IP Library Granted Patent US 12,043,284
Granted Patent B2
US 12,043,284 · App. 17/049,112 · Granted Jul 23, 2024

Trajectory planning

Inventor: Christopher Ostafew (Mountain View, CA)
Assignee: Nissan North America, Inc.
B60W60/0011B60W30/09B60W30/0956B60W30/146B60W40/04B60W60/001B60W60/0015B60W60/00274G06V20/58B60W2554/20B60W2554/4044B60W2554/4045
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 12,043,284
App. No.
17/049,112
Granted
Jul 23, 2024
Kind
B2
Abstract

Object avoidance by an autonomous vehicle (AV) is disclosed. A method includes detecting a first object along a coarse driveline of a drivable area of the AV; receiving a predicted path of the first object; determining, based on the predicted path of the first object, an adjusted drivable area; and determining a trajectory of the AV through the adjusted drivable area. A system includes a trajectory planner configured to detect a first object along a coarse driveline of a drivable area of the AV; receive a predicted path of the first object; determine, based on the predicted path of the first object, an adjusted drivable area; and determine a trajectory of the AV through the adjusted drivable area.

Claims (100)

1. A method for object avoidance by an autonomous vehicle (AV), comprising:

identifying nearby objects to the AV;

extracting a drivable area of the AV;

cutting out, to obtain an initial adjusted drivable area, cutout portions of the drivable area, the cutout portions corresponding to nearby objects that are classified as static objects, wherein the cutting out comprises:

dividing the drivable area into bins, each having original hard boundaries and soft boundaries, wherein the original hard boundaries correspond to nearby objects and the soft boundaries correspond to lawful restraints, defined with respect to a longitudinal direction of travel of the AV;

assigning each of the nearby objects to a respective bin based on a respective location of the each of the nearby objects determined using sensors of the AV;

tracking the nearby objects within the respective bin over time; and

for each bin that is assigned at least one of the nearby objects and while maintaining the soft boundaries, adjusting the original hard boundaries of the bin to exclude an area occupied by the at least one of the nearby objects thereby defining a modified drivable area within the bin, wherein the excluded area constitutes a cutout portion;

identifying a first object along a coarse driveline of the initial adjusted drivable area of the AV, the first object being a nearby object that is classified as a dynamic object;

receiving a predicted path of the first object;

determining, based on the predicted path of the first object, an adjusted drivable area by adjusting, based on the predicted path of the first object, the initial adjusted drivable area to obtain the adjusted drivable area;

determining a trajectory of the AV through the adjusted drivable area by prioritizing confining the trajectory within the adjusted hard boundaries but permitting crossing the soft boundaries when the trajectory cannot be confined within the adjusted hard boundaries due to collisions with the first object, wherein crossing the soft boundaries is permitted if crossing the soft boundaries is safe; and

controlling the AV according to the trajectory.

2. The method of claim 1 , wherein determining, based on the predicted path of the first object, the adjusted drivable area comprises:

assigning the first object to one bin of the bins; and

determining the adjusted drivable area based on the one bin.

3. The method of claim 2 , wherein determining, based on the predicted path of the first object, the adjusted drivable area comprises:

determining respective object locations of the first object at discrete time intervals;

determining respective AV locations of the AV at the discrete time intervals; and

determining the adjusted drivable area based on the respective object locations and the respective AV locations.

4. The method of claim 3 , wherein the first object is determined to be an oncoming longitudinal constraint; and wherein controlling the AV according to the trajectory comprises:

stopping the AV until a second time subsequent to a first time that an object location of the first object meets an AV location of the AV.

5. The method of claim 3 , wherein the first object is determined to be a longitudinal constraint; and wherein controlling the AV according to the trajectory comprises:

slowing the AV to drive behind the first object.

6. The method of claim 2 , further comprising:

classifying the first object as a lateral constraint, a longitudinal constraint, an oncoming lateral constraint, or an oncoming longitudinal constraint.

7. The method of claim 1 , wherein determining the trajectory of the AV through the adjusted drivable area comprises:

determining that the first object causes a static blockage; and wherein controlling the AV according to the trajectory comprises:

stopping the AV at the static blockage.

8. The method of claim 1 , further comprising:

detecting a second object, wherein determining the trajectory of the AV through the adjusted drivable area comprises:

determining that the first object is passable on a first side; and

determining that the second object is passable on the first side and a second side; and

wherein controlling the AV according to the trajectory comprises:

controlling the AV to pass the first object on the first side and to pass the second object on the first side.

9. A system for object avoidance by an autonomous vehicle (AV), comprising:

a processor configured to:

identify nearby objects to the AV;

extract a drivable area of the AV;

cut out, to obtain an initial adjusted drivable area, cutout portions of the drivable area, the cutout portions corresponding to nearby objects that are classified as static objects, wherein to cut out the cutout portions comprises to:

divide the drivable area into bins, each having original hard boundaries and soft boundaries, wherein the original hard boundaries correspond to nearby objects and the soft boundaries correspond to lawful restraints, defined with respect to a longitudinal direction of travel of the AV;

assign each of the nearby objects to a respective bin based on a respective location of the each of the nearby objects;

track the nearby objects within the respective bin over time; and

for each bin that is assigned at least one of the nearby objects, maintain the soft boundaries and adjust the original hard boundaries of the bin to exclude an area occupied by the at least one of the nearby objects thereby defining a modified drivable area within the bin, wherein the excluded area constitutes a cutout portion;

identify a first object along a coarse driveline of the initial adjusted drivable area of the AV, the first object being a nearby object that is classified as a dynamic object;

receive a predicted path of the first object;

determine, based on the predicted path of the first object, an adjusted drivable area by adjusting, based on the predicted path of the first object, the first initial adjusted drivable area to obtain the adjusted drivable area;

determine a trajectory of the AV through the adjusted drivable area by prioritizing confining the trajectory within the adjusted hard boundaries but permitting crossing the soft boundaries when the trajectory cannot be confined within the adjusted hard boundaries due to collisions with the first object, wherein crossing the soft boundaries is permitted if crossing the soft boundaries is safe; and

control the AV according to the trajectory.

