KR20230137080A - A mobile robot - Google Patents

Abstract

An embodiment of the present invention for achieving the above object includes a traveling unit that moves the main body within the traveling area; a detection unit that acquires distance detection information regarding the distance to an object outside the main body; and generate a grid map for the driving area from the distance detection information, search for open nodes for a plurality of driving nodes on the driving path from the lidar information among the distance detection information, and select each of the open nodes from the grid map. We present a mobile robot that includes a control unit that re-determines whether or not it is open and generates a topology map. Therefore, efficiency can be improved by minimizing unnecessary driving when searching for a space that requires additional driving. In addition, during additional search driving, the search for unsearched areas through the grid map in the topology map is performed once again, thereby minimizing duplication and ensuring reliability.

Description

A MOBILE ROBOT}

The present invention relates to mobile robots, and more specifically, to detection and control technology for mobile robots for generating driving maps.

The field of robot applications continues to expand, with medical robots, aerospace robots, etc. being developed, and household robots that can be used in general homes are also being created. Among these robots, those that can travel on their own are called mobile robots.

A representative example of a mobile robot used at home is a mobile robot.

Various technologies are known for detecting the environment and users around a mobile robot through various sensors provided in the mobile robot. Additionally, technologies are known in which mobile robots learn and map cleaning areas on their own and determine their current location on the map. Mobile robots that travel and clean a cleaning area in a preset manner are known.

In order to perform a set task such as cleaning, it is necessary to accurately create a map of the driving area and accurately determine the current location of the mobile robot on the map in order to move to a certain location within the driving area.

In the case of prior art (Korean Patent Publication No. 10-2010-0031878), a path is generated based on the uncertainty of the positions of feature points extracted from images obtained when a mobile robot explores an unknown environment, and according to the generated path, A driving technology is introduced. A path based on the uncertainty of feature points is created to increase the accuracy of the feature point map of a mobile robot or to increase the accuracy of self-location recognition.

In addition, another prior art, Korean Patent Publication No. 2021-0009011, discloses conducting a search drive to create a map for driving during cleaning.

In this other prior art, when driving to create a map, data within a real-time sensing range is acquired for driving in an unknown area, nodes are set accordingly, and a grid map is calculated based on the node information. For such a grid map, map information is generated by searching for boundaries based on image data and providing an updated final grid map.

However, with other conventional technologies, only data within the sensing range is acquired in real time while the robot is driving, and a basic map is created based on that, so it is possible to drive between the current robot position and the boundary found in the image. There is a risk that obstacles may arise that are not set in stone.

Additionally, in other prior art, while the robot is driving, a topology node is added to the path at a location where additional search is required to indicate that future search driving is necessary.

However, in the prior art, when a node is created on the driving path where additional exploration is required, the direction requiring additional driving at that node is divided into sections, and the initial direction of the robot is determined by the angle of the structure inside the building and the additional driving direction. If it is not correct, the direction setting may be misaligned, making driving difficult.

In addition, since only data within the sensing range is acquired while the mobile robot is traveling, there may be cases where it is not possible to drive from the current mobile robot position to the boundary found in the image.

Meanwhile, a method of driving while the robot is driving by adding a topology node to the path at a location that requires additional exploration to indicate the need for further exploration was also proposed. However, when using a topology map in this way, unsearched areas may occur because not all unsearched areas are represented.

Korean Patent Publication No. 10-2010-0031878 A (2010.03.25.) Korean Patent Publication No. 10-2021-0009011 A (2021.01.26.)

In order to solve the above problems, the purpose of the present invention is to provide a map generation method that can minimize unnecessary driving when searching for a space where driving is required.

Additionally, by updating the topology node map by cross-applying the topology node map and grid map, a reliable map is created without unexplored areas.

Additionally, the present invention proposes a method for preventing duplicate searches when switching to manual mapping while performing automatic mapping and when the grid map is expanded.

An embodiment of the present invention for achieving the above object includes a traveling unit that moves the main body within the traveling area; a detection unit that acquires distance detection information regarding the distance to an object outside the main body; and generate a grid map for the driving area from the distance detection information, search for open nodes for a plurality of driving nodes on the driving path from the lidar information among the distance detection information, and select each of the open nodes from the grid map. We present a mobile robot that includes a control unit that re-determines whether or not it is open and generates a topology map.

The sensor unit may include a LiDAR sensor that irradiates a laser to an object outside the main body and calculates the distance detection information by reflected light.

The control unit may perform the ray casting for each of a plurality of traveling nodes to search for open space, and may set the traveling node where the open space exists as the open node.

The control unit may close the open node when the open node has an open space facing the direction in which the open space of the open node exists based on the grid map.

For each open node, the control unit may divide the area around the open node from the grid map, determine whether an unexplored area exists around the open node, and determine whether to close the open node.

The control unit divides the driving area into a plurality of sub-areas and generates the grid map and the topology map for each sub-area, and the sub-area allows the mobile robot to travel the driving area for a predetermined time or a predetermined distance. It can be divided into the area when driving.

When it is re-determined whether the open node is open for the sub-area, the control unit may determine whether an unsearched area for the sub-area exists from the grid map.

If the unsearched area exists, the controller may move to the driving node around the unsearched area and perform a search drive to see if an open node exists in the unsearched area.

The control unit may simultaneously perform node-based search and grid map-based search while driving along the travel node.

The control unit may determine that the mobile robot can travel by ray casting from the travel node and that a space in which the mobile robot has not previously driven is the open space.

The control unit may perform 360-degree ray casting around the travel node.

The sensor unit may further include an obstacle detection sensor for generating the grid map.

According to the present invention, one or more of the following effects are achieved.

First, efficiency can be improved by minimizing unnecessary driving when searching for spaces that require additional driving.

The present invention can secure reliability while minimizing duplication by performing additional search of untraveled areas through a grid map in the topology map during additional search driving.

Additionally, the accuracy of open nodes can be improved by using information from the grid map when checking open nodes.

In addition, when switching to manual mapping in some sections during automatic mapping, search and calculation are simplified by creating a topology map again from that location without repeated learning of the expanded grid map.

Meanwhile, the effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description of the claims.

