Granted Patent B2

US 12,517,521 · App. 17/982,959 · Granted Jan 6, 2026

Inventors: Mathias Andreas Hintermann (Playa Vista, CA); Justin Huang (Los Angeles, CA); Russell Douglas Patton (Playa Vista, CA)

Assignee: Snap Inc.

G05D1/042G06V10/82G06V20/17G06V40/28G08G5/30B64U10/13B64U30/20B64U2201/10

Abstract

Claims (74)

1 . A method performed on an apparatus of an autonomous drone, the method comprising:

accessing a fly instruction from a person;

capturing, a first image, the first image depicting a particular face of the person;

sending the first image to an external computing device;

receiving, from the external computing device in response to sending the first image, weights for a neural network to identify the particular face of the person;

processing a second image using the neural network with the received weights, wherein a result of the processing indicates the person has the particular face;

in response to the person having the particular face, changing a state of a navigation system to unlocked;

lifting off the autonomous drone in response to the fly instruction;

receiving sensor data;

processing the sensor data to identify a gesture that indicates that the autonomous drone is to land; and

verifying the gesture was performed by the person by processing a third image using the neural network with the received weights, wherein a result of the processing indicates the person has the particular face.

2 . The method of claim 1 wherein the sensor data is first sensor data from a height sensor, the first sensor data indicating the autonomous drone is a first height above a ground, and second sensor data from the height sensor, the second sensor data indicating the autonomous drone is a second height above the ground, and wherein the method further comprises:

determining a difference between the first height and the second height is larger than a first threshold and the second height is not larger than a second threshold; and

landing the autonomous drone.

3 . The method of claim 1 wherein the sensor data comprises the third image, and wherein the method further comprises:

capturing the third image.

4 . A non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of an apparatus of an autonomous drone, cause the at least one processor to perform operations comprising:

accessing a fly instruction from a person;

capturing, a first image, the first image depicting a particular face of the person;

sending the first image to an external computing device;

receiving, from the external computing device in response to sending the first image, weights for a neural network to identify the particular face of the person;

processing a second image using the neural network with the received weights, wherein a result of the processing indicates the person has the particular face;

in response to the person having the particular face, changing a state of a navigation system to unlocked;

lifting off the autonomous drone in response to the fly instruction;

receiving sensor data;

processing the sensor data to identify a gesture that indicates that the autonomous drone is to land; and

verifying the gesture was performed by the person by processing a third image using the neural network with the received weights, wherein a result of the processing indicates the person has the particular face.

5 . An apparatus of an autonomous drone comprising:

at least one processor; and

a memory storing instructions that, when executed by the at least one processor, configure the processor to perform operations comprising:

accessing a fly instruction from a person;

capturing, a first image, the first image depicting a particular face of the person;

sending the first image to an external computing device;

receiving, from the external computing device in response to sending the first image, weights for a neural network to identify the particular face of the person;

processing a second image using the neural network with the received weights, wherein a result of the processing indicates the person has the particular face;

in response to the person having the particular face, changing a state of a navigation system to unlocked;

lifting off the autonomous drone in response to the fly instruction;

receiving sensor data;

processing the sensor data to identify a gesture that indicates that the autonomous drone is to land; and

verifying the gesture was performed by the person by processing a third image using the neural network with the received weights, wherein a result of the processing indicates the person has the particular face.

6 . The apparatus of claim 5 wherein the sensor data is first sensor data from a height sensor, the first sensor data indicating the autonomous drone is a first height above a ground, and second sensor data from the height sensor, the second sensor data indicating the autonomous drone is a second height above the ground, and wherein the operations further comprise:

determining a difference between the first height and the second height is larger than a first threshold and the second height is not larger than a second threshold; and

landing the autonomous drone.

7 . The apparatus of claim 6 wherein the operations further comprise:

before the receiving the sensor data, completing a flight plan, navigating the autonomous drone to the person, and hovering within a threshold distance from the person.

8 . The apparatus of claim 5 wherein the sensor data comprises the third image, and wherein the operations further comprises:

capturing the third image.

9 . The apparatus of claim 8 wherein the gesture is an open hand gesture that indicates the autonomous drone is to land on an open hand of the person.

10 . The apparatus of claim 9 wherein the operations further comprise:

flying the autonomous drone to land on the open hand of the person.

11 . The apparatus of claim 8 wherein the gesture indicates that the autonomous drone is to start flying to a landing site.

12 . The apparatus of claim 8 wherein the gesture is a hand of the person being placed under the autonomous drone, and wherein the operations further comprise:

landing the autonomous drone on the hand by flying the autonomous drone downward.

13 . The apparatus of claim 8 wherein the operations further comprise:

capturing a fourth image of a hand of the person before capturing the third image; and

using the fourth image of the hand to identify the hand of the person.

14 . The apparatus of claim 8 wherein the operations further comprise:

capturing a fourth image of a hand of the person a threshold duration after capturing the second image;

processing the fourth image to determine whether the third image comprises the gesture from the person that indicates that the autonomous drone is to land, wherein the processing comprises determining the hand of the person is not grasping an object unsuitable for the autonomous drone to land on; and

in response to the fourth image comprising the gesture from the person, landing the autonomous drone.

15 . The apparatus of claim 8 wherein the operations further comprise:

processing the third image using an additional neural network to identify the gesture.

16 . The apparatus of claim 8 wherein the operations further comprise:

processing a portion of the third image comprising the particular face to identify the gesture that indicates that the autonomous drone is to land, wherein the gesture is performed within a threshold distance of the autonomous drone; and

navigating to a landing of the autonomous drone.

17 . The apparatus of claim 8 wherein the gesture is a first gesture, and wherein the operations further comprise:

capturing a fourth image; and

processing the fourth image to identify a second gesture from the person that indicates that the autonomous drone is to continue on a flight plan.

18 . The apparatus of claim 8 wherein the operations further comprise:

navigating the autonomous drone to hover at a height.

19 . The apparatus of claim 18 wherein the third image is captured by an image capturing device mounted horizontally relative to an axis of propellers of the autonomous drone and a height sensor is mounted vertically relative to the axis of the propellers and is directed downward.

20 . The apparatus of claim 5 wherein the operations further comprise:

determining a land flight plan based on a position of the person relative to the autonomous drone; and

navigating the land flight plan based on identifying the person in subsequent images captured by an image capturing device.

Assignments (1)

Continuity (2)

References Cited (205)