Cost map fusion for lane selection

Patent metadata

Field	Value
Publication number	US-12427987-B2
Application number	US-202318235792-A
Country	US
Kind code	B2
Filing date	Aug 18, 2023
Priority date	Aug 18, 2023
Publication date	Sep 30, 2025
Grant date	Sep 30, 2025

Abstract

Systems and methods of generating lane selection cost maps to control autonomous vehicles are disclosed. An autonomous vehicle system can receive a plurality of cost values from a plurality of components of an autonomous vehicle; generate a lane selection cost map based on the plurality of cost values; determining that the autonomous vehicle should change lanes based on the lane selection cost map; and transmit a command that causes the autonomous vehicle to change lanes responsive to determining that the autonomous vehicle should change lanes.

First claim

What is claimed is:

1. A system, comprising:
at least one processor coupled to non-transitory memory, the at least one processor configured to:
receive a plurality of cost values from a plurality of components of an autonomous vehicle, each cost value representing a cost for traveling in a particular region of a particular lane;
determine an overall cost of a particular region based on a weighted sum of at least two cost values for the particular region;
generate a lane selection cost map including the particular region determined overall cost and at least one other region including a determined overall cost;
responsive to a trajectory of the autonomous vehicle in an adjacent lane having a lower cost value than a trajectory of the autonomous vehicle in a current lane, determine that the autonomous vehicle should change lanes based on the lane selection cost map; and
transmit a command that causes the autonomous vehicle to change lanes responsive to determining that the autonomous vehicle should change lanes.

2. The system of claim 1 , wherein the at least one processor is further configured to determine that the autonomous vehicle should change lanes based on execution of a pathfinding algorithm using the lane selection cost map.

3. The system of claim 1 , wherein the at least one processor is further configured to generate the lane selection cost map in response to a detected object in an environment surrounding the autonomous vehicle.

4. The system of claim 1 , wherein at least one of the plurality of cost values is determined based on a location of the autonomous vehicle.

5. The system of claim 1 , wherein at least one of the plurality of cost values is determined based on map data comprising one or more static map features.

6. The system of claim 1 , wherein at least one of the plurality of cost values is determined based on a detected obstacle.

7. A method, comprising:
receiving, by one or more processors coupled to non-transitory memory, a plurality of cost values from a plurality of components of an autonomous vehicle, each cost value representing a cost for traveling in a particular region of a particular lane;
determine an overall cost of a particular region based on a weighted sum of at least two cost values for the particular region;
generating, by the one or more processors, a lane selection cost map including the particular region determined overall cost and at least one other region including a determined overall cost;
responsive to a trajectory of the autonomous vehicle in an adjacent lane having a lower cost value than a trajectory of the autonomous vehicle in a current lane, determining, by the one or more processors, that the autonomous vehicle should change lanes based on the lane selection cost map; and
transmitting, by the one or more processors, a command that causes the autonomous vehicle to change lanes responsive to determining that the autonomous vehicle should change lanes.

8. The method of claim 7 , further comprising determining, by the one or more processors, that the autonomous vehicle should change lanes based on execution of a pathfinding algorithm using the lane selection cost map.

9. The method of claim 7 , further comprising generating, by the one or more processors, the lane selection cost map in response to a detected object in an environment surrounding the autonomous vehicle.

10. The method of claim 7 , wherein at least one of the plurality of cost values is determined based on a location of the autonomous vehicle.

11. The method of claim 7 , wherein at least one of the plurality of cost values is determined based on map data comprising one or more static map features.

12. The method of claim 7 , wherein at least one of the plurality of cost values is determined based on a detected obstacle.

13. An autonomous vehicle comprising one or more processors configured to:
identify a plurality of cost values generated by a plurality of components of the autonomous vehicle, each cost value representing a cost for traveling in a particular region of a particular lane;
determine an overall cost of a particular region based on a weighted sum of at least two cost values for the particular region;
generate a lane selection cost map including the particular region determined overall cost and at least one other region including a determined overall cost;
responsive to a trajectory of the autonomous vehicle in an adjacent lane having a lower cost value than a trajectory of the autonomous vehicle in a current lane, determine that the autonomous vehicle should change lanes based on the lane selection cost map; and
cause the autonomous vehicle to change lanes responsive to determining that the autonomous vehicle should change lanes.

Assignees

• Torc Robotics Inc

Inventors

• Shoemaker Adam
• Blankenhorn John
• Madsen Garrett
• Petrin Joshua
• Tulsyan Ajay
• Brown Paul
• Pereira Savio
• Davis William
• Fernández Daniel
• Hao Yexuan

Classifications

• B60W2300/145

  Semi-trailers · CPC title

• B60W2556/35

  Data fusion · CPC title

• B60W2552/53

  Road markings, e.g. lane marker or crosswalk · CPC title

• B60W2556/40

  High definition maps · CPC title

• B60W60/001 Primary

  Planning or execution of driving tasks · CPC title