KR102471487B1 - Cleaning robot and controlling method thereof - Google Patents

Abstract

The disclosed invention relates to a cleaning robot and a control method thereof, comprising: a data acquisition unit that obtains actual sensor data by measuring a distance from a current location to a measurement target; a local map acquisition unit acquiring a local map by scanning the vicinity of a current location based on a pre-stored environment map; and a processor performing matching between the local map and real sensor data to determine the coordinates of the current location on the local map, and determining the driving direction based on the current location by calculating angles of main segments of lines existing on the local map. can include

Description

Cleaning robot and its control method {CLEANING ROBOT AND CONTROLLING METHOD THEREOF}

It relates to a cleaning robot and its control method.

A cleaning robot is a device that cleans impurities such as dust from the floor while traveling on its own in an area to be cleaned without a user's manipulation, and performs cleaning tasks while traveling according to a preset driving pattern. In addition, the cleaning robot determines the distance to obstacles such as furniture, walls, and appliances installed in the cleaning area through sensors, and changes direction by itself by selectively driving the left and right motors.

In order for the above-described cleaning robot to perform a smooth cleaning operation in various cleaning environments, it is important to accurately determine its location.

In addition, when the cleaning robot is placed at an angle to the wall in a situation where it starts cleaning, if cleaning is started while maintaining the initial position, an inefficient driving path is created, and a number of areas where cleaning is omitted may occur. .

Accordingly, there is a need to accurately determine the traveling direction of the cleaning robot so that the cleaning robot can travel in all areas of the cleaning space.

Disclosed is to provide a cleaning robot and a control method for determining a running direction of a cleaning robot using a map of an environment in which the cleaning robot performs cleaning and a position of the cleaning robot on the environment map.

According to one aspect, a cleaning robot may include: a data acquisition unit configured to obtain real sensor data by measuring a distance from a current location to a measurement target; a local map acquisition unit acquiring a local map by scanning the vicinity of the current location based on a pre-stored environment map; and performing matching between the local map and the real sensor data to determine the coordinates of the current position with respect to the local map, and calculating the angle of a main segment of a line segment existing in the local map to determine the driving direction based on the current position. A processor that determines; may include.

In addition, the processor may classify a plurality of straight lines existing in the local map according to angles to determine main line segment angles.

Also, the processor may rotate the driving direction of the cleaning robot in place based on the driving direction determined based on the current location coordinates.

In addition, the data acquisition unit acquires the real sensor data while the cleaning robot is stopped, or acquires the real sensor data while the cleaning robot rotates in place at a certain angle, or the cleaning robot is in place. Real sensor data for all directions can be acquired while rotating 360°.

In addition, the local map acquisition unit extracts virtual sensor data by performing a ray casting technique in all directions while rotating 360° from a virtual sensor data extraction location selected with the current location of the cleaning robot as the center. You can get a local map.

Also, the local map acquisition unit may extract a local map having a predetermined size around each of a plurality of local map extraction locations selected around the current location of the cleaning robot from the environment map.

In addition, the processor may include a position estimating unit for determining current position coordinates of the cleaning robot with respect to the local map; a line segment angle calculation unit configured to calculate a main line segment angle of a line segment existing in the local map; and a direction adjuster configured to allow the cleaning robot to rotate in a driving direction according to an angle of the main line segment at a current location.

In addition, the position estimating unit may include a corresponding point obtaining unit acquiring a plurality of corresponding points through data matching between the real sensor data and the local map; a relative position calculating unit calculating a relative position of the real sensor data with respect to the local map using the obtained plurality of corresponding points; a similarity calculation unit calculating a similarity between a plurality of local maps and the real sensor data; and a location determiner configured to determine the relative location of the cleaning robot with respect to the local map having the greatest similarity as the current location of the cleaning robot with respect to the environment map, and determine current location coordinates.

Also, the similarity calculation unit may calculate that the similarity is high as the number of data points commonly included in the local map and the real sensor data increases.

The line segment angle calculation unit may include a straight line acquisition unit acquiring a plurality of straight lines from the local map; and a main line segment angle calculation unit configured to determine a main line segment angle by classifying the plurality of straight lines according to angles.

Also, the straight line acquisition unit may obtain the plurality of straight lines from the local map using a Hough Transform.

The main line segment angle calculation unit may classify the plurality of straight lines according to angles to generate an angle histogram, and determine the main line segment angles by considering the distribution of the angle histogram.

In addition, the main line segment angle calculation unit may generate the angle histogram by merging orthogonal or parallel straight lines with each other.

The main line segment angle calculation unit may determine the main line segment angle of the local map by removing noise from the angle histogram using a low-pass filter.

The cleaning robot may further include a display for outputting in the form of a message that the driving direction of the cleaning robot has been adjusted according to the main line segment angle.

The cleaning robot may further include an audio output unit for outputting in the form of a voice that the driving direction of the cleaning robot has been adjusted according to the angle of the main line.

In addition, the cleaning robot may include a sensor for measuring a distance; The data acquisition unit may obtain actual sensor data by measuring a distance from the sensor to the measurement target using the sensor.

In one aspect, a control method of a cleaning robot may include acquiring a local map by scanning an area around a current location; acquiring real sensor data by measuring a distance from the current location to a measurement object; performing matching between the local map and the real sensor data to determine current location coordinates with respect to the local map; determining a driving direction by calculating an angle of a main line segment of a line segment existing in the local map; and adjusting the driving direction according to the calculated angle of the main line segment.

