Parking-lot-navigation system and method

Patent metadata

Field	Value
Publication number	US-10481609-B2
Application number	US-201615373845-A
Country	US
Kind code	B2
Filing date	Dec 9, 2016
Priority date	Dec 9, 2016
Publication date	Nov 19, 2019
Grant date	Nov 19, 2019

Abstract

A system and method for assisted or autonomous parking of a vehicle is disclosed. The method may begin when the vehicle approaches a feeder lane within a parking lot. At that point, a computer system may decide whether the vehicle should enter the feeder lane. The computer system may use at least one of machine learning, computer vision, and range measurements to determining whether a condition precedent for entering the feeder lane exists. The condition precedent may include an in-bound arrow on the feeder lane or parking lines and/or a parked vehicle adjacent the feeder lane defining a departure angle less than or equal to ninety degrees. If the condition precedent exists, the vehicle may enter the feeder lane. If the condition precedent does not exist, the vehicle may move on to another feeder lane.

First claim

The invention claimed is:

1. A method for determining whether to advance into a feeder lane within a parking lot, the method comprising:
approaching, by an autonomous vehicle, the feeder lane within the parking lot;
identifying, by the autonomous vehicle, using at least one of machine learning and computer vision, at least one of
an in-bound arrow on the feeder lane, and
parking lines or a parked vehicle adjacent the feeder lane defining a departure angle less than or equal to ninety degrees, wherein the angle of departure is a measure of a change of direction necessary for the autonomous vehicle to appropriately enter one or more parking spaces directly adjacent the feeder lane if the autonomous vehicle were already traveling down the feeder lane; and

advancing, by the autonomous vehicle in response to the identifying, into the feeder lane.

2. The method of claim 1 , further comprising using one or more sensors on-board the autonomous vehicle to collect data characterizing at least a portion of the feeder lane.

3. The method of claim 2 , wherein the identifying comprises applying, by a computer system carried on-board the autonomous vehicle, the at least one of machine learning and computer vision to the data.

4. The method of claim 3 , wherein the one or more sensors comprise a camera.

5. The method of claim 4 , wherein the data comprises at least one image frame output by the camera.

6. The method of claim 5 , wherein the identifying further comprises using, by the computer system, computer vision to select at least one region of interest comprising less than all of the at least one image frame.

7. The method of claim 6 , wherein the identifying further comprises generating, by the computer system, at least one cropped subset of the at least one image frame by cropping the at least one region of interest from the at least one image frame.

8. The method of claim 7 , wherein the identifying further comprises using, by the computer system, an artificial neural network to classify the at least one cropped subset.

9. The method of claim 8 , wherein:
the identifying further comprises using, by the computer system, the artificial neural network to determine an affinity score between the at least one cropped subset and each class of a plurality of classes; and
the plurality of classes includes
a no arrow class,
an in-bound arrow only class,
an out-bound arrow only class, and
an in-bound and out-bound arrow class.

10. The method of claim 8 , wherein:
the identifying further comprises using, by the computer system, the artificial neural network to determine an affinity score between the at least one cropped subset and each class of a plurality of classes; and
the plurality of classes includes
a no parking lines class,
a parking lines with departure angle of less than or equal to ninety degrees class, and
a parking lines with departure angle of greater than ninety degrees class.

11. The method of claim 8 , wherein:
the identifying further comprises using, by the computer system, the artificial neural network to determine an affinity score between the at least one cropped subset and each class of a plurality of classes; and
the plurality of classes includes
a no parked vehicle class,
a parked vehicle with departure angle of less than or equal to ninety degrees class, and
a parked vehicle with departure angle of greater than ninety degrees class.

12. A method for determining whether to advance into a first feeder lane within a parking lot, the method comprising:
approaching, by an autonomous vehicle, the first feeder lane within the parking lot;
collecting, by one or more sensors on-board the autonomous vehicle, first data characterizing at least a portion of the first feeder lane;
confirming, by a computer system carried on-board the autonomous vehicle, an existence of a condition precedent for advancing into the first feeder lane;
the confirming comprising detecting the condition precedent by at least one of applying a machine-learning algorithm to the first data, applying a computer-vision technique to the first data, and extracting first range measurements from the first data;
the confirming wherein the condition precedent for advancing into the first feeder lane comprises at least one of
an in-bound arrow on the first feeder lane,
parking lines adjacent the first feeder lane defining a departure angle less than or equal to ninety degrees, wherein the angle of departure defined by the parking lines adjacent the first feeder lane is a measure of a change of direction necessary for the autonomous vehicle to appropriately enter a parking space defined by the parking lines adjacent the first feeder lane if the autonomous vehicle were already traveling down the first feeder lane, and
at least one parked vehicle adjacent the first feeder lane defining a departure angle less than or equal to ninety degrees, wherein the angle of departure defined by the at least one parked vehicle adjacent the first feeder lane is a measure of a change of direction necessary for the autonomous vehicle to turn and park alongside the at least one parked vehicle adjacent the first feeder lane if the autonomous vehicle were already traveling down the first feeder lane; and

advancing, by the autonomous vehicle, into the first feeder lane in response to the confirming.

13. The method of claim 12 , further comprising approaching, by the autonomous vehicle, a second feeder lane within the parking lot.

14. The method of claim 13 , further comprising collecting, by the one or more sensors, second data characterizing at least a portion of the second feeder lane.

15. The method of claim 14 , further comprising confirming, by the computer system carried on-board the autonomous vehicle, an absence of a condition precedent for advancing into the second feeder lane.

16. The method of claim 15 , wherein the confirming the absence comprises at least one of applying the machine-learning algorithm to the second data, applying the computer-vision technique to the second data, and extracting second range measurements from the second data.

17. The method of claim 16 , wherein the condition precedent for advancing into the second feeder lane comprises at least one of:
an in-bound arrow on the second feeder lane;
parking lines adjacent the second feeder lane defining a departure angle less than or equal to ninety degrees, wherein the angle of departure defined by the parking lines adjacent the second feeder lane is a measure of a change of direction necessary for the autonomous vehicle to appropriately enter a parking space defined by the parking lines adjacent the second feeder lane if the autonomous vehicle were already traveling down the second feeder lane; and
at least one parked vehicle adjacent the second feeder lane defining a departure angle less than or equal to ninety degrees, wherein the angle of departure defined by the at least one parked vehicle adjacent the second feeder lane is a measure of a change of direction necessary for the autonomous vehicle to turn and park alongside the at least one parked vehicle adjacent the second feeder lane if the autonomous vehicle were already traveling down the second feeder lane.

18. The method of claim 17 , further comprising avoiding, by the autonomous vehicle, the second feeder lane in response to the confirming the absence.

19. The method of claim 12 , wherein:
the one or more sensors comprise a camera;
the data comprises at least one image frame output by the camera;
the identifying f

Assignees

• Ford Global Tech Llc

Inventors

• Moosaei Maryam
• Jain Jinesh J
• Banvait Harpreetsingh

Classifications

• B60W2710/20

  Steering systems · CPC title

• G08G1/168

  Driving aids for parking, e.g. acoustic or visual feedback on parking space · CPC title

• B60W2420/54

  Audio sensitive means, e.g. ultrasound · CPC title

• G08G1/142 Primary

  external to the vehicles · CPC title

• B60W30/06

  Automatic manoeuvring for parking · CPC title