Figure 1 is a perspective view showing a mobile robot and a charging base for charging the mobile robot according to an embodiment of the present invention.
FIG. 2 is an elevation view of the mobile robot of FIG. 1 viewed from above.
FIG. 3 is an elevation view of the mobile robot of FIG. 1 viewed from the front.
FIG. 4 is an elevation view of the mobile robot of FIG. 1 viewed from below.
Figure 5 is a diagram showing an example of a mobile robot according to another embodiment of the present invention.
FIG. 6 is a block diagram showing the control relationship between the main components of the mobile robot of FIG. 1 or FIG. 5.
Figure 7 is a diagram explaining the operation of a LiDAR sensor provided in a mobile robot according to an embodiment of the present invention.
Figure 8 is a flow chart of a map creation process for a mobile robot according to an embodiment of the present invention.
FIGS. 9A to 9E are diagrams showing the map creation process of the mobile robot according to the flowchart of FIG. 8.
Figure 10 is a flow chart of a map creation process for a mobile robot according to another embodiment of the present invention.
Figure 11a shows open space search according to a comparative example, and Figure 18b shows open space search according to the present invention.

In the magnitude comparison expressed verbally/mathematically throughout this description, 'less than or equal to (less than)' and 'less than' are easily interchangeable from the point of view of an ordinary technician, and 'greater than or equal to' From the point of view of a person skilled in the art, '(more than)' and 'greater than (greater than)' are degrees that can be easily substituted for each other, and of course, even if they are substituted in implementing the present invention, there is no problem in achieving the effect.

The mobile robot 100 of the present invention refers to a robot that can move on its own using wheels, etc., and may be a home helper robot or a mobile robot.

1 to 4 are diagrams showing the appearance of a mobile robot and a charging station for charging the mobile robot according to an embodiment of the present invention.

Figure 1 is a perspective view showing a mobile robot and a charging base for charging the mobile robot according to an embodiment of the present invention, Figure 2 is an elevation view of the mobile robot of Figure 1 viewed from above, and Figure 3 is a mobile robot of Figure 1. It is an elevation view viewed from the front, and FIG. 4 is an elevation view of the mobile robot of FIG. 1 viewed from the bottom.

The mobile robot 100 includes a main body 110. Hereinafter, in defining each part of the main body 110, the part facing the ceiling in the travel area is defined as the upper surface part (see Figure 2), and the part facing the floor in the driving area is defined as the bottom part (see Figure 4). And, of the portion forming the circumference of the main body 110 between the upper and lower surfaces, the portion facing the traveling direction is defined as the front portion (see FIG. 3). Additionally, the portion facing the opposite direction from the front portion of the main body 110 may be defined as the rear portion. The main body 110 may include a case 111 that forms a space in which various parts constituting the mobile robot 100 are accommodated.

The mobile robot 100 may include, for example, at least one driving wheel 136 that moves the main body 110. The driving wheel 136 may be driven and rotated by, for example, at least one motor (not shown) connected to the driving wheel 136.

For example, the driving wheels 136 may be provided on the left and right sides of the main body 110, respectively, and are hereinafter referred to as the left wheel 136 (L) and the right wheel 136 (R), respectively.

The left wheel 136(L) and the right wheel 136(R) may be driven by a single drive motor, but if necessary, the left wheel drive motor and the right wheel 136(R) drive the left wheel 136(L). ) may each be provided with a right-wheel drive motor that drives the motor. The driving direction of the main body 110 can be switched to the left or right by making a difference in the rotation speed of the left wheel 136(L) and the right wheel 136(R).

The mobile robot 100 may include, for example, a suction unit 330 for sucking foreign substances, brushes 154 and 155 for mopping, a dust bin for storing collected foreign substances, and a mopping unit for mopping. there is.

For example, an intake port 150h through which air is sucked may be formed on the bottom of the main body 110, and a suction port 150h may be formed inside the main body 110 to provide suction force so that air can be sucked in through the intake port 150h. The device may be provided with a dust bin that collects dust sucked in with air through the inlet 150h.

For example, the mobile robot 100 may include a case 111 that forms a space in which various parts constituting the mobile robot 100 are accommodated. An opening (not shown) for inserting and removing the dust bin may be formed in the case 111, and a dust bin cover 112 that opens and closes the opening may be rotatable with respect to the case 111.

For example, the mobile robot 100 includes a roll-shaped main brush 154 having brushes exposed through the suction port 150h, and a plurality of radially extending wings located on the front side of the bottom of the main body 110. An auxiliary brush 155 having a brush made of may be provided. Dust is separated from the floor in the driving area by the rotation of these brushes 154 and 155, and the dust separated from the floor is sucked through the suction port 150h and can be introduced into the dust bin through the suction unit 330. .

Air and dust may be separated from each other while passing through the filter or cyclone of the dust bin, the separated dust is collected in the dust bin, and the air is discharged from the dust bin and then passes through the exhaust passage (not shown) inside the main body 110 and finally It can be discharged to the outside through an exhaust port (not shown).

For example, the battery 138 may supply power required for the overall operation of the mobile robot 100 as well as the drive motor. Meanwhile, when the battery 138 is discharged, the mobile robot 100 can return to the charging station 200 for charging, and during this return driving, the mobile robot 100 moves to the charging station 200 by itself. The location can be detected.

The charging base 200 may include, for example, a signal transmitting unit (not shown) that transmits a predetermined return signal. The return signal may be, for example, an ultrasonic signal or an infrared signal, but is not necessarily limited thereto.

The mobile robot 100 may include, for example, a signal detection unit (not shown) that receives a return signal.

For example, the signal detector may include an infrared sensor that detects an infrared signal, and may receive an infrared signal transmitted from the signal transmitter of the charging base 200. At this time, the mobile robot 100 may move to the location of the charging station 200 and dock with the charging station 200 according to the infrared signal transmitted from the charging station 200. Through this docking, the charging terminal 133 of the mobile robot 100 and the charging terminal 210 of the charging stand 200 can be brought into contact, and the battery 138 can be charged.

The mobile robot 100 may be equipped with a configuration to detect internal/external information of the mobile robot 100.

For example, the mobile robot 100 may be equipped with a camera 120 that acquires image information about the driving area.

For example, the mobile robot 100 may be equipped with a front camera 120a to acquire an image of the front of the main body 110.

For example, the mobile robot 100 may include an upper camera 120b that is provided on the upper surface of the main body 110 and acquires an image of the ceiling within the travel area.