10. The system of claim 9 , wherein determining, based on the predicted path of the first object, the adjusted drivable area comprises:

assigning the first object to one bin of the bins; and

determining the adjusted drivable area based on the one bin.

11. The system of claim 10 , wherein determining, based on the predicted path of the first object, the adjusted drivable area comprises:

determining respective object locations of the first object at discrete time intervals;

determining respective AV locations of the AV at the discrete time intervals; and

determining the adjusted drivable area based on the respective object locations and the respective AV locations.

12. The system of claim 11 ,

wherein the first object is determined to be an oncoming longitudinal constraint; and

wherein to control the AV according to the trajectory comprises to:

stop the AV until a second time subsequent to a first time that an object location of the first object meets an AV location of the AV.

13. The system of claim 11 ,

wherein the first object is determined to be a longitudinal constraint; and

wherein to control the AV according to the trajectory comprises to:

slow the AV to drive behind the first object.

14. The system of claim 10 , wherein the processor is further configured to:

classify the first object as a lateral constraint, a longitudinal constraint, an oncoming lateral constraint, or an oncoming longitudinal constraint.

15. The system of claim 9 ,

wherein to determine the trajectory of the AV through the adjusted drivable area comprises to:

determine that the first object is a static blockage; and wherein to control the AV according to the trajectory comprises to:

stop the AV.

16. The system of claim 9 , wherein the processor is further configured to:

detect a second object,

wherein to determine the trajectory of the AV through the adjusted drivable area comprises to:

determine that the first object is passable on a first side; and

determine that the second object is passable on the first side and a second side; and

wherein to control the AV according to the trajectory comprises to:

control the AV to pass the first object on the first side and to pass the second object on the first side.

17. A method for object avoidance by an autonomous vehicle (AV), comprising:

identifying nearby objects to the AV;

determining an adjusted drivable area by:

extracting a drivable area of the AV;

cutting out, to obtain an initial adjusted drivable area, cutout portions of the drivable area, the cutout portions corresponding to nearby objects that are classified as static objects, wherein cutting out the cutout portions comprises:

dividing the drivable area into bins, each having original hard boundaries and soft boundaries, wherein the original hard boundaries correspond to nearby objects and the soft boundaries correspond to lawful restraints, defined with respect to a longitudinal direction of travel of the AV;

assigning each of the nearby objects to a respective bin based on a respective location of the each of the nearby objects;

tracking the nearby objects within the respective bin over time; and

for each bin that is assigned at least one of the nearby objects and while maintaining the soft boundaries, adjusting the original hard boundaries of the bin to exclude an area occupied by the at least one of the nearby objects thereby defining a modified drivable area within the bin, wherein the excluded area constitutes a cutout portion;

identifying a first object along a coarse driveline of the initial adjusted drivable area of the AV, the first object being a nearby object that is classified as a dynamic object;

assigning the first object to one bin of the bins; and

determining the adjusted drivable area based on the one bin and the initial adjusted drivable area;

determining a trajectory of the AV through the adjusted drivable area by prioritizing confining the trajectory within the adjusted hard boundaries but permitting crossing the soft boundaries when the trajectory cannot be confined within the adjusted hard boundaries due to collisions with the first object, wherein crossing the soft boundaries is permitted if crossing the soft boundaries is safe; and

controlling the AV according to the trajectory.

18. The method of claim 17 , wherein determining the trajectory of the AV through the adjusted drivable area comprises:

determining whether a gap defined by boundaries of the one bin are such that the AV fits through the gap.

19. The method of claim 18 , further comprising:

in a case where the AV does not fit though the gap, setting a static blockage at the one bin, wherein controlling the AV according to the trajectory comprises:

stopping the AV before the one bin.

20. The method of claim 17 , further comprising:

detecting a second object along the coarse driveline;

classifying the second object as an oncoming longitudinal constraint; and

stopping the AV until a second time subsequent to a first time, wherein a first location of the AV and a second location of the second object meet at the first time.

Assignments (4)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 29, 2024
From: NISSAN NORTH AMERICA, INC.
To: NISSAN MOTOR CO., LTD.
Reel/Frame 069271/0395 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 25, 2021
From: RENAULT S.A.S.
To: NISSAN NORTH AMERICA, INC.
Reel/Frame 055713/0449 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 20, 2020
From: NISSAN NORTH AMERICA, INC.
To: NISSAN NORTH AMERICA, INC.; RENAULT S.A.S.
Reel/Frame 054108/0091 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 20, 2020
From: OSTAFEW, CHRISTOPHER
To: NISSAN NORTH AMERICA, INC.
Reel/Frame 054139/0955 →
Continuity (1)
Related Publication 20210237769A1 · Aug 5, 2021
Cited By (2)
US 12,233,856 US 12,330,633