IP Library Granted Patent US 12,594,964
Granted Patent B2
US 12,594,964 · App. 17/126,888 · Granted Apr 7, 2026

Method of and system for generating reference path of self driving car (SDC)

Inventor: Pavel Sergeevich Hud (Minsk, BY)
Assignee: DIRECT CURSUS TECHNOLOGY L.L.C
B60W60/0015G01C21/3461G01C21/3492G05D1/0088G05D1/0214B60W30/09B60W2420/408B60W2556/10
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 12,594,964
App. No.
17/126,888
Granted
Apr 7, 2026
Kind
B2
Abstract

A method and device for amending a reference path associated with a vehicle are disclosed. The method includes: (i) acquiring road segment data and reference path data, the reference path data having a plurality of anchor points defining the reference path along the road segment, (ii) using the road segment data and the reference path data to determine a safe deviation interval for each one of the plurality of anchor points, a given safe deviation interval for a given anchor point being indicative of an acceptable deviation of the vehicle from the reference path; and (iii) using the safe deviation intervals to determine an amended path for the vehicle instead of the reference path, such that by following the amended path the vehicle falls within the constraints of the road segment.

Claims (56)

1 . A computer-implemented method of amending a reference path associated with a vehicle, the reference path being a path along a road segment that the vehicle is estimated to follow, the method comprising:

acquiring road segment data and reference path data,

the road segment data being indicative of constraints of the road segment, and

the reference path data having a plurality of anchor points defining the reference path along the road segment;

using the road segment data and the reference path data to determine a respective safe deviation interval for each one of the plurality of anchor points, by:

generating a simulated representation of the vehicle for a respective anchor point from the plurality of anchor points on a simulated representation of the road segment, the generating comprising:

generating a bounding box occupying a substantially same area on the simulated representation of the road segment as a surface occupied by the vehicle:

generating a simulated representation of a rear axle of the vehicle within the bounding box;

determining a rotation axis of the vehicle as an axis extending through a midpoint of the simulated representation of the rear axle in the bounding box,

the rotation axis extending upwardly and perpendicular to a ground surface on which the vehicle is travelling;

rotating, at each one of the plurality of anchor points, the bounding box around the rotation axis by a pre-determined angle to generate an enlarged projection of the vehicle onto the road segment;

determining the simulated representation of the vehicle as the enlarged projection thereof; and

determining the respective safe deviation intervals of the respective anchor points from the plurality of anchor points based on the simulated representation of the vehicle on the simulated representation of the road segment,

a given safe deviation interval for the respective anchor point of the plurality of anchor points being an interval within which the respective anchor point is to be located in order for the vehicle to fall within the constraints of the road segment when passing the respective anchor point;

determining whether any anchor points of the plurality of anchor points fall outside of the respective safe deviation intervals;

in response to determining that one or more anchor points of the plurality of anchor points fall outside of their respective safe deviation intervals:

moving each of the one or more anchor points that fall outside of their respective safe deviation intervals to within their respective safe deviation interval; and

generating an amended path as a path extending through the plurality of anchor points after the moving.

2 . The method of claim 1 , wherein the respective safe deviation intervals form a safe deviation corridor for the vehicle, the safe deviation corridor defining a section of the road segment in which the vehicle falls within the constraints of the road segment.

3 . The method of claim 2 , wherein the amended path falls within the safe deviation corridor.

4 . The method of claim 1 , wherein the constraints comprise at least one of (i) physical constraints on the road segment, and (ii) road rule constraints on the road segment.

5 . The method of claim 1 , wherein the method is executed by an electronic device that is associated with a Self-Driving Car (SDC).

6 . The method of claim 1 , wherein the moving the one or more anchor points that fall outside the respective safe deviation intervals comprises moving the one or more anchor points to respective midpoints of the respective safe deviation intervals.

7 . The method of claim 1 , wherein the method further comprises generating the reference path.

8 . The method of claim 5 , wherein the SDC is different from the vehicle, and wherein the method further comprises generating the reference path by: accessing a database of historical maneuvers associated with the vehicle and generating a prediction of the reference path.

9 . The method of claim 5 , wherein the SDC is located in proximity to the vehicle on the road segment.

10 . The method of claim 9 , wherein the method further comprises:

using the amended path of the vehicle for modifying a current path of the SDC along the road segment.

11 . The method of claim 1 , wherein the simulated representation of the vehicle covers an artificially increased surface on the simulated representation of the road segment,

the artificially increased surface being larger than an actual surface that the vehicle covers of the road segment.

12 . The method of claim 1 , wherein the generating the amended path comprises:

determining amended anchor points for the respective safe deviation intervals,

such that a given amended anchor point falls within the respective safe deviation interval; and

determining, by the electronic device the amended path as being a sequence of the amended anchor points.

13 . The method of claim 12 , wherein the given amended anchor point corresponds to a midpoint of the respective safe interval.

14 . The method of claim 12 , wherein the sequence of the amended anchor points comprises a polyline.

15 . The method of claim 1 , further comprising defining the plurality of anchor points.

16 . The method of claim 15 , wherein the defining is based on at least a velocity of the vehicle.