In the obtaining of the local map, the local map may be obtained by scanning a vicinity of the current location based on a pre-stored environment map in which the cleaning robot travels.

Also, the environment map may be a 2D grid map or a 3D grid map.

In addition, the acquiring of the local map may include performing a ray casting technique in all directions while rotating 360 ° at a virtual sensor data extraction location selected with the current location of the cleaning robot as the center to obtain virtual sensor data. The local map may be obtained by extraction, or a local map having a preset size may be extracted from the environment map around each of a plurality of local map extraction locations selected around the current location of the cleaning robot.

The step of determining the current location coordinates may include acquiring a plurality of corresponding points through data matching between the real sensor data and the local map; calculating a relative position of the real sensor data with respect to the local map using the obtained plurality of corresponding points; calculating a similarity between a plurality of local maps and the real sensor data; and determining the current location of the cleaning robot on the local map having the greatest similarity as the current location of the cleaning robot on the environment map.

In addition, in the step of calculating the degree of similarity, it may be determined that the degree of similarity increases as the number of data points commonly included in the local map and the real sensor data increases.

The determining of the driving direction may include obtaining a plurality of straight lines from the local map; and calculating main line segment angles by classifying the plurality of straight lines according to angles.

The calculating of the main line segment angles may include generating an angle histogram by classifying the plurality of straight lines according to angles; and determining the angle of the main line segment in consideration of the distribution of the angle histogram.

The calculating of the main line segment angles may further include merging orthogonal or parallel straight lines in the angle histogram.

Since the disclosed invention determines the driving direction of the cleaning robot using the environment map where the cleaning robot performs tasks and the position of the cleaning robot on the environment map, it is possible to determine the driving direction more accurately, and thus, the area where cleaning is omitted. Since this does not occur, an effect of improving the reliability of the cleaning operation can be expected.

1 is a diagram showing the overall configuration of a cleaning robot system according to an embodiment.
2 is a diagram schematically illustrating an appearance of a cleaning robot according to an exemplary embodiment.
3 is a control block diagram illustrating a configuration of a cleaning robot according to an exemplary embodiment.
4 is a diagram illustrating an environment map in which a cleaning robot travels according to an exemplary embodiment.
5 is a diagram illustrating a location from which virtual sensor data is extracted from a current location of a cleaning robot according to an embodiment.
FIG. 6 is a diagram illustrating a method of calculating a virtual sensor data extraction position of FIG. 5 .
7 is a diagram illustrating an example of a ray throwing technique according to an embodiment.
8 is a diagram illustrating an example of virtual sensor data extracted using the ray throwing technique of FIG. 7 .
9 is a diagram illustrating a location from which a small local map is extracted from a current location of a cleaning robot according to an embodiment.
10 is a diagram showing an example of a small local map extracted using a small local map extraction position.
FIG. 11 is a control block diagram showing details of the processor of FIG. 3 .
FIG. 12 is a control block diagram showing the position estimation unit of FIG. 11 in detail.
13 is a diagram illustrating an example of real sensor data obtained from a cleaning robot according to an exemplary embodiment.
14 is a diagram illustrating an example of virtual sensor data obtained from a cleaning robot according to an exemplary embodiment.
15 is a diagram illustrating a method of calculating a relative position of real sensor data with respect to a local map according to an embodiment.
16 is a diagram showing a state in which a local map is expanded according to an embodiment.
17 and 18 are diagrams for explaining a state in which a local map and real sensor data are matched according to an embodiment.
FIG. 19 is a diagram showing in detail the configuration of the line segment angle calculator in FIG. 5 .
20 and 21 are angle histograms generated by classifying straight lines obtained from an edge local map according to angles.
22 and 23 are diagrams for explaining a method for determining a driving direction of a cleaning robot.
24 is a flowchart illustrating a method of controlling a cleaning robot according to an exemplary embodiment.
FIG. 25 is a flowchart for explaining in detail a method for estimating the current location of the cleaning robot of FIG. 24 .
26 is a flowchart for explaining in detail the line segment angle extraction method of FIG. 24 .

Objects, specific advantages and novel features of the present invention will become more apparent from the following detailed description and preferred embodiments taken in conjunction with the accompanying drawings. In adding reference numerals to components of each drawing in this specification, it should be noted that the same components have the same numbers as much as possible, even if they are displayed on different drawings. In addition, in describing the present invention, if it is determined that a detailed description of a related air technology may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted. In this specification, terms such as first and second are used to distinguish one component from another component, and the component is not limited to the above terms.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram showing the overall configuration of a cleaning robot system according to an embodiment.

Referring to FIG. 1 , a cleaning robot system 1 includes a cleaning robot 100 that autonomously moves in a certain area while performing tasks, and a device that is separated from the cleaning robot 100 and controls the cleaning robot 100 remotely. 200 and a charging station 300 that is separated from the cleaning robot 100 and charges battery power of the cleaning robot 100 .

The cleaning robot 100 is a device that receives a control command from the device 200 and performs an operation corresponding to the control command, has a rechargeable battery, and has an obstacle sensor capable of avoiding obstacles while driving to clean a work area. It can drive and work autonomously.

In addition, the cleaning robot 100 recognizes its location without prior information on the surrounding environment through a camera or various sensors, and performs localization and map-building to create a map from information on the environment. process can be performed.

