Algorithms for avoiding automotive crashes at left and right turn intersections

Patent metadata

Field	Value
Publication number	US-9751506-B2
Application number	US-201514924585-A
Country	US
Kind code	B2
Filing date	Oct 27, 2015
Priority date	Oct 27, 2015
Publication date	Sep 5, 2017
Grant date	Sep 5, 2017

Abstract

A system and method for warning a driver of a host vehicle of a potential collision with other vehicles when turning at an intersection. The method includes determining if the host vehicle is likely to turn at the intersection, and if so, segmenting the intersection into a plurality of different regions, where each region has a different level of collision threat risk. The method obtains the speed, velocity and position of the host vehicle and any relevant remote vehicles in the intersection. The method determines a predicted path of the host vehicle and the predicted path of the remote vehicles. The method then determines whether the host vehicle and the remote vehicles will simultaneously occupy a collision zone in the intersection based on the predicted paths, and if so, issues a warning to the driver of the collision risk or apply vehicle controls to avoid or mitigate the collision risk.

First claim

What is claimed is:

1. A method for collision threat assessment when a host vehicle is turning left or right at an intersection, said method comprising:
determining if the host vehicle is likely to turn left or right at the intersection with a predetermined level of confidence;
segmenting the intersection and surrounding area into a plurality of different regions where each region has a different level of collision threat risk if the host vehicle is likely to turn left or right at the intersection;
obtaining speed, velocity and position data of the host vehicle using sensors on the host vehicle;
obtaining speed, velocity and position data of one or more remote vehicles and/or other objects in or near the intersection, including using information and data from one or more of a map database, a long range radar sensor on the vehicle, a short range radar sensor on the vehicle, cameras on the vehicle, a lidar sensor on the vehicle, V2X communications, Internet communications and satellite communications;
determining, by a controller including a processor, a predicted path of the host vehicle based on the speed, velocity and position data as the host vehicle travels through the intersection for each of the different segmented regions;
determining, by the controller, a predicted path of one or more of the remote vehicles based on the speed, velocity and position data as the one or more vehicles travel through the intersection;
determining, by the controller, whether the host vehicle and the one or more remote vehicles will simultaneously occupy a collision zone in the intersection based on the predicted path of the host vehicle and the predicted path of the one or more remote vehicles;
issuing, by the controller, a warning to a driver of the host vehicle if the host vehicle and the one or more remote vehicles will occupy the collision zone during the same time, including determining a time at current vehicle speed and predicted path of when the host vehicle will enter the collision zone, and issuing the warning if the time to the collision zone is below a first predetermined threshold; and
providing automatic braking of the host vehicle, by the controller, if the time to the collision zone is less than a second predetermined threshold that is shorter than the first predetermined threshold.

2. The method according to claim 1 wherein segmenting the intersection into different regions includes identifying an approach region before the intersection that the host vehicle will travel through, identifying a waiting region in the intersection that the host vehicle will travel through when making the turn, and identifying a no-waiting region in the intersection that the host vehicle will travel through when making the turn.

3. The method according to claim 1 wherein determining the predicted path of the host vehicle also includes considering geometry of the intersection that is used to refine the predicted path.

4. The method according to claim 1 wherein determining whether the host vehicle and the one or more remote vehicles will simultaneously occupy a collision zone in the intersection includes determining a duration that the host vehicle will occupy the collision zone based on the speed and the predicted path of the host vehicle, determining a duration that the one or more remote vehicles will occupy the collision zone based on the speed and predicted path of the one or more remote vehicles, and determining if the durations overlap.

5. The method according to claim 1 wherein determining whether the host vehicle and the one or more remote vehicles will simultaneously occupy a collision zone in the intersection includes determining whether the host vehicle and the one or more remote vehicles will simultaneously occupy the collision zone before the host vehicle enters the intersection.

6. The method according to claim 1 wherein determining whether the host vehicle and the one or more remote vehicles will simultaneously occupy a collision zone in the intersection includes determining whether the host vehicle and the one or more remote vehicles will simultaneously occupy the collision zone when the host vehicle is in the intersection when the one or more remote vehicles is a opposite direction traveling remote vehicle.

7. The method according to claim 1 further comprising determining that the host vehicle will not stop in the intersection when making the turn.

8. The method according to claim 1 wherein the first threshold is determined experimentally considering driver reaction time and system delays.

9. The method according to claim 1 wherein the second threshold for providing automatic braking is determined based on system delays.

10. The method according to claim 1 further comprising providing information of a potential collision to the host vehicle if the time to the collision zone is greater than the first predetermined threshold, but less than a third predetermined threshold.

11. The method according to claim 1 further comprising determining that the host vehicle is coming to or has come to a stop in the intersection in anticipation of making a turn after opposing traffic has cleared, wherein issuing a warning includes determining a likely path the host vehicle will make when making the turn, calculating a closeness coefficient between the likely path of the host vehicle and the predicted path of the host vehicle, and issuing the warning if the closeness coefficient is greater than a predetermined closeness threshold, a time that the one or more remote vehicles will enter the collision zone is less than a predetermined time threshold and brakes of the host vehicle have been released.

12. The method according to claim 11 wherein determining the closeness coefficient between the predicted path and the likely path includes using the equation:
C PL =C X *C Y ,

where:

C
X

=

N
*
Σ
⁢

⁢

X
Pi

*

X
Li

-

Σ
⁢

⁢

X
Pi

*
Σ
⁢

⁢

X
Li

N
*

Σ
⁡

(

X
Pi

Assignees

• Gm Global Tech Operations Llc

Inventors

• Mudalige Upali Priyantha
• Elvitigala Thanura Ranmal

Classifications

• B60W2554/80

  Spatial relation or speed relative to objects · CPC title

• G08G1/07

  Controlling traffic signals · CPC title

• G08G1/166

  for active traffic, e.g. moving vehicles, pedestrians, bikes · CPC title

• G08G1/09623

  Systems involving the acquisition of information from passive traffic signs by means mounted on the vehicle (G08G1/0967 takes precedence) · CPC title

• G01S13/931

  of land vehicles · CPC title