17 . An electronic device for amending a reference path associated with a vehicle, the reference path being a path along a road segment that the vehicle is estimated to follow, the electronic device comprising at least one processor and at least one non-transitory computer-readable memory storing executable instructions, which, when executed by the at least one processor, cause the electronic device to:

acquire road segment data and reference path data,

the road segment data being indicative of constraints of the road segment,

the reference path data having a plurality of anchor points defining the reference path along the road segment;

use the road segment data and the reference path data to determine a respective safe deviation interval for each one of the plurality of anchor points, by:

generating a simulated representation of the vehicle for a respective anchor point from the plurality of anchor points on a simulated representation of the road segment, the generating comprising:

generating a bounding box occupying a substantially same area on the simulated representation of the road segment as a surface occupied by the vehicle;

generating a simulated representation of a rear axle of the vehicle within the bounding box;

determining a rotation axis of the vehicle as an axis extending through a midpoint of the simulated representation of the rear axle in the bounding box,

the rotation axis extending upwardly and perpendicular to a ground surface on which the vehicle is travelling:

rotating, at each one of the plurality of anchor points, the bounding box around the rotation axis by a pre-determined angle to generate an enlarged projection of the vehicle onto the road segment;

determining the simulated representation of the vehicle as the enlarged projection thereof; and

determining the respective safe deviation intervals of the respective anchor points from the plurality of anchor points based on the simulated representation of the vehicle on the simulated representation of the road segment,

a given safe deviation interval for the respective anchor point of the plurality of anchor points being an interval within which the respective anchor point is to be located in order for the vehicle to fall within the constraints of the road segment when passing the respective anchor point;

determine whether any anchor points of the plurality of anchor points fall outside of the respective safe deviation intervals;

in response to determining that one or more anchor points of the plurality of anchor points fall outside of their respective safe deviation intervals:

move each of the one or more anchor points that fall outside of their respective safe deviation intervals to within their respective safe deviation interval; and

generate an amended path as a path extending through the plurality of anchor points after the moving.

Assignments (5)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 9, 2024
From: DIRECT CURSUS TECHNOLOGY L.L.C
To: Y.E. HUB ARMENIA LLC
Reel/Frame 068534/0619 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 2, 2023
From: YANDEX SELF DRIVING GROUP LLC
To: DIRECT CURSUS TECHNOLOGY L.L.C
Reel/Frame 065447/0048 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 18, 2023
From: HUD, PAVEL SERGEEVICH
To: YANDEXBEL LLC
Reel/Frame 062415/0601 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 18, 2023
From: YANDEXBEL LLC
To: YANDEX.TAXI LLC
Reel/Frame 062415/0605 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 18, 2023
From: YANDEX.TAXI LLC
To: YANDEX SELF DRIVING GROUP LLC
Reel/Frame 062415/0608 →
Priority Claims (1)
RU RU2019143946 · Dec 25, 2019 · national
Continuity (1)
Related Publication 20210197862A1 · Jul 1, 2021
References Cited (31)
US 9227632B1 · Lee · 2016 [cited by examiner]
US 9310804B1 · Ferguson · 2016 [cited by examiner]
US 9911346B2 · Klinger et al. · 2018 [cited by applicant]
US 10928820B1 · Tao · 2021 [cited by examiner]
US 11513519B1 · Akella · 2022 [cited by examiner]
US 20050043881A1 · Brulle-Drews · 2005 [cited by examiner]
US 20110035148A1 · Machino · 2011 [cited by examiner]
US 20130090802A1 · Curtis · 2013 [cited by examiner]
US 20170247032A1 · Lee et al. · 2017 [cited by applicant]
US 20180188734A1 · Zhu · 2018 [cited by examiner]
US 20180189578A1 · Yang · 2018 [cited by examiner]
US 20190016339A1 · Ishioka · 2019 [cited by examiner]
US 20190278277A1 · Tao · 2019 [cited by examiner]
US 20200094825A1 · Kato · 2020 [cited by examiner]
US 20200174490A1 · Ogale · 2020 [cited by examiner]
US 20200398894A1 · Hudecek · 2020 [cited by examiner]
US 20210074037A1 · Zhu · 2021 [cited by examiner]
US 20210096568A1 · Kumar · 2021 [cited by examiner]
US 20210181741A1 · Yao · 2021 [cited by examiner]
US 20210192748A1 · Morales Morales · 2021 [cited by examiner]
CN 107563040A · 2018 [cited by applicant]
CN 108657179A · 2018 [cited by examiner]
CN 106647773B · 2019 [cited by applicant]
JP 5362691B2 · 2013 [cited by applicant]
JP 2015191650A · 2015 [cited by examiner]
KR 20170050499A · 2017 [cited by applicant]
Suzuki “Automatic Two-Lane Path Generation for Autonomous Vehicles Using Quartic B-Spline Curves”, IEEE Transactions on Intelligent Vehicles, vol. 3, No. 4, pp. 547-558, Dec. 2018, doi: 10.1109/TIV.2018.2874532. [cited by applicant]
Russian Search Report dated Feb. 12, 2021 issued in respect of the counterpart Russian Patent Application No. RU2019143946. [cited by applicant]
Jian, “High-Definition Map Combined Local Motion Planning and Obstacle Avoidance for Autonomous Driving”, 2019 EEE Intelligent Vehicles Symposium (IV), Jun. 9-12, 2019, pp. 2180-2186, doi: 10.1109/IVS.2019.8814229. [cited by applicant]
Kuwata, “Real-Time Motion Planning With Applications to Autonomous Urban Driving,” IEEE Transactions on Control Systems Technology, vol. 17, No. 5, pp. 1105-1118, Sep. 2009, doi: 10.1109/TCST.2008.2012116. [cited by applicant]
European Search Report dated Apr. 29, 2021 issued in respect of the counterpart European Patent Application No. EP20206620.5. [cited by applicant]