Artificial intelligence device for guiding furniture placement and method of operating the same

1. An artificial intelligence device for guiding furniture placement, the artificial intelligence device comprising:

a transceiver configured to receive indoor area map data and indoor area image data from at least one external device; and

a processor configured to generate first furniture placement map data in which a placement position of at least one piece of furniture is mapped, provide the first furniture placement map data to a furniture placement model, and obtain second furniture placement map data in which the placement position of the at least one piece of furniture is changed, based on the indoor area map data and the indoor area image data,

wherein the processor is further configured to:

extract two points as gaze measurement points based on the second furniture placement map data;

extract a reward parameter based on the at least one piece of furniture not being located between the two points to secure a gaze of a user;

extract a penalty parameter based on the at least one piece of furniture being located between the two points and interfering with the gaze of the user; and

provide the reward parameter or the penalty parameter to the furniture placement model to train the furniture placement model,

wherein the two points are selected as the gaze measurement points from a plurality of representative points of at least one user activity area and a plurality of representative points of at least one indoor/outdoor passage area.

2. The artificial intelligence device of claim 1 , wherein the furniture placement model is a neural network trained by labeling at least one training indoor area map data with at least one training furniture placement location.

3. The artificial intelligence device of claim 1 , wherein the transceiver is further configured to receive indoor moving route data from the at least external device, and

wherein the processor is further configured to extract a parameter for training the furniture placement model based on the second furniture placement map data and the indoor moving route data, and provide the parameter to the furniture placement model to train the furniture placement model.

4. The artificial intelligence device of claim 3 , wherein the processor is configured to extract a moving distance between indoor areas based on the second furniture placement map data and the indoor moving route data, extract a second reward parameter or a second penalty parameter based on the extracted inter-area moving distance, and provide the second reward parameter or the second penalty parameter to the furniture placement model to train the furniture placement model.

5. The artificial intelligence device of claim 3 , wherein the processor is configured to extract a moving distance between a plurality of pieces of furniture based on the second furniture placement map data and the indoor moving route data, extract a second reward parameter or a second penalty parameter based on the extracted moving distance between the pieces of furniture, and provide the second reward parameter or the second penalty parameter to the furniture placement model to train the furniture placement model.

6. The artificial intelligence device of claim 1 , further comprising:

a display configured to display a placement guide screen that displays the second furniture placement map data.

7. The artificial intelligence device of claim 1 , wherein the transceiver is further configured to receive indoor area map data and indoor area image data which are obtained during driving by a robot cleaner.

8. A method of operating an artificial intelligence device for guiding furniture placement, the method comprising:

receiving indoor area map data and indoor area image data from at least one external device;

generating first furniture placement map data in which a placement location of at least one piece of furniture is mapped based on the received indoor area map data and the received indoor area image data;

providing the first furniture placement map data to a furniture placement model to obtain second furniture placement map data in which the placement location of at the least one piece of furniture is changed;

extracting two points as gaze measurement points based on the second furniture placement map data;

extracting a reward parameter based on the at least one piece of furniture not being located between the two points to secure a gaze of a user;

extracting a penalty parameter based on the at least one piece of furniture being located between the two points and interfering with the gaze of the user; and

providing the reward parameter or the penalty parameter to the furniture placement model to train the furniture placement model,

wherein the two points are selected as the gaze measurement points from a plurality of representative points of at least one user activity area and a plurality of representative points of at least one indoor/outdoor passage area.

9. The method of claim 8 , wherein the furniture placement model is a neural network trained by labeling at least one training indoor area map data with at least one training furniture placement location.

10. The method of claim 8 , further comprising:

receiving indoor moving route data from the at least one external device;

extracting a parameter for training the furniture placement model based on the second furniture placement map data and the indoor moving route data; and

training the furniture placement model by providing the parameter to the furniture placement model.

11. The method of claim 10 , wherein the extracting of the parameter includes:

extracting a moving distance between indoor areas based on the second furniture placement map data and the indoor moving route data; and

extracting a second reward parameter or a second penalty parameter based on the extracted inter-area moving distance, and

wherein the training of the furniture placement model includes training the furniture placement model by providing the second reward parameter or the second penalty parameter to the furniture placement model.

12. The method of claim 10 , wherein the extracting of the parameter includes:

extracting moving distances between a plurality of pieces of furniture based on the second furniture placement map data and the indoor moving route data; and

extracting a second reward parameter or a second penalty parameter based on the extracted inter-furniture moving distance, and

wherein the training of the furniture placement model includes training the furniture placement model by providing the second reward parameter or the second penalty parameter to the furniture placement model.

13. The method of claim 8 , further comprising:

displaying a placement guide screen that displays the second furniture placement map data.

14. The method of claim 8 , wherein receiving indoor area map data and indoor area image data includes receiving indoor area map data and indoor area image data obtained during driving by a robot cleaner.