The device 200 is a remote control device that wirelessly transmits control commands for controlling the movement of the cleaning robot 100 or performing tasks of the cleaning robot 100, such as a mobile phone, a smart phone, or a portable terminal ( It may include personal digital assistants (PDAs), portable multimedia players (PMPs), laptop computers, digital broadcasting terminals, netbooks, tablets, navigation devices, and the like.

In addition, the device 200 includes all devices capable of implementing various functions using various application programs, such as a digital camera and a camcorder having a built-in wired/wireless communication function.

Also, the device 200 may be a general remote control in a simple form. The remote controller may transmit/receive signals with the cleaning robot 100 using infrared data association (IrDA).

In addition, the device 200 includes Radio Frequency (RF), Wireless Fidelity (Wi-Fi), Bluetooth, Zigbee, near field communication (NFC), Ultra Wide Band: Wireless communication signals can be transmitted and received with the cleaning robot 100 using various methods such as UWB) communication, and if the device 200 and the cleaning robot 100 can transmit and receive wireless communication signals, which method is used? also free

In addition, the device 200 includes a power button for controlling the power of the cleaning robot 100 on and off, a charging return button for instructing the cleaning robot 100 to return to the charging station 300 to charge the battery, and , a mode button for changing the control mode of the cleaning robot 100, a start and stop button for starting and stopping the operation of the cleaning robot 100 or starting, canceling, and confirming control commands, and a dial. can

The charging station 300 is configured to charge the battery of the cleaning robot 100, and a guide member (not shown) guiding the docking of the cleaning robot 100 is provided, and the guide member (not shown) includes a cleaning robot (not shown). A connection terminal (not shown) is provided to charge the power supply unit 130 (see FIG. 2) provided in 100).

2 is a diagram schematically illustrating an appearance of a cleaning robot according to an exemplary embodiment.

As shown in FIG. 2 , the cleaning robot 100 supplies a main body 110 forming an exterior, a cover 120 covering the top of the main body 110, and driving power for driving the main body 110. It includes a power supply unit 130 and a driving unit 140 for moving the main body 110.

The main body 110 forms the exterior of the cleaning robot 100 and supports various parts installed therein.

The power supply unit 130 includes a battery that is electrically connected to the driving unit 140 and other loads for driving the main body 110 to supply driving power. The battery is provided as a rechargeable secondary battery, and is charged by receiving power from the charging station 300 when the main body 110 completes work and is coupled to the charging station 300 .

In addition, the power supply unit 130 is charged by receiving charging current from the charging station 300 when the remaining charge amount is insufficient.

In addition, a caster wheel (not shown) may be installed in front of the main body 110 to change a rotating angle according to the state of the floor on which the cleaning robot 100 moves. The caster wheel supports the cleaning robot 100 by being used to stabilize the posture of the cleaning robot 100 and prevent falling, and is composed of rollers or caster-shaped wheels.

The driving unit 140 is provided on both sides of the central portion of the main body 110 to enable movement operations such as forward, backward and rotational driving while the main body 110 travels.

The driving unit 140 rotates in a forward or backward direction, respectively, according to a command of a processor 180 (see FIG. 3 ) to be described later, so that the cleaning robot 100 can move forward, backward, or rotate. For example, by rotating the driving unit 140 in a forward or backward direction, the cleaning robot 100 travels forward or backward. In addition, while rotating the left driving unit 140 in the backward direction, the right driving unit 140 is rotated in the forward direction so that the cleaning robot 100 rotates in the left direction based on the front, and the right driving unit 140 is rotated in the backward direction. While rotating, the left drive unit 140 is rotated in a forward direction so that the cleaning robot 100 rotates in a right direction with respect to the front.

3 is a control block diagram illustrating a configuration of a cleaning robot according to an exemplary embodiment.

As shown in FIG. 3 , the cleaning robot 100 includes a data acquisition unit 151, a local map acquisition unit 153, a memory 160, an input unit 171, and a display 173 in addition to the components shown in FIG. 2 . ), an audio output unit 175 and a processor 180.

The data acquisition unit 151 is a component that acquires real sensor data of the space where the cleaning robot 100 is currently located, and acquires real sensor data by measuring the distance from the current location of the cleaning robot 100 to the measurement target. can

More specifically, the cleaning robot 100 includes a sensor (not shown) for measuring a distance, and measures the distance from the sensor to the measurement object according to the scan of the 2D sensor or 3D sensor installed in the cleaning robot 100. This is to acquire real sensor data of a real environment where the cleaning robot 100 is located. The sensor may be a 2D sensor or a 3D sensor.

At this time, the 2D sensor may display the distance to the measurement object as (x, y) coordinates with respect to the sensor's reference coordinate system, and the 3D sensor may display the distance to the measurement object as (x, y, z) with respect to the sensor's reference coordinate system ) can be expressed as coordinates. The number of distance data output from the 2D sensor or the 3D sensor may vary depending on the field of view (FoV) and resolution of the sensor.

That is, the data acquisition unit 151 may change the actual sensor data acquisition method of the cleaning robot 100 according to the field of view (FoV) of the 2D sensor or the 3D sensor. When the field of view of the sensor is sufficiently secured, the cleaning robot 100 acquires real sensor data in a stationary state. Alternatively, real sensor data is acquired while rotating at a certain angle in place, or real sensor data for all directions is obtained while rotating 360° in place.