For example, the mobile robot 100 may further include a lower camera 179 that is provided on the bottom of the main body 110 and acquires an image of the floor.

Meanwhile, the number, placement location, and shooting range of the cameras 120 provided in the mobile robot 100 are not necessarily limited thereto, and may be placed in various locations to obtain image information about the driving area. there is.

For example, the mobile robot 100 may include a camera (not shown) that is disposed at an angle on one side of the main body 110 and is configured to capture both the front and the top.

For example, the mobile robot 100 may be provided with a plurality of front cameras 120a and/or upper cameras 120b, and may be provided with a plurality of cameras configured to capture both the front and the top cameras.

According to various embodiments of the present invention, cameras 120 are installed in some parts (e.g., front, rear, bottom) of the mobile robot 100, and images can be continuously acquired while driving or cleaning. Several such cameras 120 may be installed in each area for filming efficiency, and the images captured by the cameras 120 are used to recognize the types of substances such as dust, hair, floor, etc. present in the space, whether or not they are cleaned, Alternatively, it can be used to check the timing of cleaning.

The mobile robot 100 may include a light detection and ranging (LiDAR) sensor 175 that acquires topographic information outside the main body 110 using a laser.

The LiDAR sensor 175 outputs a laser and receives the laser reflected from the object, thereby obtaining information such as the distance from the object that reflected the laser, location direction, and material, and obtaining topographic information of the driving area. can do. The mobile robot 100 can acquire 360-degree geometry information based on information acquired through the LiDAR sensor 175.

The mobile robot 100 may also include sensors 171, 172, and 179 that sense various data related to the operation and status of the mobile robot 100.

The mobile robot 100 may include an obstacle detection sensor 171 that detects an obstacle in front, a cliff detection sensor 172 that detects the presence of a cliff on the floor within the driving area, and the like.

The mobile robot 100 may include a manipulation unit 137 that can input various commands such as turning on/off the power of the mobile robot 100, and the mobile robot 100 may be operated through the manipulation unit 137. ) can receive various control commands necessary for the overall operation.

The mobile robot 100 may include an output unit (not shown) and may display reservation information, battery status, operation mode, operation status, error status, etc.

Meanwhile, Figure 5 is a diagram showing an example of a mobile robot according to another embodiment of the present invention.

The mobile robot 100 shown in FIG. 5 has the same or similar configurations as the mobile robot 100 shown in FIGS. 1 to 4, and functions as a mobile robot applicable to a large space. A detailed description of this will be omitted.

Figure 6 is a block diagram showing the control relationship between the main components of the mobile robot according to an embodiment of the present invention.

Referring to FIG. 6, the mobile robot 100 includes a storage unit 305, an image acquisition unit 320, an input unit 325, a suction unit 330, a control unit 350, a traveling unit 360, and a sensor unit ( 370), an output unit 380, and/or a communication unit 390.

The storage unit 305 can store various types of information necessary for controlling the mobile robot 100.

The storage unit 305 may include volatile or non-volatile recording media. A recording medium stores data that can be read by a microprocessor, and is not limited to its type or implementation method.

The storage unit 305 may store a map of the driving area. The map stored in the storage unit 305 may be input from an external terminal or server that can exchange information with the mobile robot 100 through wired or wireless communication, or the mobile robot 100 learns on its own and It may have been created.

The storage unit 305 can store data for the sub-area. Here, the sub-area may refer to a divided area having a certain distance or a certain area in the driving area. The data for the sub-area may include LiDAR detection data obtained while driving in the sub-area, information on each node for the corresponding LiDAR detection data, and information on the direction of movement in each node.

The storage unit 305 can store various map information.

The map may display the locations of rooms within the driving area. Additionally, the current location of the mobile robot 100 may be displayed on the map, and the current location of the mobile robot 100 on the map may be updated during the driving process.

The storage unit 305 may store cleaning history information. Such cleaning history information may be generated each time cleaning is performed.

The map for the driving area stored in the storage unit 305 includes, for example, a navigation map used for driving during cleaning, a simultaneous localization and mapping (SLAM) map used for location recognition, When an obstacle is encountered, the information is stored and a learning map used for learning and cleaning, a global topological map used for global location recognition, a grid map based on cell data, and information about recognized obstacles are stored. It may be a recorded obstacle recognition map, etc.

Meanwhile, maps can be stored and managed by classifying them in the storage unit 305 according to purpose, but the maps may not be clearly divided by purpose. For example, a plurality of information may be stored in one map so that it can be used for at least two purposes.

The image acquisition unit 320 may acquire images around the mobile robot 100. The image acquisition unit 320 may include at least one camera (eg, camera 120 of FIG. 1).

The image acquisition unit 320 may include, for example, a digital camera. A digital camera includes an image sensor (e.g., CMOS image sensor) that includes at least one optical lens and a plurality of photodiodes (e.g., pixels) that produce images by light passing through the optical lens. , may include a digital signal processor (DSP) that configures an image based on signals output from photodiodes. A digital signal processor, for example, is capable of generating not only still images but also moving images composed of frames composed of still images.

The image acquisition unit 320 may capture a situation of an obstacle or cleaning area in front of the mobile robot 100 in its traveling direction.

According to one embodiment of the present invention, the image acquisition unit 320 can acquire a plurality of images by continuously photographing the surroundings of the main body 110, and the plurality of acquired images can be stored in the storage unit 305. there is.

The mobile robot 100 can increase the accuracy of obstacle recognition by using a plurality of images, or by selecting one or more images from among the plurality of images and using effective data.

The input unit 325 may be provided with an input device (eg, key, touch panel, etc.) capable of receiving user input. The input unit 325 may include a manipulation unit 137 that can input various commands such as turning on/off the power of the mobile robot 100.

The input unit 325 can receive user input through an input device and transmit a command corresponding to the received user input to the control unit 350.

The suction unit 330 may suction air containing dust. The suction unit 330 includes, for example, a suction device (not shown) for sucking foreign substances, brushes 154 and 155 for scrubbing, and a suction device or brush (e.g., brushes 154 and 155 in FIG. 3). It may include a dust bin (not shown) for storing foreign substances collected by the dustbin, an intake port through which air is sucked (e.g., the intake port 150h in FIG. 4), etc.

