Information processing apparatus for controlling flight of an aerial vehicle with a generated learning model

Patent metadata

Field	Value
Publication number	US-12154009-B2
Application number	US-201917283166-A
Country	US
Kind code	B2
Filing date	Oct 28, 2019
Priority date	Dec 14, 2018
Publication date	Nov 26, 2024
Grant date	Nov 26, 2024

Abstract

A server device performs machine learning on the relationship between the content of piloting of an aerial vehicle and the behavior of the aerial vehicle in response to the content of the piloting, and generates a learning model for automatically piloting the aerial vehicle. However, the aerial vehicle is piloted in various environments and conditions, and there are environments and conditions that are unsuited for achieving highly accurate and stable automatic piloting. Therefore, the server device performs the machine learning only in an environment or a condition suited for realizing the automatic piloting.

First claim

The invention claimed is:

1. An information processing apparatus comprising:
a processor configured to:
learn a relationship between piloting of an aerial vehicle and a behavior of the aerial vehicle in response to the piloting;
determine whether a flight of the aerial vehicle is a flight that satisfies a condition determined as being not for the learning;
perform the learning to generate a learning model with less weight given to a relationship between the piloting and the behavior of the aerial vehicle in a period in which it is determined that the flight satisfies the condition, as compared with a period in which it is determined that the flight does not satisfy the condition;
wherein a flight that satisfies the condition is a low-visibility flight, in which the aerial vehicle flies in an environment in which the visibility of the aerial vehicle from an operator is lower than a predetermined level; and
wherein the processor is further configured to:
determine whether a flight of an aerial vehicle is the low-visibility flight;
perform the learning to generate the learning model with less weight given to a relationship between the piloting and the behavior of the aerial vehicle in a period in which a flight is determined as being the low-visibility flight, as compared with a period in which a flight is determined as being not the low-visibility flight; and
control the flight of the aerial vehicle with the generated learning model.

2. An information processing apparatus comprising:
a processor configured to:
learn a relationship between piloting of an aerial vehicle and a behavior of the aerial vehicle in response to the piloting;
determine whether a flight of the aerial vehicle is a flight that satisfies a condition determined as being not for the learning;
perform the learning to generate a learning model with less weight given to a relationship between the piloting and the behavior of the aerial vehicle in a period in which it is determined that the flight satisfies the condition, as compared with a period in which it is determined that the flight does not satisfy the condition;
wherein a flight that satisfies the condition is an anomaly-affected flight, in which the aerial vehicle flies with an anomaly occurring in the aerial vehicle; and
wherein the processor is further configured to:
determine whether the flight of the aerial vehicle is the anomaly-affected flight;
perform the learning to generate the learning model with less weight given to a relationship between the piloting and the behavior of the aerial vehicle in a period in which the flight is determined as being the anomaly-affected flight, as compared with a period in which the flight is determined as being not the anomaly-affected flight; and
control the flight of the aerial vehicle with the generated learning model.

3. An information processing apparatus comprising:
a processor configured to:
learn a relationship between piloting of an aerial vehicle and a behavior of the aerial vehicle in response to the piloting;
determine whether a flight of the aerial vehicle is a flight that satisfies a condition determined as being not for the learning;
perform the learning to generate a learning model with less weight given to a relationship between the piloting and the behavior of the aerial vehicle in a period in which it is determined that the flight satisfies the condition, as compared with a period in which it is determined that the flight does not satisfy the condition;
wherein a flight that satisfies the condition is a signal-missing flight, in which the aerial vehicle flies with a missing radio signal for controlling the aerial vehicle; and
wherein the processor is further configured to:
determine whether the flight of the aerial vehicle is the signal-missing flight; and
perform the learning to generate the learning model with less weight given to a relationship between the piloting and the behavior of the aerial vehicle in the period in which the flight is determined as being the signal-missing flight, as compared with a period in which the flight is determined as being not the signal-missing flight; and
control the flight of the aerial vehicle with the generated learning model.

4. The information processing apparatus according to claim 1 ,
wherein the processor is further configured to perform the learning to generate the learning model with less weight given to a relationship between the piloting and the behavior of the aerial vehicle in a period in which it is determined that the flight satisfies the condition and further in a specific period in which the behavior of the aerial vehicle is excessively large or small in response to the piloting of the aerial vehicle, as compared with a period that is not the specific period.

5. An information processing apparatus comprising:
a processor configured to:
learn a relationship between piloting of an aerial vehicle and a behavior of the aerial vehicle in response to the piloting;
determine whether a flight of the aerial vehicle is a flight that satisfies a condition determined as being not for the learning;
perform the learning to generate a learning model with less weight given to a relationship between the piloting and the behavior of the aerial vehicle in a period in which it is determined that the flight satisfies the condition, as compared with a period in which it is determined that the flight does not satisfy the condition; and
wherein the processor is further configured to:
perform the learning to generate the learning model with less weight given to a relationship between the piloting and the behavior of the aerial vehicle in a period in which it is determined that the flight satisfies the condition and further in a specific period in which the behavior of the aerial vehicle differs from an expected behavior by a threshold value or more, as compared with a period that is not the specific period;
wherein the threshold value varies according to the piloting skill level of an operator; and
control the flight of the aerial vehicle with the generated learning model.

6. The information processing apparatus according to claim 1 ,
wherein the processor is further configured to perform the learning to generate the learning model with less weight given to a relationship between the piloting and the behavior of the aerial vehicle in a period in which it is determined that the flight satisfies the condition and further in a specific period in which an external environment at the time of the flight of the aerial vehicle is a predetermined environment, as compared with a period that is not the specific period.

7. The information processing apparatus according to claim 6 , wherein the predetermined environment is an environment where a gust blows onto the aerial vehicle and the aerial vehicle takes a sudden unintended behavior.

8. The information processing apparatus according to claim 6 , wherein the predetermined environment is an environment where at least one of a strong wind is constantly blowing, rain is falling, or snow is falling.

Assignees

• Ntt Docomo Inc

Inventors

• Takami Tadao
• Ishii Koji
• Woo Jooin
• Kawakami Hiroshi
• Niihata Kaori
• Segawa Yuichiro
• Kitamura Yasuhiro

Classifications

• G06V20/597

  Recognising the driver's state or behaviour, e.g. attention or drowsiness · CPC title

• B64F5/00

  Designing, manufacturing, assembling, cleaning, maintaining or repairing aircraft, not otherwise provided for; Handling, transporting, testing or inspecting aircraft components, not otherwise provided for · CPC title

• G05D1/00

  Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots (drive control systems specially adapted for autonomous road vehicles B60W60/00) · CPC title

• B64C13/18

  using automatic pilot · CPC title

• G05D1/10

  Simultaneous control of position or course in three dimensions (G05D1/12 takes precedence) · CPC title