Method for drawing map of specific area, robot and electronic device implementing thereof

1. A robot for drawing a map of a specific area, the robot comprising:

a distance measurer configured to measure an accumulated moving distance of the robot moving in the specific area from a first position to a second position;

a feature point matcher configured to perform feature point matching;

a determiner configured to determine whether loop closing is required at the second position based on:

the measured accumulated moving distance, and

information on matching of feature points at the second position; and

a controller configured to control movement of the robot based on a result determined by the determiner,

wherein the determiner is configured to determine that the loop closing is required when the measured accumulated moving distance is greater than a predetermined maximum accumulated moving distance.

2. The robot of claim 1 , wherein the first position is an initial position of the robot in the specific area or an end position of a loop closing which is most recently performed, and

wherein the second position is a current position of the robot.

3. The robot of claim 1 , wherein the controller is further configured to control the robot to move from the second position to the first position when the determiner determines that the loop closing is required.

4. The robot of claim 1 , wherein the determiner is further configured to, when the measured accumulated moving distance is greater than a predetermined minimum accumulated moving distance and is not greater than the predetermined maximum accumulated moving distance:

compare a value associated with the information on matching of feature points at the second position and a predetermined threshold value, and

determine whether the loop closing is required.

5. The robot of claim 4 , further comprising an image acquirer configured to acquire an image when the robot moves,

wherein the value associated with the information on matching of the feature points corresponds to a number of pairs of matching feature points between images divided by a number of feature points extracted from the images, or corresponds to the number of matching feature points in the images, and

wherein the determiner is configured to determine that the loop closing is required when the value associated with the information on matching of the feature points is less than the predetermined threshold value.

6. The robot of claim 4 , further comprising a lidar sensor configured to measure a distance between the robot and an object,

wherein the value associated with the information on matching of the feature points corresponds to an average value of accumulated distance differences in an overlapping area of a lidar scan data, and

wherein the determiner is configured to determine that the loop closing is required when the value associated with the information on matching of the feature points is greater than the predetermined threshold value.

7. The robot of claim 1 , further comprising a communicator configured to communicate with a user terminal that controls movement of the robot,

wherein the controller is further configured to, when the determiner determines that the loop closing is required, generate information on a moving path of the loop closing,

wherein the communicator is configured to:

transmit the generated information on the moving path to the user terminal, and

receive a movement control command of the robot corresponding to the generated information on the moving path from the user terminal, and

wherein the controller is further configured to control the movement of the robot according to the movement control command of the robot.

8. The robot of claim 1 , wherein the feature point matcher is configured to perform feature point matching using an algorithm model based on an artificial neural network, and

wherein the algorithm model comprises:

an input layer comprising an input node,

an output layer comprising an output node, and

one or more hidden layers disposed between the input layer and the output layer and comprising a hidden node,

wherein a weight of an edge that connects nodes and a bias of the nodes are updated through learning.

9. The robot of claim 8 , wherein at least one of sensing data sensed at different time points from each other and information on feature points in images acquired at different time points from each other is input to the input layer of the learned algorithm model, and

wherein the information on matching of the feature points is output to the output layer of the learned algorithm model.

10. A method of drawing, by a robot, a map of a specific area, the method comprising:

measuring, by a distance measurer, an accumulated moving distance of the robot moving in the specific area from a first position to a second position;

matching, by a feature point matcher, feature points in the specific area;

determining, by a determiner, whether loop closing is required at the second position based on:

the measured accumulated moving distance, and

information on matching of a feature point at the second position; and

controlling, by a controller, a movement of the robot based on a result determined by the determiner,

wherein the determining whether the loop closing is required at the second position is performed when the measured accumulated moving distance is greater than a predetermined maximum accumulated moving distance.

11. The method of drawing of claim 10 , wherein the first position is an initial position of the robot in the specific area or an end position of a loop closing which is most recently performed, and

wherein the second position is a current position of the robot.

12. The method of drawing of claim 10 , wherein the determining whether the loop closing is required at the second position includes, when the measured accumulated moving distance is greater than a predetermined minimum accumulated moving distance and is not greater than the predetermined maximum accumulated moving distance:

comparing a value associated with the information on matching of the feature point at the second position with a predetermined threshold value; and

determining whether the loop closing is required.