Controlling an autonomous vehicle based upon computed lane boundaries

Patent metadata

Field	Value
Publication number	US-12036991-B2
Application number	US-202017039943-A
Country	US
Kind code	B2
Filing date	Sep 30, 2020
Priority date	Aug 2, 2018
Publication date	Jul 16, 2024
Grant date	Jul 16, 2024

Abstract

Described herein are technologies relating to controlling an autonomous vehicle based upon a computed lane boundary. The computed lane boundary defines a travel path for the autonomous vehicle when the autonomous vehicle turns across a lane of traffic at an intersection of two or more roadways. The autonomous vehicle navigates turn based upon the computed lane boundary, which allows the autonomous vehicle to enter the intersection without crossing into a lane of oncoming traffic.

First claim

What is claimed is:

1. An autonomous vehicle (AV) comprising:
a processor; and
computer-readable storage storing:
a map employed by the AV to navigate roadways, where the map includes a previously computed lane boundary; and
instructions that, when executed by the processor, cause the processor to perform acts comprising:
when the AV is stopped in a first lane of a first roadway at a stop line in the first lane of the first roadway, detecting that a traffic light has changed from red to green, wherein the traffic light is placed at an intersection of the first roadway and a second roadway, first traffic flows in a first direction in the first lane of the first roadway, the AV is to turn at the intersection across a second lane of the first roadway and into the second roadway, and further wherein second traffic flows in a second direction in the second lane of the first roadway, the second direction opposing the first direction;
subsequent to detecting that the traffic light has changed state from red to green, detecting that a vehicle is travelling towards the AV in the second direction in the second lane of the first roadway;
while continuing to detect that the vehicle is travelling towards the AV, moving in the first direction and entering the intersection while adhering to a travel boundary that crosses the second lane of the first roadway, wherein the travel boundary is represented by the previously computed lane boundary in the map employed by the AV to navigate roadways, and further wherein the travel boundary linearly extends from a painted lane boundary that defines a boundary between the first lane of the first roadway and the second lane of the first roadway and further linearly extends into the intersection such that the AV enters the intersection without crossing into the second lane of the first roadway;
detecting, based upon sensors of the AV, that the vehicle has passed through the intersection and the AV is able to turn through the intersection across the second lane of the first roadway and into the second roadway without impeding the second traffic that flows in the second direction; and
upon detecting that the AV is able to turn through the intersection across the second lane of the first roadway and into the second roadway without impeding the second traffic that flows in the second direction, controlling at least one of a propulsion system or a steering system to cause the AV to turn through the intersection across the second lane of the first roadway and into the second roadway.

2. The AV of claim 1 , wherein the intersection is free of an intersection painted lane boundary between the painted lane boundary and another painted lane boundary in the second roadway.

3. The AV of claim 1 , the acts further comprising: prior to reaching the stop line and while travelling in the first direction in the first roadway, controlling at least one of the propulsion system, the steering system, or a braking system to adhere to the painted lane boundary.

4. The AV of claim 1 , wherein detecting that the AV is able to turn through the intersection across the second lane of the first roadway and into the second roadway without impeding the second traffic that flows in the second direction comprises:
detecting that the traffic light has transitioned from green to red.

5. The AV of claim 1 , wherein the previously computed lane boundary is computed based upon at least one of a width of the intersection, a length of the intersection, or a turn radius of the AV.

6. The AV of claim 1 , wherein the sensors comprise a lidar sensor and a camera.

7. The AV of claim 1 , where the map includes several previously computed lane boundaries for several intersections.

8. A method performed by an autonomous vehicle (AV), the method comprising:
autonomously navigating roadways in a region, where autonomously navigating the roadways comprises:
accessing a computer-implemented map of the region, where the computer- implemented map includes a previously computed lane boundary;
stopping in a first lane of a first roadway at a stop line in the first lane of the first roadway;
when the AV is stopped in the first lane of the first roadway at the stop line in the first lane of the first roadway, detecting that a traffic light has changed from red to green, wherein the stop line and the traffic light are for an intersection between the first roadway and a second roadway that intersects the first roadway, first traffic flows in a first direction in the first lane of the first roadway, the AV is to turn at the intersection across a second lane of the first roadway and into the second roadway, and further wherein second traffic flows in a second direction in the second lane of the first roadway, the second direction being opposite the first direction;
subsequent to detecting that the traffic light has changed from red to green, detecting that a vehicle is travelling towards the AV in the second direction in the second lane of the first roadway;
while continuing to detect that the vehicle is travelling towards the AV, moving in the first direction and entering the intersection while adhering to a travel boundary that crosses the second lane of the first roadway, wherein the travel boundary is represented by the previously computed lane boundary in the computer-implemented map, and further wherein the travel boundary linearly extends from a painted lane boundary that defines a boundary between the first lane of the first roadway and the second lane of the first roadway such that the AV, while adhering to the travel boundary, enters the intersection without crossing into the second lane of the first roadway;
detecting, based upon sensors of the AV, that the vehicle has passed through the intersection and the AV is able to turn through the intersection across the second lane of the first roadway and into the second roadway without impeding the second traffic that flows in the second direction; and
upon detecting that the AV is able to turn through the intersection across the second lane of the first roadway and into the second roadway without impeding the second traffic that flows in the second direction, turning through the intersection across the second lane of the first roadway and into the second roadway.

9. The method of claim 8 , wherein the intersection is free of an intersection painted lane boundary between the painted lane boundary and another painted lane boundary in the second roadway.

10. The method of claim 8 , further comprising:
prior to reaching the stop line and while travelling in the first direction in the first roadway, controlling at least one of the propulsion system, the steering system, or a braking system to adhere to the painted lane boundary.

11. The method of claim 8 , wherein detecting that the AV is able to turn through the intersection across the second lane of the first roadway and into the second roadway without impeding the second traffic that flows in the second direction comprises:
detecting that the traffic light has transitioned from green to red.

12. The method of claim 8 , wherein the previously computed lane boundary is computed based upon at least one of a width of the intersection, a length of the intersection, or a turn radius of the AV.

13. The method of claim 8 , wherein the sensors comprise a lidar sensor and a camera.

14. The method of claim 8 , wherein the computer-implemented map is stored in computer-readable memory of the AV and includes several computed lane boundaries for intersecting roadways.

15. An autonomous vehicle (AV) comprising a computer-readable storage medium, where

Assignees

• Gm Global Tech Operations Llc
• Gm Cruise Holdings Llc

Inventors

• Allan Jeremy
• Chen Daniel

Classifications

• G05D1/646

  Following a predefined trajectory, e.g. a line marked on the floor or a flight path · CPC title

• G05D1/227

  Handing over between remote control and on-board control; Handing over between remote control arrangements · CPC title

• B60W2420/408

  Radar; Laser, e.g. lidar · CPC title

• B60W2420/403

  Image sensing, e.g. optical camera · CPC title

• B60W2552/00

  Input parameters relating to infrastructure · CPC title