IP Library › Granted Patent US 12,241,996
Granted Patent B1
US 12,241,996 · App. 17/375,707 · Granted Mar 4, 2025

Antenna array calibration for vehicle radar systems

Inventors: Ryan Fan (Mountain View, CA); Kenneth Ho (Mountain View, CA)
Assignee: Waymo LLC
G01S7/40G01S7/4017G01S7/4021G01S7/412G01S7/415
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 12,241,996
App. No.
17/375,707
Granted
Mar 4, 2025
Kind
B1
Abstract

An example method for using antenna array calibration to adjust radar unit operation involves receiving radar data from a radar unit coupled to a vehicle during vehicle operation in an environment, where the radar unit receives the radar data from the environment via an antenna array of the radar unit. The method also involves detecting an object in the environment based on the radar data, determining that the detected object satisfies a set of conditions, and, in response to the set of conditions being satisfied, estimating a first phase array offset for the antenna array. The method also involves comparing the first phase array offset with a second phase array offset that represents a prior calibration for the radar unit, and, based on a difference between the first and second phase array offsets exceeding a threshold difference, adjusting operation of the radar unit according to the first phase array offset.

Claims (65)

1. A method comprising:

receiving, at a computing device, radar data from a radar unit coupled to a vehicle during vehicle operation in an environment, wherein the radar unit receives the radar data from the environment via an antenna array of the radar unit;

detecting an object in the environment based on the radar data;

determining that the detected object satisfies a set of conditions;

responsive to determining that the detected object satisfies the set of conditions, estimating a first phase array offset for the antenna array;

performing a comparison between the first phase array offset and a second phase array offset, wherein the second phase array offset represents a prior calibration for the radar unit;

based on a difference between the first phase array offset and the second phase array offset exceeding a threshold difference, adjusting operation of the radar unit according to the first phase array offset; and

based on the difference between the first phase array offset and the second phase array offset being less than the threshold difference, operating the radar unit according to the second phase array offset.

2. The method of claim 1 , wherein determining that the detected object satisfies the set of conditions comprises:

determining that radar data indicative of the detected object exceeds both a signal-to-noise ratio (SNR) threshold and a range threshold; and

determine that the radar data indicative of the detected object represents a Doppler for the detected object that is within a predefined Doppler range.

3. The method of claim 1 , wherein detecting the object in the environment comprises:

identifying a set of peaks such that each peak represents a different object in the environment; and

selecting training data corresponding to the identified set of peaks based on a plurality of channels that depend on operation of the antenna array of the radar unit.

4. The method of claim 3 , wherein estimating the first phase array offset for the antenna array of the radar unit comprises:

performing a smoothing process to each peak of the identified set of peaks, wherein the smoothing process includes one or more of a phase calibration, a motion compensation, and a linear phase compensation; and

based on performing the smoothing process to each peak of the identified set of peaks, extracting error data from the training data, wherein the error data represents respective portions of the training data that differ from the identified set of peaks; and

estimating the first phase array offset based on the extracted error data.

5. The method of claim 4 , wherein estimating the first phase array offset based on the extracted error data comprises:

determining a spatially invariant correction and a spatially variant correction based on the extracted error data; and

estimating the first phase array offset based on the determined spatially invariant correction and the determined spatially variant correction.

6. The method of claim 5 , wherein determining the spatially invariant correction and the spatially variant correction comprises:

determining the spatially invariant correction and the spatially variant correction using a singular value decomposition process.

7. The method of claim 5 , wherein adjusting operation of the radar unit according to the first phase array offset comprises:

adjust operation of the radar unit based on the determined spatially invariant correction and the determined spatially variant correction.

8. The method of claim 5 , further comprising:

based on the determined spatially invariant correction and the determined spatially variant correction, generating an array impulse response; and

estimating a peak sidelobe level and a root-mean-square (RMS) sidelobe level; and

wherein performing the comparison between the first phase array offset and the second phase array offset comprises:

performing a given comparison between the estimated peak sidelobe level and the RMS sidelobe level to an existing peak sidelobe level and an existing RMS sidelobe level associated with the second phase array offset.

9. The method of claim 1 , wherein the second phase array offset is based on an offline calibration process.

10. A system comprising:

a radar unit coupled to a vehicle; and

a computing device coupled to the vehicle, wherein the computing device is configured to:

receive radar data from the radar unit coupled to the vehicle during vehicle operation in an environment, wherein the radar unit receives the radar data from the environment via an antenna array of the radar unit;

detect an object in the environment based on the radar data;

determine that the detected object satisfies a set of conditions;

responsive to determining that the detected object satisfies the set of conditions, estimate a first phase array offset for the antenna array;

perform a comparison between the first phase array offset and a second phase array offset, wherein the second phase array offset represents a prior calibration for the radar unit;

based on a difference between the first phase array offset and the second phase array offset exceeding a threshold difference, adjust operation of the radar unit according to the first phase array offset; and

based on the difference between the first phase array offset and the second phase array offset being less than the threshold difference, operate the radar unit according to the second phase array offset.