The traveling unit 360 can move the mobile robot 100. The traveling unit 360 may include, for example, at least one driving wheel 136 that moves the mobile robot 100 and at least one motor (not shown) that rotates the driving wheel.

The sensor unit 370 may include a distance measurement sensor that measures the distance to an object outside the main body 110. The distance measurement sensor may include the lidar sensor 175 described above.

The mobile robot 100 according to an embodiment of the present invention can generate a map by determining the distance, location, and direction of objects sensed by the LiDAR sensor 175.

The mobile robot 100 according to an embodiment of the present invention can obtain topographic information of the driving area by analyzing the laser reception pattern, such as the time difference or signal strength of the laser reflected and received from the outside. Additionally, the mobile robot 100 may generate a map using terrain information acquired through the LiDAR sensor 175.

For example, the mobile robot 100 according to the present invention may perform a LiDAR slam to determine the direction of movement by analyzing surrounding terrain information acquired at the current location through the LiDAR sensor 175.

More preferably, the mobile robot 100 according to the present invention effectively recognizes obstacles through vision-based location recognition using a camera, lidar-based location recognition technology using a laser, and an ultrasonic sensor, and provides optimal movement with a small amount of change. Map creation can be performed by extracting directions.

The sensor unit 370 may include an obstacle detection sensor 171 that detects an obstacle in front, a cliff detection sensor 172 that detects the presence of a cliff on the floor within the driving area, and the like.

A plurality of obstacle detection sensors 171 may be arranged at regular intervals on the outer peripheral surface of the mobile robot 100. The obstacle detection sensor 171 may include an infrared sensor, an ultrasonic sensor, a radio frequency (RF) sensor, a geomagnetic sensor, a position sensitive device (PSD) sensor, etc.

The obstacle detection sensor 171 may be a sensor that detects the distance to an indoor wall or obstacle, and the present invention is not limited to its type, but will be described below by taking an infrared sensor as an example.

The obstacle detection sensor 171 can detect objects, especially obstacles, present in the driving (movement) direction of the mobile robot 100 and transmit obstacle information to the control unit 350. That is, the obstacle detection sensor 171 detects the movement path of the mobile robot 100, protrusions present in front or on the side of the mobile robot 100, household fixtures, furniture, walls, wall corners, etc., and provides the information. It can be transmitted to the control unit 350.

The sensor unit 370 may further include a driving detection sensor (not shown) that detects the driving motion of the mobile robot 100 and outputs motion information. The driving sensor may include, for example, a gyro sensor, a wheel sensor, an acceleration sensor, etc.

The gyro sensor can detect the rotation direction and rotation angle when the mobile robot 100 moves according to the driving mode. The gyro sensor can detect the angular velocity of the mobile robot 100 and output a voltage value proportional to the angular velocity.

The wheel sensor is connected to the driving wheel 136, for example, the left wheel 136 (L) and the right wheel 136 (R) of FIG. 4 and can detect the rotation speed of the driving wheel 136.

The acceleration sensor can detect changes in speed of the mobile robot 100. The acceleration sensor may be attached to a position adjacent to the driving wheel 136 or may be built into the control unit 350.

The output unit 380 may include an audio output unit 381 that outputs an audio signal. Under the control of the control unit 350, the sound output unit outputs alarm messages such as warning sounds, operation modes, operation states, and error states, information corresponding to the user's command input, and processing results corresponding to the user's command input as sound. You can.

The audio output unit 381 can convert the electrical signal from the control unit 150 into an audio signal and output it. For this purpose, speakers, etc. may be provided.

The output unit 380 may include a display 382 that displays information corresponding to the user's command input, processing results corresponding to the user's command input, operation mode, operation state, error state, etc. as images.

Depending on the embodiment, the display 382 may be configured as a touch screen by forming a layer structure with the touch pad. In this case, the display 382, which is configured as a touch screen, can be used as an input device capable of inputting information by a user's touch in addition to an output device.

The communication unit 390 may include at least one communication module (not shown) and may transmit and receive data with an external device. Among the external devices that communicate with the mobile robot 100, an external terminal includes, for example, an application for controlling the mobile robot 100, and maps the driving area to be cleaned by the mobile robot 100 through execution of the application. , and you can designate an area to clean a specific area on the map.

The communication unit 390 can transmit and receive signals through wireless communication methods such as Wi-Fi, Bluetooth, beacon, zigbee, and radio frequency identification (RFID).

The power supply unit may supply driving power and operating power to each component of the mobile robot 100.

The mobile robot 100 may detect the remaining battery capacity and charging state of the battery 138 and may further include a battery detection unit (not shown) that transmits the detection result to the control unit 350.

The control unit 350 may be connected to each component provided in the mobile robot 100. For example, the control unit 350 can transmit and receive signals between each component provided in the mobile robot 100 and control the overall operation of each component.

The control unit 350 may determine the internal/external status of the mobile robot 100 based on information acquired through the sensor unit 370.

The control unit 350 can calculate the rotation direction and angle using the voltage value output from the gyro sensor.

The control unit 350 may calculate the rotational speed of the driving wheel 136 based on the rotational speed output from the wheel sensor. Additionally, the control unit 350 may calculate the rotation angle based on the difference in rotation speed between the left wheel 136(L) and the right wheel 136(R).

The control unit 350 may determine changes in the state of the mobile robot 100, such as starting, stopping, changing direction, or colliding with an object, based on the value output from the acceleration sensor. Meanwhile, the control unit 350 can detect the amount of impact due to a change in speed based on the value output from the acceleration sensor, so the acceleration sensor can also perform the function of an electronic bumper sensor.

The control unit 350 may detect the location of an obstacle based on at least two signals received through an infrared sensor and control the movement of the mobile robot 100 according to the location of the detected obstacle.

Depending on the embodiment, the obstacle detection sensor 171 provided on the outer surface of the mobile robot 100 may be configured to include a transmitter and a receiver.

For example, an infrared sensor may be provided with at least one transmitting unit and at least two receiving units staggered from each other. Accordingly, the transmitter can radiate infrared signals at various angles, and at least two receivers can receive the infrared signals reflected by obstacles at various angles.

Depending on the embodiment, the signal received from the infrared sensor may undergo signal processing such as amplification and filtering, and then the distance and direction to the obstacle may be calculated.

