DIGITAL ASSISTANT AND ASSOCIATED METHODS FOR A TRANSPORTATION VEHICLE

• US 20200135183A1
• Filed: 12/27/2019
• Published: 04/30/2020
• Est. Priority Date: 08/19/2016
• Status: Active Grant

• Current Assignee: Panasonic Avionics Corporation (Panasonic Holdings Corporation)
• Original Assignee: Panasonic Avionics Corporation (Panasonic Holdings Corporation)
• Inventors: Hilal, Rawad, Khabrani, Gurmukh, Perng, Chin
• Granted Patent: US 11,048,869 B2
• Time in Patent Office: Days
• CPC Class Codes: B60R 11/0235 of flat type, e.g. LCD
• CPC Class Codes: B60R 11/0264 for control means
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30
• CPC Class Codes: G06N 3/04 Architecture, e.g. interconnection topology
• CPC Class Codes: G06N 3/084 Backpropagation, e.g. using gradient descent
• CPC Class Codes: G06N 5/01 Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/16 using artificial neural networks
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command

Abstract

Methods and systems for a transportation vehicle are provided. One method includes receiving a user input for a valid communication session by a processor executable, digital assistant at a device on a transportation vehicle; tagging by the digital assistant, the user input words with a grammatical connotation; generating an action context, a filter context and a response context by a neural network, based on the tagged user input; storing by the digital assistant, a key-value pair for a parameter of the filter context at a short term memory, based on an output from the neural network; updating by the digital assistant, the key-value pair at the short term memory after receiving a reply to a follow-up request and another output from the trained neural network; and providing a response to the reply by the digital assistant.

Claims

1. A method comprising:

training a neural network for an action category on an aircraft associated with an intended user request by defining at least a filter parameter and a response context based on the filter parameter;

generating by the neural network an action context based on the action category, a filter context and a response context for a tagged user input;

using a short term memory configured as a least recently used memory for storing a key-value pair associated with the filter parameter of the filter context;

generating an output by the neural network in response to a follow-up request from the user;

updating the key-value pair at the short term memory based on the output from the neural network; and

providing a response to the follow-up request.

2. The method of claim 1, wherein the action context involves using a seat function controller that controls functions associated with an aircraft seat.

3. The method of claim 1, wherein the action context is for executing a shopping request using the seat device on the aircraft.

4. The method of claim 1, wherein the action context is for providing flight data to the user.

5. The method of claim 1, wherein the action context is for providing assistance by a crew member on the aircraft.

6. The method of claim 1, wherein the action context is to set a reminder for the user on the aircraft.

7. The method of claim 1, wherein a natural language processor generates the tagged user input.

8. A method comprising:

training a neural network for an action category on an aircraft associated with an intended user request by defining at least a filter parameter and a response context based on the filter parameter;

generating by the neural network an action context that maps to an action application programming interface (API) based on the action category, a filter context and a response context for a tagged user input, wherein the tagged user input is a user input transformed by a tag indicating a grammatical connotation;

wherein the action context identifies a type of action associated with the action category, the filter context defines one or more filter parameters that define a query parameter for the action API, and the response context provides information to extract from the action context to execute an action;

using a short term memory configured as a least recently used memory for storing a key-value pair associated with one or more of the filter parameters of the filter context and any missing parameter for the action API for a most current action context of a user session;

generating an output by the neural network in response to a follow-up request for the user session;

updating the key-value pair at the short term memory based on the output from the neural network; and

providing a response to the follow-up request.

9. The method of claim 8, wherein the action context involves using a seat function controller that controls functions associated with an aircraft seat.

10. The method of claim 8, wherein the action context is for executing a shopping request using the seat device on the aircraft.

11. The method of claim 8, wherein the action context is for providing flight data to the user.

12. The method of claim 8, wherein the action context is for providing assistance by a crew member on the aircraft.

13. The method of claim 8, wherein the action context is to set a reminder for the user on the aircraft.

14. The method of claim 8, wherein the missing parameter is retrieved from a memory device different from the short term memory.

15. A non-transitory machine-readable storage medium having stored thereon instructions for performing a method comprising machine executable code, which when executed by at least one machine, causes the machine to:

train a neural network for an action category on an aircraft associated with an intended user request by defining at least a filter parameter and a response context based on the filter parameter;

generate by the neural network an action context that maps to an action application programming interface (API) based on the action category, a filter context and a response context for a tagged user input, wherein the tagged user input is a user input transformed by a tag indicating a grammatical connotation;

wherein the action context identifies a type of action associated with the action category, the filter context defines one or more filter parameters that define a query parameter for the action API, and the response context provides information to extract from the action context to execute an action;

use a short term memory configured as a least recently used memory for storing a key-value pair associated with one or more of the filter parameter of the filter context and any missing parameter for the action API for a most current action context of a user session;

generate an output by the neural network in response to a follow-up request for the user session;

update the key-value pair at the short term memory based on the output from the neural network; and

provide a response to the follow-up request.

16. The non-transitory machine-readable storage medium of claim 15, wherein the action context involves using a seat function controller that controls functions associated with an aircraft seat.

17. The non-transitory machine-readable storage medium of claim 15, wherein the action context is for executing a shopping request using the seat device on the aircraft.

18. The non-transitory machine-readable storage medium of claim 15, wherein the action context is for providing flight data to the user.

19. The non-transitory machine-readable storage medium of claim 15, wherein the action context is for providing assistance by a crew member on the aircraft.

20. The non-transitory machine-readable storage medium of claim 15, wherein the action context is to set a reminder for the user on the aircraft.