In this case, when the cleaning robot 100 acquires real sensor data while rotating at a certain angle in place, real sensor data outside the field of view of the sensor may be acquired.

The local map acquisition unit 153 may obtain a local map by scanning the vicinity of the current location based on a pre-stored environment map.

More specifically, the local map acquisition unit 153 uses virtual sensor data extracted based on the environment map stored in the memory 160 to obtain a local map around the current location of the cleaning robot 100. get a map).

In this case, an environment map in which the cleaning robot operates may be the same as that of FIG. 4 . It is assumed that the environmental map 500 in which the cleaning robot 100 operates is given in advance. As shown in FIG. 4 , the shape of the environment map 500 may be a 2-dimensional grid map or a 3-dimensional grid map. A grid map is a map in which the surrounding environment of the cleaning robot 100 is divided into small grids and probabilistically expresses the probability that an object is present in each grid, and is also referred to as a probability grid map.

Hereinafter, a case where the cleaning robot 100 is located in region I in the environment map 500 will be described as an example.

In the disclosed invention, as an embodiment of obtaining a local map, three methods using a 2D grid map are proposed. It goes without saying that this method can be extended and applied to a 3D grid map.

5 is a diagram showing a location from which virtual sensor data is extracted from the current location of a cleaning robot according to an embodiment, FIG. 6 is a diagram showing a method for calculating a virtual sensor data extraction location of FIG. 5, and FIG. It is a diagram showing an example of a ray throwing technique according to an embodiment, and FIG. 8 is a diagram showing an example of virtual sensor data extracted using the ray throwing technique of FIG. 7 .

First, the local map acquisition unit 153 may obtain a local map through a method of extracting virtual sensor data.

The local map acquisition unit 153 extracts virtual sensor data by performing a ray casting technique in all directions while the cleaning robot 100 rotates 360 ° at the virtual sensor data extraction location selected with the current location as the center. to obtain a local map.

As shown in FIG. 5 , the cleaning robot 100 selects a plurality of virtual sensor data extraction locations (X) on a map within a predetermined distance from the current location (S).

The virtual sensor data extraction positions (X) are orthogonal to each other with the current position (S) of the cleaning robot 100 as the center, and the end of a point having a radius R is determined as the extraction position (X) (see FIG. 6 ).

In FIG. 6 , four end points of an X character having a side length of 2R become the extraction positions (X) of virtual sensor data.

Thereafter, as shown in FIG. 7, a ray casting technique is performed in all directions of 360 ° from each virtual sensor data extraction position (X), and as shown in FIG. 8, the virtual sensor extract data 8 illustrates an example of virtual sensor data extracted from a virtual sensor data extraction position (X).

FIG. 9 is a diagram showing a location from which a small local map is extracted from a current location of a cleaning robot according to an embodiment, and FIG. 10 is a diagram showing an example of a small local map extracted using the small local map extraction location.

Second, the local map acquisition unit 153 extracts a local map having a preset size around each of a plurality of local map extraction locations selected around the current location of the cleaning robot 100 from the environment map 500. can

In this case, the local map may be divided into a small local map and a large local map, which are wider than the small local map, according to a preset size.

Referring to FIG. 9 , the local map acquisition unit 153 locates a plurality of small local map extraction locations (X) on the map within a predetermined distance centered on the current location (S) of the cleaning robot 100. select

Referring to FIG. 10 , the local map acquisition unit 153 determines the horizontal (W) and vertical (H) lengths of the small local map based on the maximum detection distance of the sensor mounted on the cleaning robot 100. A local map of a predetermined size (W x H) centered on each small local map extraction position (X) is extracted from the environment map. 10 shows an example of a small local map extracted from the displayed small local map extraction position (X).

When the large local map extraction method is applied, the local map acquisition unit 153 sets the current location S of the cleaning robot 100 as the center of the large local map extraction and is wider than the small local map. Areas are extracted from the environment map.

The processor 180 performs matching between the local map and actual sensor data to determine the coordinates of the current position on the local map, calculates the angle of the main line segment of the line segment existing on the local map, and determines the driving direction based on the current position. can At this time, the current location coordinates may be defined as (x, y, θ) with respect to the reference coordinate system {W} displayed on the environment map shown in FIG. 4 . The x and y are position coordinates with respect to the reference coordinate system and θ is a rotation angle.

That is, the processor 180 is a component that controls the overall operation of the cleaning robot 100, recognizes the relative position of the cleaning robot 100 with respect to the local map by performing matching between the local map and sensor data, A travel direction of the cleaning robot 100 in a relative position may be determined.

Also, the processor 180 may classify a plurality of straight lines existing in the local map according to angles to determine main line segment angles.

Also, the processor 180 may rotate the driving direction of the cleaning robot 100 in place based on the driving direction determined based on the current location coordinates.

Referring to FIG. 11 , the processor 180 includes a position estimator 181 for determining the coordinates of the current position of the cleaning robot 100 on a local map, and a line segment for calculating angles of main segments existing on the local map. It may include an angle calculation unit 183 and a direction adjustment unit 185 that allows the cleaning robot 100 to rotate its running direction according to the angle of the main line at its current position.

FIG. 12 is a control block diagram showing the position estimation unit of FIG. 11 in detail.