Meanwhile, the control unit 350 may include a driving control module 351, a map generation module 352, a location recognition module 353, and/or an obstacle recognition module 354. In this drawing, for convenience of explanation, the driving control module 351, map generation module 352, location recognition module 353, and/or obstacle recognition module 354 are separately described, but the present invention is not limited thereto. no.

The location recognition module 353 and the obstacle recognition module 354 can be integrated into one recognizer to form one recognition module 355. In this case, a recognizer is trained using a learning technique such as machine learning, and the learned recognizer can later classify input data to recognize properties of areas, objects, etc.

Depending on the embodiment, the map generation module 352, the location recognition module 353, and the obstacle recognition module 354 may be configured as one integrated module.

The travel control module 351 can control the travel of the mobile robot 100 and control the driving of the travel unit 360 according to travel settings.

The travel control module 351 can determine the travel path of the mobile robot 100 based on the operation of the travel unit 360. The travel control module 351 can determine the current or past movement speed and traveled distance of the mobile robot 100 based on the rotation speed of the driving wheel 136, and the travel of the mobile robot 100 determined in this way. Based on the information, the location of the mobile robot 100 on the map may be updated.

The map generation module 352 can generate a map for the driving area.

The map generation module 352 can generate and/or update a map in real time based on information acquired while the mobile robot 100 is traveling.

The map generation module 352 can set multiple movement directions. For example, when a function for generating a map of a driving area (hereinafter referred to as map generation function) is executed, the map generation module 352 determines the direction in which the front of the mobile robot 100 faces at the time the function is executed. It can be set as the first movement direction. In addition, at the time the function is executed, the map generation module 352 uses the direction in which the left side of the mobile robot 100 faces as the second movement direction, and the direction in which the right side of the mobile robot 100 faces as the third movement direction and the first direction in which the map generation module 352 faces. The direction in which the rear of the mobile robot 100 faces, which is the opposite direction, can be set as the fourth movement direction.

Meanwhile, in this drawing, the plurality of movement directions are described as being set to four directions, but the present invention is not limited to this, and according to various embodiments, they may be set to various directions such as 8 or 16. .

The map creation module 352 can create a map based on information acquired through the LiDAR sensor 175.

The map generation module 352 is output through the LiDAR sensor 175 and analyzes reception patterns such as reception time difference and signal strength of the laser reflected from external objects and can obtain topographic information of the driving area. . Topographic information of the driving area may include, for example, the location, distance, and direction of objects existing around the mobile robot 100.

The map generation module 352 can store information about a plurality of nodes while generating a global topological map based on the topographic information of the driving area acquired through the LiDAR sensor 175. Information may be defined as first map data.

Additionally, the map generation module 352 may generate a grid map based on cell data based on a distance detection signal from an infrared sensor, which is the obstacle detection sensor 171.

The map generation module 352 divides the driving area into a plurality of sub-areas and generates a grid map with different cell data for areas where obstacles exist and areas where obstacles do not exist through an infrared sensor while driving in each sub-area. can do.

The location recognition module 353 can determine the location of the mobile robot 100. The location recognition module 353 can determine the location of the mobile robot 100 while the mobile robot 100 is traveling.

The location recognition module 353 may determine the location of the mobile robot 100 based on the acquired image obtained through the image acquisition unit 320.

For example, while the mobile robot 100 is driving, the location recognition module 353 uses the characteristics of each location in the driving area detected from the acquired image based on the map data generated by the map generation module 352. Each location can be mapped, and data on the characteristics of each location in the driving area mapped to each location on the map can be stored in the storage unit 305 as location recognition data.

Meanwhile, the location recognition module 353 compares the characteristics of the driving area detected from the acquired image with the characteristics of each location of the driving area included in the location recognition data stored in the storage unit 305, The similarity (probability) for each location can be calculated, and based on the calculated similarity (probability) for each location, the location with the greatest similarity can be determined as the location of the mobile robot 100.

According to one embodiment of the present invention, the mobile robot 100 extracts features from the image acquired through the image acquisition unit 320, and substitutes the extracted features into the mapped grid map to create a mobile robot ( 100) can be determined.

Meanwhile, the mobile robot 100 learns the map through the driving control module 351, map generation module 352, and/or obstacle recognition module 354 without the location recognition module 353 to determine the current location. It may be possible.

The obstacle recognition module 354 can detect obstacles around the mobile robot 100. For example, the obstacle recognition module 354 detects obstacles around the mobile robot 100 based on the acquired image acquired through the image acquisition unit 320 and/or the sensing data acquired through the sensor unit 370. can be detected.

For example, the obstacle recognition module 354 may detect obstacles around the mobile robot 100 based on topographic information of the driving area obtained through an infrared sensor.

The obstacle recognition module 354 can determine whether there is an obstacle that interferes with the movement of the mobile robot 100 while the mobile robot 100 is traveling.

When it is determined that an obstacle exists, the obstacle recognition module 354 can determine a driving pattern such as going straight or turning according to the attributes of the obstacle, and can transmit the determined driving pattern to the driving control module 351.

The mobile robot 100 according to an embodiment of the present invention is capable of recognizing and avoiding people and objects based on machine learning. Here, machine learning can mean that a computer learns through data without a person directly instructing the computer to use logic, and through this, the computer solves problems on its own.

Deep Learning is a method of teaching computers how to think like humans, based on artificial neural networks (ANN) to construct artificial intelligence. It can mean intelligent technology. Artificial neural networks (ANNs) can be implemented in software form or in hardware form such as chips.

The obstacle recognition module 354 may include an artificial neural network (ANN) in the form of software or hardware that learns the properties of obstacles.

For example, the obstacle recognition module 354 may include a deep neural network (DNN) such as a convolutional neural network (CNN), a recurrent neural network (RNN), and a deep belief network (DBN) learned through deep learning. there is.

For example, the obstacle recognition module 354 may determine the properties of an obstacle included in input image data based on weights between nodes included in a deep neural network (DNN).

Additionally, the travel control module 351 may control the driving of the travel unit 360 based on the attributes of the recognized obstacle.

The storage unit 305 may store input data for determining obstacle properties and data for learning the deep neural network (DNN). The storage unit 305 may store the original image acquired by the image acquisition unit 320 and the extracted images from which a predetermined area is extracted. Weights and biases forming a deep neural network (DNN) structure may be stored in the storage unit 305. For example, weights and biases that make up the deep neural network structure may be stored in the embedded memory of the obstacle recognition module 354.

