Merge behavior systems and methods for mainline vehicles

Patent metadata

Field	Value
Publication number	US-11117584-B2
Application number	US-201815965411-A
Country	US
Kind code	B2
Filing date	Apr 27, 2018
Priority date	Apr 27, 2018
Publication date	Sep 14, 2021
Grant date	Sep 14, 2021

Abstract

A merge behavior system assists a host vehicle positioned in a mainline lane that is adjacent a merge lane. The merge behavior system includes an identification module that identifies at least one proximate vehicle in the merge lane. The merge behavior system also includes a prediction module that selects one or more models based on the at least one proximate vehicle, calculates one or more merge factors corresponding to the one or more models, and predicts a merge location based on the one or more merge factors. The merge behavior system further includes a control module configured to adjust a kinematic parameter of the host vehicle to create a gap at the predicted merge location.

First claim

The invention claimed is:

1. A merge behavior system for assisting a host vehicle positioned in a mainline lane that is adjacent a merge lane, the merge behavior system comprising:
an identification module configured to identify at least one proximate vehicle in the merge lane, a preceding vehicle positioned in the mainline lane ahead of the host vehicle, and a following vehicle positioned in the mainline lane behind the host vehicle;
a prediction module configured to:
select a behavior model generated using machine learning based on naturalistic datasets that approximate human behavioral outcomes associated with the at least one proximate vehicle, wherein the behavior model includes a preceding vehicle model and a following vehicle model,
calculate one or more merge factors corresponding to the behavior model, wherein the one or more merge factors include a first merge factor based on the preceding vehicle model and a second merge factor based on the following vehicle model,
compare the one or more merge factors to at least one threshold value; and
predict a merge location based on the one or more merge factors based on the comparison; and

a control module configured to adjust a kinematic parameter of the host vehicle to create a gap at the predicted merge location.

2. The merge behavior system of claim 1 , wherein the behavior model is selected based on a relative location of the at least one proximate vehicle to the host vehicle.

3. The merge behavior system of claim 1 , further comprising a merge module configured to determine whether the gap at the predicted merge location is a sufficient size for the at least one proximate vehicle, and in response to determining that the gap does not have a sufficient size, initiating the prediction module to predict an updated merge location.

4. The merge behavior system of claim 1 , wherein the preceding vehicle model is based on a preceding gap length between the host vehicle and the preceding vehicle, a host vehicle velocity, and a relative preceding velocity of the preceding vehicle relative to the host vehicle.

5. The merge behavior system of claim 4 , wherein the preceding vehicle model incorporates a following vehicle variable that has a binary value based on the following vehicle in the mainline lane.

6. The merge behavior system of claim 1 , wherein the following vehicle model is based on a following gap length between the host vehicle and the following vehicle, a host vehicle velocity, and a relative following velocity of the following vehicle relative to the host vehicle.

7. The merge behavior system of claim 1 , wherein the kinematic parameter is a velocity of the host vehicle.

8. The merge behavior system of claim 1 , wherein the control module adjusts the kinematic parameter with an advanced driver-assistance system.

9. A merge behavior system for assisting a host vehicle positioned in a mainline lane that is adjacent a merge lane, the merge behavior system comprising:
an identification module configured to identify at least one proximate vehicle in the merge lane, a preceding vehicle positioned in the mainline lane ahead of the host vehicle, and a following vehicle positioned in the mainline lane behind the host vehicle;
a prediction module configured to:
select a behavior model generated using machine learning that incorporate naturalistic datasets that approximate human behavioral outcomes associated with a relative location of the at least one proximate vehicle to the host vehicle, wherein the behavior model includes a preceding vehicle model and a following vehicle model,
calculate one or more merge factors corresponding to the behavior model, wherein the one or more merge factors include a first merge factor based on the preceding vehicle model and a second merge factor based on the following vehicle model,
compare the one or more merge factors to at least one threshold value; and
predict a merge location based on the comparison, wherein the merge location is ahead or behind the host vehicle; and
a control module configured to adjust a kinematic parameter of the host vehicle to create a gap at the predicted merge location.

10. The merge behavior system of claim 9 , wherein the kinematic parameter is a velocity of the host vehicle.

11. The merge behavior system of claim 9 , wherein the control module adjusts the kinematic parameter with an advanced driver-assistance system.

12. The merge behavior system of claim 9 , wherein in response to determining that the gap does not have a sufficient size, initiating a merge module configured to determine whether a gap at the predicted merge location is a sufficient size for the host vehicle, and in response to determining that the gap has a sufficient size, initiating a merge maneuver.

13. A merge behavior method for assisting a host vehicle positioned in a mainline lane that is adjacent a merge lane, the merge behavior method comprising:
identifying at least one proximate vehicle in the merge lane, a preceding vehicle positioned in the mainline lane ahead of the host vehicle, and a following vehicle positioned in the mainline lane behind the host vehicle;
selecting a behavior model generated using machine learning based on naturalistic datasets that approximate human behavioral outcomes associated with the at least one proximate vehicle, wherein the behavior model includes a preceding vehicle model and a following vehicle model,
calculating one or more merge factors corresponding to the behavior model, wherein the one or more merge factors include a first merge factor based on the preceding vehicle model and a second merge factor based on the following vehicle model,
predicting a merge location based on the one or more merge factors based on the comparison; and
adjusting a kinematic parameter of the host vehicle to create a gap at the predicted merge location.

14. The merge behavior method of claim 13 , further comprising:
determining whether a gap at the predicted merge location is a sufficient size for the host vehicle, and
predicting an updated merge location.

15. The merge behavior method of claim 13 , wherein the behavior model is selected based on a relative location of the at least one proximate vehicle to the host vehicle.

Assignees

• Honda Motor Co Ltd

Inventors

• D'Sa Jovin
• Rajab Samer
• Saigusa Shigenobu
• Bai Xue

Classifications

• B60W2554/802

  Longitudinal distance · CPC title

• B60W2554/4046

  Behavior, e.g. aggressive or erratic · CPC title

• B60W2554/804

  Relative longitudinal speed · CPC title

• B60W2554/4049

  Relationship among other objects, e.g. converging dynamic objects · CPC title

• B60W30/18163 Primary

  Lane change; Overtaking manoeuvres · CPC title