As shown in FIG. 12 , the position estimating unit 410 may include a corresponding point obtaining unit 411 , a relative position calculating unit 413 , a similarity calculating unit 415 and a position determining unit 417 .

The corresponding point acquisition unit 411 acquires a plurality of corresponding points through data matching between the real sensor data acquired by the data acquisition unit 151 and the local map acquired by the local map acquisition unit 153.

FIG. 13 is a diagram illustrating an example of real sensor data obtained from a cleaning robot according to an exemplary embodiment, and FIG. 14 is a diagram illustrating an example of virtual sensor data obtained from a cleaning robot according to an exemplary embodiment.

Referring to FIGS. 13 and 14 , each sensor data is composed of a plurality of data points. For example, if a certain point P is included as P1 in real sensor data and P2 in virtual sensor data, P1 and P2 are corresponding points. In the case of a 2D sensor, data points constituting real sensor data are each defined as (x, y) with respect to the reference coordinate system of the cleaning robot 100 . Data points constituting the virtual sensor data are each defined as (x, y) with respect to the reference coordinate system {W} (see FIG. 4) of the environment map.

In this way, the corresponding point acquisition unit 411 obtains data points of real sensor data defined as (x, y) with respect to the reference coordinate system of the cleaning robot 100 and (x, y) with respect to the reference coordinate system of the environment map, respectively. Acquire a plurality of data points of virtual sensor data defined by

In addition, the corresponding point acquisition unit 411 may remove outliers from among the obtained plurality of corresponding points. In this case, the outlier is defined as a data point that is not commonly included in real sensor data and virtual sensor data.

The relative position calculation unit 413 calculates the relative position of the real sensor data with respect to the local map, that is, coordinates transformation parameters (Coordinates Transformation Parameter, Rotation and Translation), using the plurality of corresponding points obtained by the corresponding point acquisition unit 411 do.

The relative position calculating unit 413 calculates the remaining corresponding points (Inliers) after removing the outliers as relative positions of real sensor data with respect to the local map.

At this time, the relative position is coordinates transformation coefficient (Coordinates Transformation Parameter, Rotation and Translation) can be obtained by obtaining. Since each data point of the local map is defined with respect to the reference coordinate system of the environment map, the coordinate conversion coefficient for fitting the actual sensor data represents the relative position of the cleaning robot 100 with respect to the environment map.

Referring to FIG. 15 , the {A} coordinate system may be referred to as the reference coordinate system of the local map, and the {B} coordinate system may be referred to as the reference coordinate system of the sensor.

As shown in FIG. 15 , when data points PA constituting a local map and data points PB of real sensor data correspond, a relationship between them can be expressed by [Equation 1].

[Equation 1]

PA = R·PB + t

In this case, R and t are coordinate conversion coefficients, R means a rotation matrix, and t means a translation vector.

That is, if the {A} coordinate system is rotated according to R and moved by t, it becomes the same as the {B} coordinate system. In the disclosed embodiment, R and t are calculated using the least squares method.

The similarity calculating unit 415 calculates the similarity between the plurality of local maps and the real sensor data by using the relative positions of the real sensor data with respect to the local map calculated by the relative position calculating unit 413 .

The local map shown in FIG. 14 can be regarded as a binary image. Therefore, the local map is dilated as shown in FIG. 16 using the dilation method among digital image processing techniques, and the similarity is calculated by fitting real sensor data to the dilated local map.

As shown in FIG. 17 , when the local map and the real sensor data are well-fitted, the number of overlapping parts between the real sensor data and the expanded local map increases, resulting in a high degree of similarity.

At this time, the similarity calculation unit 415 calculates that the similarity is high as the number of data points commonly included in the local map and the real sensor data increases.

On the other hand, as shown in FIG. 18, if the number of data points commonly included in the local map and real sensor data is significantly less or absent than in FIG. 17, or if the coordinate conversion coefficient is incorrectly calculated, the local map and real sensor data The similarity of the data is low.

The location determination unit 417 converts the relative position of the cleaning robot 100 to the local map having the greatest similarity to the current location of the cleaning robot 100 to the environment map using the similarity calculated by the similarity calculation unit 415. Determine the coordinates of the current location.

Referring to FIG. 19 , the line segment angle calculation unit 430 may include a straight line acquisition unit 431 and a main line segment angle calculation unit 433 .

The straight line acquisition unit 431 may obtain a plurality of straight lines from the local map acquired by the local map acquisition unit 153 . In this case, the straight line acquisition unit 431 may acquire a plurality of straight lines from the local map using Hough Transform.

The plurality of straight lines may be a plurality of point sets forming straight lines on the local map or a plurality of point sets forming the local map itself.

More specifically, the straight line acquisition unit 431 obtains a plurality of straight lines for the local map by calculating a voting matrix by performing a Hough transform on a plurality of points constituting the plurality of straight lines extracted from the local map. At this time, each row and column of the voting matrix corresponds to the angle of the line segment and the distance from the origin of the reference coordinate system. A plurality of straight lines obtained according to the Hough transform have angles.

The main line segment angle calculation unit 433 may calculate the main line segment angle by classifying a plurality of straight lines according to angles.

More specifically, the main line segment angle calculating unit 433 may classify a plurality of straight lines according to angles to generate an angle histogram, and determine the main line segment angles by considering the distribution of the angle histogram.