For example, whenever the obstacle recognition module 354 extracts a partial area of the image acquired by the image acquisition unit 320, it performs a learning process using the extracted image as training data, or performs a predetermined The learning process can be performed after more than a number of extracted images are acquired.

In other words, the obstacle recognition module 354 updates the deep neural network (DNN) structure, such as weights, by adding recognition results every time it recognizes an obstacle, or trains secured after a predetermined number of training data are secured. By performing a learning process with data, you can update the deep neural network (DNN) structure, including weights.

Alternatively, the mobile robot 100 may transmit the original image or extracted image acquired by the image acquisition unit 320 to a predetermined server through the communication unit 390 and receive data related to machine learning from the predetermined server. At this time, the mobile robot 100 may update the obstacle recognition module 354 based on data related to machine learning received from a certain server.

Figure 7 is a diagram referenced in the description of a lidar sensor provided in a mobile robot according to an embodiment of the present invention.

Referring to FIG. 7, the LiDAR sensor 175 can output a laser in all directions of 360 degrees, and by receiving the laser reflected from the object, the distance to the object that reflected the laser, position direction, material, etc. Information can be obtained, and topographical information of the driving area can be obtained.

The mobile robot 100 can acquire terrain information within a certain distance according to the performance and settings of the lidar sensor 175. The mobile robot 100 may acquire terrain information within a circular area 610 with a radius of a certain distance 610 based on the LiDAR sensor 175.

The mobile robot 100 obtains terrain information within the original area 610, and at this time, it can acquire terrain information for a predetermined sub-area.

At this time, the terrain information of the LiDAR sensor 175 may store the terrain information within the circle area 610 detected at each point where the mobile robot 100 travels a predetermined distance or a predetermined time. At this time, the information stored may be stored by matching the traveling nodes (n1, n2...), which are the current location of the mobile robot 100, which is the center of the corresponding circular area 610.

Therefore, when the mobile robot 100 travels in each divided sub-area, topographical information of each original area 610 is obtained for a plurality of consecutive travel nodes (n1, n2...) within each sub-area. and saved.

The mobile robot 100 can extract open space from topographic information about the sub-area obtained through the lidar sensor 175. Here, open space may refer to information about the space between objects that reflect the laser through which the mobile robot 100 can travel.

After driving in the allocated sub-area ends, the mobile robot 100 continuously reads the detection data of the lidar sensor 175 for each driving node (n1, n2...) within the sub-area (ray casting). You can perform an open direction detection algorithm that checks whether there is an open direction by performing casting).

Information about the open space may include information about the traveling nodes (n1, n2...) and traveling direction of the mobile robot 100 searching for the open space, and the width of the open space. The open node may include information about the width of the open space, and the open node may include information about the width of the open space. It may include information about the open space and the traveling nodes (n1, n2...) of the mobile robot 100.

After detecting all open directions for each traveling node in the sub-area, the mobile robot 100 determines whether an unexplored area exists in the open direction for each traveling node based on the grid map of each sub-area. , If there is no unexplored area, the corresponding driving node is closed.

The mobile robot 100 can generate a topology map when all the openings of the traveling nodes (n1, n2...) of the mobile robot 100 are closed for one sub-area.

The mobile robot 100 searches for the existence of an unexplored area through the grid map again in the sub-area where the topology map was created. If an unsearched area exists within the corresponding sub-area, the mobile robot 100 moves to a traveling node adjacent to the unsearched area, searches the unsearched area, and then updates the grid map and topology map with the corresponding information.

In this way, for one sub-area, a topology map is formed by checking whether there is an unexplored area based on information on the lidar sensor 175, and the corresponding sub-area is additionally determined by the grid map on the obstacle detection sensor 171. By re-checking whether there is an unsearched area in the sub-area, mapping of the sub-area is possible without an unsearched area.

Hereinafter, the process of generating a map for each sub-area of the mobile robot 100 of the present invention will be described with reference to FIGS. 8 and 9.

FIG. 8 is a flow chart of a map creation process for a mobile robot according to an embodiment of the present invention, and FIGS. 9A to 9E are diagrams showing a map creation process for a mobile robot according to the flowchart of FIG. 8 .

Referring to Figures 8 and 9, the mobile robot 100 can generate a topology map and a grid map while moving.

Specifically, as shown in FIG. 8, the mobile robot 100 may divide the driving area into a plurality of sub-areas.

At this time, the sub-area to be divided may be divided based on the driving distance or driving area, and may be an arbitrary area for forming a topology map of the moved area after moving a certain distance or for a certain time.

In other words, when the mobile robot 100 drives an unknown driving area for cleaning for a predetermined distance or a certain time, rather than an equally divided area, the area traveled for that distance or time can be defined as one sub-area. there is.

At this time, the sub-area may be defined from the current location of the mobile robot 100 until an obstacle is discovered in front, and can be set in various ways depending on the user's settings.

When the map creation function is executed (S100), the mobile robot 100 drives a line to map an unknown driving area for cleaning work (S101) and uses the lidar of the divided sub-area through the lidar sensor 175. A topology map can be formed by collecting information, and a grid map can be created by collecting ultrasonic sensor or infrared sensor information through the obstacle detection sensor 171.

At this time, ray casting at the driving node to generate the topology map checks whether each driving node is an open node, and rechecks whether an unexplored area exists in the grid map of the corresponding area to update the topology map.

The grid map is an OGM (Occupancy grid map), which is a map utilizing a grid generated by obtained regional information.

The grid map is a map for recognizing the surrounding environment of the mobile robot 100. It can refer to a map in which the sub-area (A) is expressed as a grid or cells (hereinafter referred to as cells) of the same size and the presence or absence of an object is indicated in each cell. there is. The presence or absence of an object can be indicated through color. For example, a white cell may represent an area without an object, and a gray cell may represent an area with an object. Therefore, the line connecting the gray cells can represent the border of any space (wall, obstacle, etc.). The color of a cell can change through image processing.

At this time, the control unit 350 can generate a grid map while driving in an unknown sub-area. In contrast, the control unit 350 may generate a grid map of the corresponding sub-area (A) after driving in the sub-area is completed.

