IP Library Granted Patent US 12,515,654
Granted Patent B2
US 12,515,654 · App. 17/222,015 · Granted Jan 6, 2026

Counter-steering penalization during vehicle turns

Inventors: Alice Kassar (Dearborn Heights, MI); Scott Julian Varnhagen (Ann Arbor, MI); Ramadev Burigsay Hukkeri (Pittsburgh, PA)
Assignee: Ford Global Technologies, LLC
B60W30/12B60W60/001G01S19/42B60W2552/53
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 12,515,654
App. No.
17/222,015
Granted
Jan 6, 2026
Kind
B2
Abstract

Devices, systems, and methods are provided for counter-steering penalization. A device may analyze, by one or more processors, lane geometry associated with one or more lanes at a geographic location. The device may identify one or more corners in the lane geometry. The device may select one or more desired maximum steering angles of a steering wheel of an autonomous vehicle. The device may select weight values associated with the one or more desired maximum steering angles. The device may execute a path planning optimization based on the one or more desired maximum steering angles and the weight values.

Claims (40)

1 . A method comprising:

detecting, by one or more processors, lane geometry associated with one or more lanes at a location;

identifying one or more corners in the lane geometry;

selecting one or more desired maximum steering angles of a steering wheel of an autonomous vehicle;

selecting one or more weight values associated with the one or more desired maximum steering angles; and

performing a path planning optimization to select a steering wheel angle as a function of time or distance wherein an objective function of the optimization penalizes, based on the one or more weight values, steering wheel angles that exceed, in absolute value, one or more of the desired maximum steering angles.

2 . The method of claim 1 , wherein the one or more desired maximum steering angles are associated with a time interval prior to a first corner of the one or more corners.

3 . The method of claim 1 , wherein the one or more desired maximum steering angles comprises a first desired maximum steering angle of a left turn and a second desired maximum steering angle of a right turn.

4 . The method of claim 3 , wherein the first desired maximum steering angle associated with the left turn is independent of the second desired maximum steering angle of the right turn.

5 . The method of claim 1 , wherein performing the path planning optimization comprises imposing a slack variable to control the path planning optimization.

6 . The method of claim 5 , wherein the slack value is based on an avoidance path of an object in a region associated with a corner of the one or more corners.

7 . The method of claim 1 , wherein a positive steering wheel angle is associated with left movements on the steering wheel, and a negative steering wheel angle is associated with right movements on the steering wheel.

8 . A non-transitory computer-readable medium storing computer-executable instructions, which when executed by one or more processors result in performing operations comprising:

detecting lane geometry associated with one or more lanes at a location;

identifying one or more corners in the lane geometry;

selecting one or more desired maximum steering angles of a steering wheel of an autonomous vehicle;

selecting one or more weight values associated with the one or more desired maximum steering angles; and

performing a path planning optimization to select a steering wheel angle as a function of time wherein an objective function of the optimization penalizes, one or more weight values, steering wheel angles that exceed, in absolute value, one or more of the desired maximum steering angles.

9 . The non-transitory computer-readable medium of claim 8 , wherein the one or more desired maximum steering angles are associated with a time interval prior to a first corner of the one or more corners.

10 . The non-transitory computer-readable medium of claim 8 , wherein the one or more desired maximum steering angles comprises a first desired maximum steering angle of a left turn and a second desired maximum steering angle of a right turn.

11 . The non-transitory computer-readable medium of claim 10 , wherein the first desired maximum steering angle associated with the left turn, is independent of the second desired maximum steering angle of the right turn.

12 . The non-transitory computer-readable medium of claim 8 , wherein to perform the path planning optimization further comprises applying the weight values to penalize the one or more desired maximum steering angles.

13 . The non-transitory computer-readable medium of claim 8 , wherein a steering angle of the steering wheel of the autonomous vehicle is less than the one or more desired maximum steering angles.

14 . The non-transitory computer-readable medium of claim 8 , wherein to perform the path planning optimization further comprises imposing a slack variable to control the path planning optimization.

15 . A system for controlling an operation of an autonomous vehicle, the system comprising:

a global positioning system sensor;

a memory device;

one or more autonomous driving sensors; and

an autonomous mode controller to:

determine a geographic location of the autonomous vehicle via the global positioning system sensor;

analyze lane geometry associated with one or more lanes at the geographic location;

identify one or more corners in the lane geometry;

select one or more desired maximum steering angles of a steering wheel of an autonomous vehicle;

select one or more weight values from the memory device associated with the one or more desired maximum steering angles; and

execute a path planning optimization to select a steering wheel angle as a function of time or distance wherein an objective function of the optimization penalizes, based on the one or more weight values, steering wheel angles that exceed, in absolute value, one or more of the desired maximum steering angles.