For example, in order to determine the running direction of the cleaning robot 100, since the distance of the line segment is not required, the main line segment angle calculation unit 433 converts the straight line obtained by the straight line acquisition unit 431 according to the angle as shown in FIG. 20. It can be classified and converted into an angular histogram.

Also, the main line segment angle calculation unit 433 may determine the main line segment angle of the local map by removing noise from the angle histogram using a low-pass filter.

Referring to FIG. 21 , the main line segment angle calculation unit 433 performs low pass filtering on the angle histogram and determines an angle having a maximum vote value as the main line segment angle of the local map.

Also, the main line segment angle calculation unit 433 may generate an angle histogram by merging orthogonal or parallel straight lines with each other.

Specifically, the main line segment angle calculation unit 433 may merge straight lines having different angles according to a preset condition. The main line segment angle calculation unit 433 may improve the reliability of the distribution of the main line segment angles by merging straight lines orthogonal to each other or straight lines that are horizontal to each other.

For example, the main line segment angle calculation unit 433 may classify straight lines parallel to each other (180°) as the same angle, or classify straight lines perpendicular to each other (90°) as the same angle. By classifying orthogonal or parallel straight lines as one angle, the main line segment angle calculation unit 433 can classify all straight lines into a specific range.

The main line segment angle calculation unit 433 may generate an angle histogram by merging straight lines parallel to each other or orthogonal to each other so that the main line segment angles exist between -45° and +45°. Specifically, θL and θL+90° shown in FIG. 21 are orthogonal to each other. Therefore, the main line segment angle calculating unit 433 may move θL+90° greater than +45° by 90° and classify it together with θL.

Meanwhile, the main line segment angle calculation unit 433 may calculate an angle at which the largest number of linear components are distributed as the main line segment angle.

Referring to FIG. 21 , the main line segment angle calculator 433 calculates θL where the largest number of linear components exist as the main line segment angle. That is, the main line segment angle calculation unit 433 low-pass filters the angle histogram and determines an angle having the largest vote value as the main line segment angle.

The direction adjuster 185 may determine an angle of a main line in the current position of the cleaning robot 100 as a driving direction, and may rotate the cleaning robot 100 in place in the determined driving direction.

For example, as shown in FIG. 22 , when the cleaning robot 100 is positioned in the direction of the rotation angle θR with respect to the reference coordinate system {W}, the cleaning robot 100 is positioned in the same direction as the main line segment angle θL. is to rotate

The memory 160 stores an environment map in which the cleaning robot 100 travels, an operating program for operation of the cleaning robot 100, a driving pattern, location information and obstacle information of the cleaning robot 100 obtained during the driving process. can be saved

In addition, the memory 160 stores control data for controlling the operation of the cleaning robot 100, reference data used during operation control of the cleaning robot 100, and data generated while the cleaning robot 100 performs a predetermined operation. User input information such as motion data and setting data input by the device 200 to cause the cleaning robot 100 to perform a predetermined motion may be stored.

In addition, the memory 160 may include a non-volatile memory device such as a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmed Read Only Memory (EPRM), a flash memory, It may be implemented as a storage medium such as a volatile memory device such as random access memory (RAM), a hard disk, a card-type memory (eg SD or XD memory, etc.), and an optical disk. However, the memory 160 is not limited thereto, and various storage media that designers can consider may be used.

The input unit 171 may receive various types of control information for operating the cleaning robot 100 through user manipulation.

The display 173 is a component for outputting in the form of a message that the driving direction of the cleaning robot 100 has been adjusted according to the angle of the main line. At this time, the display 173 may display using icons or LEDs. The display 173 is not limited to notifying completion of the driving direction adjustment of the cleaning robot 100, and may also output a current location detection notification even when the current location is ascertained.

In this way, in the case of an LCD UI capable of displaying icons or text, the display 173 displays the position recognition and driving direction adjustment state of the cleaning robot 100 using icons or text so that the user can easily check the display 173 .

In addition, in the case of the LED UI, the display 173 displays the driving direction adjustment state of the cleaning robot 100 using lighting or blinking and a difference in duration so that the user can easily check it.

In addition, the display 173 displays the user's input information and the operating state of the cleaning robot 100 according to the display control signal of the processor 180, and displays various information received from the cleaning robot 100 and the user through the device. It is possible to display various control commands received from

For example, the display 173 displays the time or the speed of the cleaning robot 100 when a time setting, reservation setting or direct control command is input, and in other cases, the current time, reservation information, battery level, moving distance, moving route, etc. can display.

In addition, the display 173 includes a liquid crystal display, a thin film transistor-liquid crystal display, an organic light-emitting diode, a flexible display, and a 3D display. (3D display) may include at least one.

The sound output unit 175 is a component for outputting in the form of voice that the driving direction of the cleaning robot 100 has been adjusted according to the angle of the main line. The sound output unit 175 is not limited to notifying completion of the driving direction adjustment of the cleaning robot 100, and may also output a current location detection notification even when the current location is ascertained.

In addition, the sound output unit 175 outputs the operating state of the cleaning robot 100 and the driving direction adjustment state of the cleaning robot 100 as sound (eg, a beep sound) according to the sound control signal of the processor 180. It may include a speaker that

In addition, the sound output unit 175 further includes a digital-to-analog converter (DAC) that converts the digitized electrical signal into an analog, an amplifier that amplifies the electrical signal converted into an analog by the digital-to-analog converter, and the like. can include

24 is a flowchart for explaining a control method of a cleaning robot according to an embodiment, FIG. 25 is a flowchart for explaining in detail a method for estimating a current position of the cleaning robot of FIG. 24 , and FIG. 26 is a line segment angle of FIG. 24 . It is a flowchart for explaining the extraction method in detail.