In the present invention, the grid map creation step (S102) can be set to generate a grid map while driving in a sub-area.

The mobile robot 100 collects distance information and LiDAR information of the area through the obstacle detection sensor 171 and the LiDAR sensor 175 while traveling in the sub-area (S102), and collects the collected LiDAR information. A step of searching for open nodes in each driving node while generating a topology map through (S103), a step of re-confirming the openness of each open node (S104), and if there are no more open nodes in the sub-area (S105) ), extracting an unexplored area through the grid map of the sub-area (S107), moving the mobile robot to the unexplored area (S108), and creating a new open node by searching the unexplored area to create a topology map. The process proceeds to the updating step (S109).

First, the control unit 350 of the mobile robot 100 performs line driving to generate a cleaning map for a predetermined driving area according to an external or set instruction (S100) (S101).

When driving in this way, the point where driving is completed for a predetermined distance or a certain time can be defined as a sub-area, and driving continues in the same direction (d1) until driving to the corresponding sub-area is completed.

At this time, the mobile robot 100 can obtain LiDAR information about the 360-degree circular area 610 from the LiDAR sensor 175 and obtain distance information from the obstacle detection sensor 171, as shown in FIG. 9A. (S102). The collected LIDAR information may be collected from each of a plurality of traveling nodes (n1, n2...) located on the path when the mobile robot 100 travels along a path within the sub-area. At this time, when driving along the passage, center following can be performed, moving along the center point of the passage.

Travel nodes (n1, n2...) may be set at equal intervals on the route, but are not limited to this, and may be defined as points at which the mobile robot 100 is located every predetermined time. The virtual line connecting such driving nodes (n1, n2...) may function as a driving path (DP), but is not limited to this.

Travel nodes (n1, n2...) set in one sub-area may have different numbers.

The mobile robot 100 collects LIDAR information at each traveling node (n1, n2...).

Therefore, at each driving node (n1, n2...), local information about the circle area 610 centered on the corresponding driving node (n1, n2...) is collected as LIDAR information.

Each of the LIDAR information may be recorded in the storage unit 305 by matching each driving node (n1, n2...).

Such LIDAR information may include information such as the distance, position direction, and material of the object that outputs the laser and reflects the laser within the circle area 610.

Additionally, the mobile robot 100 may obtain distance information to an obstacle located ahead from ultrasonic or infrared information from the obstacle detection sensor 171 while continuously traveling.

The mobile robot 100 may collect distance information such as lidar information and infrared information for each traveling node (n1, n2...) until it completes traveling in the sub-area (A). Once driving is completed for the sub-area, a topology node map can be created using the collected information. At this time, as shown in FIG. 9A, the generated topology node map can be completed by indicating the open and closed for consecutive traveling nodes (n1, n2...), and the topology node is formed by connecting such traveling nodes. The map is complete.

Specifically, the mobile robot 100 ray casts LiDAR information for each traveling node (n1, n2...) and performs an open space detection algorithm to check whether there is an open space.

At this time, the open space is created at the corresponding driving node (n1, n2...) by ray casting (RS) the lidar information on the original area 610 obtained for each driving node (n1, n2...) in the entire section. , that is, determine whether there is an open space where there are no obstacles.

At this time, the presence or absence of an open space may be defined as a case where the distance between the reflected and obtained LIDAR data is greater than the width of the mobile robot 100, but is not limited to this and has a numerical value for the reference width, If there is a space with a width wider than the standard width, it can be set as an open space.

In this way, by performing ray casting, that is, ray projection, of LiDAR information on the 360-degree circular area 610 for each traveling node (n1, n2...), rather than only confirming LiDAR information for a specific direction, the entire It is possible to check the presence or absence of open space in any direction.

This means that when calculating data by subdividing it into a specific direction, for example, 4 or 8 directions, the search for open space that may occur in an unfiltered area may be missed. In particular, if the direction of arrangement of the driving area for cleaning and the direction of the main passage do not match the driving direction of the mobile robot 100, there is a risk that a large number of open spaces may be missed in search.

Therefore, as in the present invention, after dividing into a predetermined sub-area, 360-degree ray casting is performed on the lidar information received at each of the plurality of traveling nodes (n1, n2...) on the route while driving in the sub-area to ensure that no direction is missed. Exploration of open space is possible.

If an open space exists through ray casting for each traveling node (n1, n2…), the corresponding traveling node (n1, n2…) is changed to an open node, and the open direction (OD) for the direction in which the open space is located is changed. ) is displayed.

In this way, when ray casting is completed for all driving nodes (n1, n2...) in the corresponding sub-area and a plurality of open spaces are searched, the information of each open space, that is, the location of the open space, the width, and the open space Information about the driving nodes (n1, n2...) corresponding to can be stored, respectively.

At this time, the control unit 350 can apply various algorithms to determine open space and set open nodes accordingly.

Next, the control unit 350 confirms the opening of the open node (converted to black), as shown in FIG. 9B (S104).

Specifically, the control unit 350 divides the area where the open node is located in the sub-area for each open node, and determines whether the area facing the open direction in each open node faces the open direction of another open node.

At this time, the division of each open node and the decision on the open direction are searched based on the grid map. Specifically, in order to check whether there is an unexplored grid map from the open node to the open direction, a local map of the corresponding section is created by dividing the section in the open direction. The demarcated area is set to be longer than the lidar sensor detection distance in the open direction, set to the left/right as the error range of the open node, and set as short as possible to the rear. Gradually expand from the open node starting point and check whether unexplored areas are found. The criterion for the presence or absence of an unexplored area is defined as an area where an unknown point and an explored point meet in an area equal to the size of the mobile robot. If no unexplored area is found, the open node is closed. Therefore, if there are open directions facing each other in the area defined for each open node, the corresponding open node is closed.

In this way, the grid map is read from each open node, and through a search in the opposite direction from the open direction of the open node, it is confirmed that there is no unsearched area and the open node is closed.

For example, if it is confirmed that the first traveling node (n1) and the ninth traveling node (n9) have open directions (OD) facing each other in the first section of FIG. 9B, the first traveling node (n1) and the ninth traveling node (n9) 9 Remove the open node setting of the driving node (n9) and close the open node.

In this determination, it is checked whether there is an open node remaining in the sub-area (S105), and if an open node exists, it is registered and stored (S106).