11. The system of claim 10 , further comprising:

a memory coupled to the vehicle, wherein the computing device is further configured to:

store the first phase array offset in the memory based on the difference between the first phase array offset and the second phase array offset exceeding the threshold difference.

12. The system of claim 10 , wherein the computing device is further configured to:

determine that radar data indicative of the detected object exceeds both a signal-to-noise ratio (SNR) threshold and a range threshold; and

determine that the radar data indicative of the detected object represents a Doppler for the detected object that is within a predefined Doppler range.

13. The system of claim 10 , wherein the computing device is further configured to:

identify a set of peaks such that each peak represents a different object in the environment; and

select training data corresponding to the identified set of peaks based on a plurality of channels that depend on operation of the antenna array of the radar unit.

14. The system of claim 13 , wherein the computing device is further configured to:

perform a smoothing process to each peak of the identified set of peaks, wherein the smoothing process includes one or more of a phase calibration, a motion compensation, and a linear phase compensation; and

based on performing the smoothing process to each peak of the identified set of peaks, extract error data from the training data, wherein the error data represents respective portions of the training data that differ from the identified set of peaks; and

estimate the first phase array offset based on the extracted error data.

15. The system of claim 14 , wherein the first phase array offset includes a spatially invariant correction and a spatially variant correction determined based on the extracted error data.

16. The system of claim 15 , wherein the computing device is further configured to:

adjust operation of the radar unit based on the spatially invariant correction and the spatially variant correction.

17. The system of claim 10 , wherein the second phase array offset is based on an offline calibration process.

18. A method for calibrating a radar unit of a vehicle during operation of the vehicle comprising:

navigating a vehicle having a radar unit past a plurality of objects at a first time frame, wherein a computing system coupled to the vehicle uses radio signals received by an antenna array of the radar unit while the vehicle is navigating to identify each object, and wherein each object is associated with a distance, a speed, and a signal-to-noise ratio;

navigating the vehicle past the plurality of objects at a second time frame, wherein the second time frame occurs subsequent to the first time frame;

performing a comparison between a first phase array offset associated with first radar signals received from the antenna array of the radar unit during the first time frame and a second phase array offset associated with second radar signals received from the antenna array of the radar unit during the second time frame;

based on a difference between the first phase array offset and the second phase array offset exceeding a threshold difference, operating the radar unit according to the second phase array offset; and

based on the difference between the first phase array offset and the second phase array offset being less than the threshold difference, operating the radar unit according to the first phase array offset.

19. The method of claim 18 , wherein the plurality of object comprises at least one of: a car, a truck, a street sign, and a metallic object having a size equal to a car or larger.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 14, 2021
From: FAN, RYAN; HO, KENNETH
To: WAYMO LLC
Reel/Frame 056855/0695 →
References Cited (25)
US 9733350B2 · Stainvas Olshansky et al. · 2017 [cited by applicant]
US 10732260B2 · Ajanoh · 2020 [cited by applicant]
US 10739438B2 · Harrison · 2020 [cited by applicant]
US 10768304B2 · Englard et al. · 2020 [cited by applicant]
US 20200041612A1 · Harrison · 2020 [cited by applicant]
US 20200064441A1 · Alcalde · 2020 [cited by examiner]
US 20200191914A1 · Kunz et al. · 2020 [cited by applicant]
US 20200249315A1 · Eshet et al. · 2020 [cited by applicant]
US 20200249316A1 · Harrison · 2020 [cited by applicant]
US 20200371198A1 · Schoor · 2020 [cited by applicant]
US 20200400810A1 · Cho et al. · 2020 [cited by applicant]
US 20200408529A1 · Zeng et al. · 2020 [cited by applicant]
US 20210003404A1 · Zeng et al. · 2021 [cited by applicant]
US 20210072350A1 · Loesch · 2021 [cited by examiner]
CN 103018727A · 2013 [cited by applicant]
CN 102360528B · 2013 [cited by applicant]
CN 110378204A · 2019 [cited by applicant]
CN 110412559A · 2019 [cited by applicant]
CN 110824912A · 2020 [cited by applicant]
CN 111694019A · 2020 [cited by applicant]
EP 3511733A3 · 2019 [cited by applicant]
WO 2020113160A2 · 2020 [cited by applicant]
Randy S. Depoy Jr. et al., “Mitigating atmospheric phase-errors in SAL data using model-based reconstruction”, 2019 IEEE National Aerospace and Electronics Conference (NAECON), Jul. 2019. [cited by applicant]
Kihong Park et al., “High-Precision Depth Estimation Using Uncalibrated LiDAR and Stereo Fusion”, IEEE Transactions on Intelligent Transportation Systems, vol. 21, Issue 1, pp. 321-335, Jan. 2020. [cited by applicant]
Shijie Bai et al., “Nonlinear correction of frequency-modulated continuous-wave lidar frequency modulation based on singular value decomposition-least square algorithm”, SOptical Engineering, vol. 59, Issue 5, May 12, 2… [cited by applicant]