Referring to FIG. 24 , the cleaning robot 100 may obtain a local map by scanning the vicinity of the current location (S110).

In this case, the cleaning robot 100 may acquire a local map by scanning the vicinity of a current location based on a pre-stored environment map in which the cleaning robot 100 travels. The environment map may be a 2D grid map or a 3D grid map.

The cleaning robot 100 obtains the local map by extracting virtual sensor data by performing a ray casting technique in all directions while rotating 360° from a virtual sensor data extraction position selected around its current position. Alternatively, a local map having a preset size may be extracted from the environment map around each of a plurality of local map extraction locations selected around the current location of the cleaning robot 100 .

Next, the cleaning robot 100 may acquire real sensor data by measuring the distance from the current location to the object to be measured, and perform matching between the local map and the real sensor data to determine the coordinates of the current location with respect to the local map (S130). ).

Referring to FIG. 24 , the cleaning robot 100 may acquire real sensor data by measuring a distance from a current location to a measurement target (S131).

More specifically, the cleaning robot 100 includes a sensor (not shown) for measuring a distance, and measures the distance from the sensor to the measurement object according to the scan of the 2D sensor or 3D sensor installed in the cleaning robot 100. This is to acquire real sensor data of a real environment where the cleaning robot 100 is located. The sensor may be a 2D sensor or a 3D sensor.

Next, the cleaning robot 100 may acquire a plurality of correspondence points through data matching between real sensor data and the local map (S133).

For example, the cleaning robot 100 uses data points of real sensor data defined as (x, y) for the reference coordinate system and virtual sensor data defined as (x, y) for the reference coordinate system of the environment map. To obtain corresponding points corresponding between a plurality of data points of .

Next, the cleaning robot 100 may remove outliers from among the plurality of corresponding points (S135).

Next, the cleaning robot 100 may calculate a relative position of real sensor data with respect to the local map using the obtained plurality of corresponding points (S137).

The cleaning robot 100 calculates a relative position of real sensor data with respect to a local map, that is, coordinates transformation parameters (Coordinates Transformation Parameter, Rotation, and Translation) by using a plurality of corresponding points.

Next, the cleaning robot 100 may calculate a similarity between the plurality of local maps and actual sensor data (S139).

In this case, it may be determined that the similarity is high as the number of data points commonly included in the local map and the real sensor data increases.

Next, the cleaning robot 100 may determine the current location of the cleaning robot on the local map having the greatest similarity as the current location of the cleaning robot on the environment map (S141). In this case, the current location of the cleaning robot may be in the form of current location coordinates.

Next, the cleaning robot 100 may calculate an angle of a main line segment of a line segment existing in the local map to determine a driving direction (S150).

Referring to FIG. 26 , the cleaning robot 100 may obtain a plurality of straight lines from the local map (S151 and S153).

Next, the cleaning robot 100 may classify the plurality of straight lines according to angles to calculate the main line segment angle (S155).

Step S155 may include generating an angle histogram by classifying a plurality of straight lines according to angles, and determining main segment angles in consideration of a distribution of the angle histogram.

Also, step S155 may further include merging orthogonal or parallel straight lines in the angle histogram.

Next, the cleaning robot 100 may adjust the driving direction according to the calculated main line segment angle (S170).

Next, the cleaning robot 100 may output in the form of a message or voice that the running direction of the cleaning robot is adjusted according to the angle of the main line (S190).

Although the present invention has been described in detail through specific examples, this is for explaining the present invention in detail, the present invention is not limited thereto, and within the technical spirit of the present invention, by those skilled in the art It is clear that modifications and improvements are possible.

All simple modifications or changes of the present invention fall within the scope of the present invention, and the specific protection scope of the present invention will be clarified by the appended claims.

100: cleaning robot 151: data acquisition unit
153: local map acquisition unit 160: memory
171: input unit 173: display
175: sound output unit 180: processor
181, 410: position estimation unit 183: line segment angle calculation unit
185: direction adjustment unit 411: correspondence point acquisition unit
413: relative position calculation unit 415: similarity calculation unit
417: positioning unit 430: line segment angle calculation unit
431: straight line acquisition unit 433: main line segment angle calculation unit

Claims (26)

