Artificial intelligence apparatus for sharing information of stuck area and method for the same

1. A method for sharing information of a stuck area, the method comprising:

receiving three-dimensional (3D) sensor data and bumper sensor data from a first cleaner;

generating surrounding situation map data based on the 3D sensor data and the bumper sensor data;

generating learning data by labeling area classification data for representing whether the surrounding situation map data corresponds to the stuck area;

training a stuck area classification model based on the learning data; and

transmitting the trained stuck area classification model to a second cleaner,

wherein the generating learning data includes:

acquiring, from the surrounding situation map data, escape path data until the first cleaner completes an escape from the stuck area;

determining classification of the stuck area based on the escape path data; and

labeling the area classification data for representing the classified stuck area, on the surrounding situation map data, and

wherein the acquiring of the escape path data includes: determining that the first cleaner completes the escape from the stuck area, when a preset number of times or less of bumper events occurs for a preset time from a time point at which the surrounding situation map data is determined as corresponding to the stuck area.

2. The method of claim 1 , wherein the generating of the surrounding situation map data includes:

generating the surrounding situation map data based on the 3D sensor data and the bumper sensor data for a preset time.

3. The method of claim 1 , wherein the generating of the learning data includes:

determining whether the surrounding situation map data corresponds to the stuck area, based on the bumper sensor; and

labeling area classification data for classifying surrounding situation map data, which corresponds to the stuck area, as the stuck area.

4. The method of claim 3 , wherein the determining of whether the surrounding situation map data corresponds to the stuck area includes:

determining that the surrounding situation map data corresponds to the stuck area for a preset time, when a bumper event occurs in a preset number of times for the preset time.

5. The method of claim 3 , wherein the bumper sensor data includes:

data on a bumper event sensed by a plurality of bumper sensors of the first cleaner, and

wherein the determining of whether the surrounding situation map data corresponds to the stuck area includes:

determining that the surrounding situation map data corresponds to the stuck area for a preset time, when the bumper event occurs in the plurality of bumper sensors of the first cleaner for the preset time.

6. The method of claim 3 , wherein the receiving of the 3D sensor data and the bumper sensor data from the first cleaner includes:

receiving path data of the first cleaner.

7. The method of claim 1 , wherein the classification of the stuck area includes:

acquiring an escape direction based on the escape path data; and

determining the classification of the stuck area in the escape direction.

8. The method of claim 1 , wherein the second cleaner receives the trained stuck area classification model from the first cleaner, stores the received and trained stuck area classification model, provides 3D sensor data, which is acquired during traveling, to the trained stuck area classification model, acquires area classification data representing whether the second cleaner is positioned in the stuck area, and performs escape travelling when the second cleaner is positioned in the stuck area.

9. An artificial intelligence (AI) apparatus, the AI apparatus comprising:

a communications interface configured to receive 3D sensor data and bumper sensor data from a first cleaner;

a processor configured to generate surrounding situation map data based on the 3D sensor data and the bumper sensor data; and

a learning processor configured to generate learning data by labeling area classification data for representing whether the surrounding situation map data corresponds to the stuck area, and to train a stuck area classification model based on the learning data; and

wherein the processor is configured to:

transmit the trained stuck area classification model to a second cleaner,

wherein the learning processor is configured to:

acquire, from the surrounding situation map data, escape path data until the first cleaner completes an escape from the stuck area, determine classification of the stuck area based on the escape path data, and label the area classification data for representing the classified stuck area, on the surrounding situation map data, and

wherein the learning processor is further configured to determine that the first cleaner completes the escape from the stuck area, when a preset number of times or less of bumper events occurs for a preset time from a time point at which the surrounding situation map data is determined as corresponding to the stuck area.

10. The AI apparatus of claim 9 , wherein the processor is configured to:

generate the surrounding situation map data based on the 3D sensor data and the bumper sensor data for a preset time.

11. The AI apparatus of claim 9 , wherein the learning processor is configured to:

determine whether the surrounding situation map data corresponds to the stuck area, based on the bumper sensor; and

label area classification data for classifying surrounding situation map data corresponding to the stuck area as the stuck area.

12. The AI apparatus of claim 11 , wherein the learning processor is configured to:

determine that the surrounding situation map data corresponds to the stuck area for a preset time, when a bumper event occurs in a preset number of times for the preset time.

13. The AI apparatus of claim 11 , wherein the bumper sensor data includes:

data on a bumper event sensed by a plurality of bumper sensors of the first cleaner, and

wherein the learning processor is configured to:

determine that the surrounding situation map data corresponds to the stuck area for a preset time, when the bumper event occurs in the plurality of bumper sensors of the first cleaner for the preset time.

14. The AI apparatus of claim 11 , wherein the communications interface receives path data of the first cleaner.

15. The AI apparatus of claim 9 , wherein the learning processor is further configured to:

acquire an escape direction based on the escape path data; and

determining the classification of the stuck area in the escape direction.

16. The AI apparatus of claim 9 , wherein the second cleaner receives the trained stuck area classification model from the first cleaner, stores the received and trained stuck area classification model, provides 3D sensor data, which is acquired d during traveling, to the trained stuck area classification model, acquires area classification data representing whether the second cleaner is positioned in the stuck area, and performs escape travelling when the second cleaner is positioned in the stuck area.