16 . The system of claim 15 , wherein the one or more desired maximum steering angles are associated with a time interval prior to a first corner of the one or more corners.

17 . The system of claim 15 , wherein the one or more desired maximum steering angles comprises a first desired maximum steering angle of a left turn and a second desired maximum steering angle of a right turn.

18 . The method of claim 1 , wherein the one or more weights are based on the presence of obstacles.

19 . The non-transitory computer-readable medium of claim 8 , wherein the one or more weights are based on the presence of obstacles.

20 . The system of claim 15 , wherein the one or more weights are based on the presence of obstacles.

Assignments (3)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 9, 2023
From: ARGO AI, LLC
To: FORD GLOBAL TECHNOLOGIES, LLC
Reel/Frame 062936/0548 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 9, 2023
From: ARGO AI, LLC
To: FORD GLOBAL TECHNOLOGIES, LLC
Reel/Frame 063025/0346 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 5, 2021
From: KASSAR, ALICE; VARNHAGEN, SCOTT JULIAN; HUKKERI, RAMADEV BURIGSAY
To: ARGO AI, LLC
Reel/Frame 055821/0511 →
Continuity (1)
Related Publication 20220314983A1 · Oct 6, 2022
References Cited (30)
US 10831210B1 · Kobilarov · 2020 [cited by applicant]
US 20040040781A1 · Fujioka · 2004 [cited by examiner]
US 20070184929A1 · Piyabongkarn et al. · 2007 [cited by applicant]
US 20170129535A1 · You · 2017 [cited by examiner]
US 20170364083A1 · Yang et al. · 2017 [cited by applicant]
US 20180253098A1 · Borinato · 2018 [cited by applicant]
US 20180281191A1 · Sinyavskiy et al. · 2018 [cited by applicant]
US 20180293449A1 · Sathyanarayana et al. · 2018 [cited by applicant]
US 20190005412A1 · Matus et al. · 2019 [cited by applicant]
US 20190107840A1 · Green et al. · 2019 [cited by applicant]
US 20190113929A1 · Mukadam et al. · 2019 [cited by applicant]
US 20190113931A1 · Zha et al. · 2019 [cited by applicant]
US 20190299978A1 · Balakrishnan et al. · 2019 [cited by applicant]
US 20190299983A1 · Shalev-Shwartz et al. · 2019 [cited by applicant]
US 20190324463A1 · Zhu · 2019 [cited by applicant]
US 20200001862A1 · Luo et al. · 2020 [cited by applicant]
US 20200159225A1 · Zeng et al. · 2020 [cited by applicant]
US 20200207368A1 · Takagi · 2020 [cited by applicant]
US 20200371518A1 · Kang · 2020 [cited by applicant]
US 20210070360A1 · Lee et al. · 2021 [cited by applicant]
US 20210078592A1 · Febbo · 2021 [cited by examiner]
US 20220009522A1 · Zhang · 2022 [cited by examiner]
US 20220063610A1 · Hafner · 2022 [cited by examiner]
WO 2020062033A1 · 2020 [cited by applicant]
International Search Report of PCT/US2022/023383 mailed Jul. 15, 2022, 3 pages. [cited by applicant]
Written Opinion of PCT/US2022/023383 mailed Jul. 15, 2022, 5 pages. [cited by applicant]
Cornet, R. “Safe Reinforcement Learning for Automated Vehicles”, Master of Science Thesis, Aug. 14, 2020, pp. 1-118, Delft University of Technology, Delft, Netherlands. [cited by applicant]
Nolte et al. “Model Predictive Control Based Trajectory Generation for Autonomous Vehicles—An Architectural Approach”, IEEE Intelligent Vehicles Symposium 2017, Aug. 10, 2017, pp. 1-8, IEEE, New York, NY. [cited by applicant]
Wolf et al. “Learning How to Drive in a Real World Simulation with Deep Q-Networks”, 2017 IEEE Intelligent Vehicles Symposium, Jun. 2017, pp. 1-8, IEEE, New York, NY. [cited by applicant]
B. Guenmn et. al., A Gentle Introduction to Optimization 1 (Cambridge University Press, 2014). [cited by applicant]