a data acquisition unit that obtains real sensor data by measuring a distance from a current location to a measurement target;
a local map acquisition unit acquiring a local map by scanning the vicinity of the current location based on a pre-stored environment map; and
Matching between the local map and the actual sensor data is performed to determine the coordinates of the current location on the local map, and the angle of a main line segment existing on the local map is calculated to determine the driving direction based on the current location. processor to determine;
including,
The local map acquisition unit,
Extracting a local map having a predetermined size around each of a plurality of local map extraction locations selected around the current location of the cleaning robot from the environment map;
The data acquisition unit,
The real sensor data is acquired while the cleaning robot is stationary, or the real sensor data is obtained while the cleaning robot is rotating at a certain angle in place, or the cleaning robot is rotating 360° in place and moving in all directions. Obtain real sensor data for
The local map acquisition unit,
The cleaning robot obtains the local map by extracting virtual sensor data by performing a ray casting technique in all directions while rotating 360° from a virtual sensor data extraction position selected around the current position of the cleaning robot.
According to claim 1,
the processor,
A cleaning robot configured to determine an angle of a main line segment by classifying a plurality of straight lines existing in the local map according to angles.
According to claim 1,
the processor,
The cleaning robot rotates the driving direction of the cleaning robot in place based on the driving direction determined based on the current location coordinates.
delete delete delete According to claim 1,
the processor,
a position estimator for determining the coordinates of the current position of the cleaning robot with respect to the local map;
a line segment angle calculation unit configured to calculate a main line segment angle of a line segment existing in the local map; and
a direction adjuster for allowing the cleaning robot to rotate in a running direction according to an angle of the main line segment at a current location;
A cleaning robot comprising a.
According to claim 7,
The location estimation unit,
a correspondence point acquisition unit acquiring a plurality of correspondence points through data matching between the real sensor data and the local map;
a relative position calculating unit calculating a relative position of the real sensor data with respect to the local map using the obtained plurality of corresponding points;
a similarity calculation unit calculating a similarity between a plurality of local maps and the real sensor data; and
a position determining unit determining a relative position of the cleaning robot with respect to a local map having the greatest similarity as a current position of the cleaning robot with respect to an environment map and identifying current location coordinates;
A cleaning robot comprising a.
According to claim 8,
The similarity calculator,
The cleaning robot calculating that the degree of similarity increases as the number of data points commonly included in the local map and the real sensor data increases.
According to claim 7,
The line segment angle calculator,
a straight line acquisition unit acquiring a plurality of straight lines from the local map; and
a main line segment angle calculator configured to determine a main line segment angle by classifying the plurality of straight lines according to angles;
A cleaning robot comprising a.
According to claim 10,
The straight line acquisition unit,
A cleaning robot that acquires the plurality of straight lines in the local map using a Hough Transform.
According to claim 10,
The main line segment angle calculator,
The cleaning robot classifies the plurality of straight lines according to angles to generate an angle histogram, and determines an angle of the main line segment in consideration of a distribution of the angle histogram.
According to claim 12,
The main line segment angle calculator,
A cleaning robot that generates the angle histogram by merging orthogonal or parallel straight lines with each other.
According to claim 12,
The main line segment angle calculator,
The cleaning robot determines an angle of a main line segment of the local map by removing noise from the angle histogram using a low-pass filter.
According to claim 1,
a display for outputting in the form of a message that the driving direction of the cleaning robot has been adjusted according to the angle of the main line;
A cleaning robot further comprising a.
According to claim 1,
an audio output unit for outputting in the form of a voice that the driving direction of the cleaning robot has been adjusted according to the angle of the main line;
A cleaning robot further comprising a.
According to claim 1,
a sensor for measuring distance; Including more,
The data obtaining unit obtains real sensor data by measuring a distance from the sensor to the measurement target using the sensor.
acquiring a local map by scanning around the current location;
acquiring real sensor data by measuring a distance from the current location to a measurement object;
performing matching between the local map and the real sensor data to determine current location coordinates with respect to the local map;
determining a driving direction by calculating an angle of a main line segment of a line segment existing in the local map; and
adjusting a driving direction according to the calculated main line segment angle;
including,
Acquiring the real sensor data,
The real sensor data is acquired while the cleaning robot is stationary, or the real sensor data is obtained while the cleaning robot rotates at a certain angle in place, or the cleaning robot rotates 360° in place and measures data in all directions. Acquiring real sensor data;
The step of obtaining the local map,
Acquiring the local map by scanning the vicinity of the current location based on a pre-stored environment map in which the cleaning robot travels;
Extracting a local map having a predetermined size from the environment map centered on each of a plurality of local map extraction locations selected around the current location of the cleaning robot;
A cleaning robot that obtains the local map by extracting virtual sensor data by performing a ray casting technique in all directions while rotating 360 ° from a virtual sensor data extraction position selected around the current position of the cleaning robot. control method.
delete According to claim 18,
The method of controlling a cleaning robot in which the environment map is a 2-dimensional grid map or a 3-dimensional grid map.
delete According to claim 18,
The step of determining the current location coordinates,
obtaining a plurality of correspondence points through data matching between the real sensor data and the local map;
calculating a relative position of the real sensor data with respect to the local map using the obtained plurality of corresponding points;
calculating a similarity between a plurality of local maps and the real sensor data; and
determining the current location of the cleaning robot on the local map having the greatest similarity as the current location of the cleaning robot on the environment map;
A control method of a cleaning robot comprising a.
The method of claim 22,
In the step of calculating the similarity,
determining that the similarity is higher as the number of data points commonly included in the local map and the real sensor data increases.
According to claim 18,
The step of determining the driving direction is,
obtaining a plurality of straight lines from the local map; and
calculating main segment angles by classifying the plurality of straight lines according to angles;
A control method of a cleaning robot comprising a.
According to claim 24,
The step of calculating the main line segment angle,
generating an angle histogram by classifying the plurality of straight lines according to angles; and
determining the angle of the main line segment in consideration of the distribution of the angle histogram;
A control method of a cleaning robot comprising a.
◈Claim 26 was abandoned upon payment of the setup registration fee.◈ According to claim 25,
The step of calculating the main line segment angle,
merging orthogonal or parallel straight lines in the angle histogram;
A control method of a cleaning robot further comprising a.

Images