Method for learning to drive growing autonomous vehicles in real time based on driver interaction

Patent metadata

Field	Value
Publication number	US-2025187627-A1
Application number	US-202418944856-A
Country	US
Kind code	A1
Filing date	Nov 12, 2024
Priority date	Dec 7, 2023
Publication date	Jun 12, 2025
Grant date	—

Abstract

Disclosed herein is a method for learning to drive growing autonomous vehicles in real time based on driver interaction, which includes (a) an autonomous vehicle performing autonomous driving learning based on driving guidance information generated using information received from a driving expert, (b) the autonomous vehicle performing external driving environment recognition matching on a result of recognition of a recognition system and a result of recognition of the driving expert, and (c) the autonomous vehicle performing reviewing and learning after driving training to improve a level of autonomous driving.

First claim

What is claimed is:

1 . A method for learning to drive growing autonomous vehicles in real time based on driver interaction performed by a system for learning to drive growing autonomous vehicles in real time based on driver interaction, the method comprising:
(a) an autonomous vehicle performing autonomous driving learning based on driving guidance information generated using information received from a driving expert; (b) the autonomous vehicle performing external driving environment recognition matching on a result of recognition of a recognition system and a result of recognition of the driving expert; and (c) the autonomous vehicle performing reviewing and learning after driving training to improve a level of autonomous driving.

2 . The method according to claim 1 , wherein the (a) an autonomous vehicle performing autonomous driving learning performs the autonomous driving learning using utterance information of the driving expert.

3 . The method according to claim 2 , wherein the utterance information comprises an instruction for movement of the autonomous vehicle, a confirmation of surrounding recognition situations, a penalty related to intervention in case of issues that occur during driving, a reward for good work, and an additional explanation of situations to watch out for.

4 . The method according to claim 2 , wherein the (a) an autonomous vehicle performing autonomous driving learning classifies a result of converting the utterance information into text, performs matching with a list of predefined driving instruction data, and generates the driving guidance information.

5 . The method according to claim 1 , wherein the (a) an autonomous vehicle performing autonomous driving learning records information on situations where control of driving is taken through intervention of the driving expert in different driving situations.

6 . The method according to claim 1 , wherein the (b) the autonomous vehicle performing external driving environment recognition matching performs the matching using the result of recognition of the recognition system that recognizes driving environments in real time during the driving training and the result of recognition of the driving expert, which is a record stored for a point in time that the driving expert looks at external driving environments.

7 . The method according to claim 1 , wherein the (c) the autonomous vehicle performing reviewing and learning receives information of determination on whether there is an abnormality from the driving expert in relation to driving status information of the autonomous vehicle and information on training instructions, receives a result of reconfirmation of commands by the driving expert, and performs the reviewing and the learning using the information of determination and the result of reconfirmation of commands.

8 . A system for learning to drive growing autonomous vehicles in real time based on driver interaction, comprising:
an input interface device configured to receive utterance information of a driving expert through a user interface; a memory configured to store a program for performing driving learning of an autonomous vehicle using the utterance information; and a processor configured to execute the program, wherein the processor performs the driving learning based on driving guidance information generated using the utterance information, performs matching on a result of recognition of a recognition system and a result of recognition of the driving expert, and performs reviewing and learning after driving training to improve a level of autonomous driving.

9 . The system according to claim 8 , wherein the input interface device receives the utterance information that comprises an instruction for movement of the autonomous vehicle, a confirmation of surrounding recognition situations, a penalty related to intervention in case of issues that occur during driving, a reward for good work, and an additional explanation of situations to watch out for.

10 . The system according to claim 8 , wherein the processor matches a result of converting the utterance information into text with a list of predefined driving instruction data, and generates the driving guidance information.

11 . The system according to claim 8 , wherein the processor records information on situations where control of driving is taken through intervention of the driving expert in different driving situations.

12 . The system according to claim 8 , wherein the processor performs the matching using the result of recognition of the recognition system that recognizes driving environments in real time during the driving training and the result of recognition of the driving expert, which is a record stored for a point in time that the driving expert looks at external driving environments.

13 . The system according to claim 8 , wherein the processor receives information of determination on whether there is an abnormality from the driving expert in relation to driving status information of the autonomous vehicle and information on training instructions, receives a result of reconfirmation of commands by the driving expert, and performs the reviewing and the learning using the information of determination and the result of reconfirmation of commands.

Assignees

• Electronics & Telecommunications Res Inst

Inventors

• Kim Jinwoo
• Kim Ki Tae
• Park Myungwook
• An Kyoung Hwan
• Ji Myungin

Classifications

• B60W2050/0088

  Adaptive recalibration · CPC title

• B60W60/001 Primary

  Planning or execution of driving tasks · CPC title

• B60W50/085

  Changing the parameters of the control units, e.g. changing limit values, working points by control input · CPC title

• B60W2540/21

  Voice · CPC title

• B60W50/14 Primary

  Means for informing the driver, warning the driver or prompting a driver intervention · CPC title