include an outer region of uncertainty and an inner region of uncertainty.

19 . The method of claim 18 , wherein the inner region of uncertainty includes radar measurements indicative of a center of a vertical portion of a radar cloud along a y-axis.

20 . The method of claim 19 , wherein the outer region of uncertainty includes radar measurements less indicative of the center of the vertical portion of the radar cloud along the y-axis compared to the inner region of uncertainty.

21 . The method of claim 1 , wherein fusing the radar uncertainty cloud with the image data is performed to reduce or mitigate at least one of the one or more angular uncertainties for vehicle perception used to operate a vehicular system.

22 . The method of claim 9 , wherein fusing the two-dimensional radar uncertainty cloud with the image data is performed to reduce or mitigate at least one of the one or more angular uncertainties for operation of a vehicular system.

23 . The method of claim 17 , wherein fusing each radar uncertainty cloud with image data from each target is performed to reduce or mitigate at least one of the one or more angular uncertainties for vehicle perception used to operate a vehicular system.