Next, if there are no more open nodes to check for openness, a secondary search is performed.

The secondary search is performed based on the grid map, and the control unit 350 reads information about the sub-area from the grid map as shown in FIG. 9C and determines whether there are unsearched areas in the sub-area other than closed open nodes. Judge (S107).

The control unit 350 updates the open node information of the topology map of the sub-area to the grid map of the sub-area and extracts unsearched segments.

As shown in FIG. 9C, when the unsearched area (A) is searched, the control unit 350 selects the traveling node closest to the searched unsearched area (A) among the traveling nodes (n1, n2,...) through an algorithm such as wavefront. Search for (n5).

As shown in Figure 9d, the mobile robot moves to the traveling node (n5) closest to the unsearched area (A) (S108), obtains sensor information from the remaining direction other than the existing robot approach direction in the unsearched area (A), and , As shown in Figure 9e, new open nodes (n10, n11) are created based on the grid map within the unsearched area (A) (S109).

At this time, when new open nodes (n10, n11) are created, sub-area division and search driving for generating a topology map start again from the new open nodes (n10, n11).

In this way, when determining whether an open node is open, the computational load can be improved by drastically reducing the redundant search for data judgment by updating based on the grid map. In this way, by minimizing the number of open nodes and additionally checking whether there is an unexplored area based on the grid map, additional exploration of the mobile robot 100 can be dramatically reduced.

Meanwhile, referring to FIG. 10, the mapping method of the present invention can be applied even when manual mapping is added.

Referring to FIG. 10, first, the control unit 350 of the mobile robot 100 performs line driving to generate a cleaning map for a predetermined driving area according to external or set instructions, which is the same as FIG. 8. do.

At this time, the mobile robot 100 obtains LiDAR information about the 360-degree circular area 610 from the LiDAR sensor 175, obtains distance information from the obstacle detection sensor 171, and obtains distance information from the obstacle detection sensor 171. Distance information to an obstacle located ahead can be obtained from ultrasonic or infrared information from the device (S200).

The mobile robot 100 may collect distance information such as lidar information and infrared information for each traveling node (n1, n2...) until it completes traveling in the sub-area (A). When driving in the sub-area (A) is completed, a topology node map can be created using the collected information. At this time, the generated topology node map can be completed by indicating open and closed for consecutive traveling nodes (n1, n2...), and the topology node map is completed by connecting such traveling nodes.

In this way, when a manual mapping instruction is received from the user during the search drive for mapping (S201), the mobile robot 100 moves to the indicated area and performs the search drive for mapping in the area.

At this time, when the user performs manual mapping through the application, like automatic mapping, detection information can be obtained from the lidar sensor 175 and the obstacle detection sensor 171 while moving according to the user's instructions, Figure 9 As shown, the opening or closing of the driving node can be determined according to the relevant information.

In this way, while driving manually, information for the topology map and grid map is received and processed to perform mapping (S202).

Meanwhile, when manual mapping is terminated in the area where manual mapping was performed (S203), the mobile robot 100 restarts automatic mapping (S204).

When automatic mapping starts again, the mobile robot 100 determines whether an open node exists among the topology driving nodes created at the corresponding location.

That is, as shown in FIG. 8, the open space is confirmed from the open space detection algorithm, and if an open space exists through ray casting for the driving node (n1, n2...), the driving node (n1, n2...) is designated as an open node. Change it and display the open direction (OD) for the direction in which the open space is located.

Next, the control unit 350 confirms the opening of each open node (S206).

That is, the control unit 350 divides the area where the open node is located in the sub-area for each open node, and determines whether the area facing the open direction in each open node faces the open direction of another open node.

At this time, the division of each open node and the decision on the open direction are searched based on the grid map. Specifically, in order to check whether there is any unexplored area in the grid map from the open node to the open direction, a local map of the partitioned area is created by partitioning in the open direction. Therefore, if there are open directions facing each other in the area defined for each open node, the corresponding open node is closed.

In this way, the grid map is read from each open node, and through a search in the opposite direction from the open direction of the open node, it is confirmed that there is no unsearched area and it is determined whether the open node is open or closed. In this judgment, it is checked whether there are any open nodes remaining in the manually mapped area, and if an open node exists, it is registered and stored.

Also, move to the corresponding open node and rotate to check additional driving nodes and other open nodes (S207).

Meanwhile, if there are no more open nodes within the manual mapping area to check for openness, a secondary search is performed.

The secondary search is performed based on the grid map, and the control unit 350 reads information about the manual mapping area from the grid map as shown in FIG. 9C and determines whether there is an unsearched area other than closed open nodes in the manual mapping area. Determine whether or not (S208).

The control unit extracts unexplored segments while updating the open node information of the topology map of the manual mapping area to the grid map of the manual mapping area.

When an unexplored area is discovered, the control unit searches for the driving node closest to the discovered unsearched area among driving nodes through an algorithm such as wavefront, moves to the driving node closest to the unsearched area, and moves the existing robot in the unsearched area. Sensor information is obtained from directions other than the approach direction, and a new open node is created based on the grid map within the unexplored area, as shown in Figure 9e.

At this time, when a new open node is created, sub-area division and search driving for creating a topology map starts again from the new open node.

In this way, when manual mapping from the user is performed during automatic mapping, open nodes are checked for the area where the manual mapping was performed, and the search is performed twice to see if there is an unsearched area, thereby eliminating duplicate search due to disconnection of the mapping area. can be prevented.

The control unit 350 can form a topology graph for each of these open nodes, and can complete a topology graph for the entire driving area by connecting the topology graphs for a plurality of sub-areas.

This topology graph can function as a basic map to form a cleaning map.

The control unit 350 can check the relationship between each node by referring to the grid map and topology graph and set the optimal path.

Additionally, the control unit 350 can fuse image data into the corresponding grid map and create an image.

In such imaging, the shading of each grid map can be displayed as a line by performing border processing algorithms and image processing through brightness differences.

The mobile robot 100 can set an optimal path based on the grid map and topology graph imaged at the current location of the mobile robot 100.

This mapping method has the following effects.

Figure 11a shows the process without checking whether the open node is open in the topology map, and Figure 11b shows the present invention and a search drive in an unsearched area according to the grid map after checking whether the open node is open.

Each node represents an additional navigation point for the mobile robot.

As shown in Figure 11a, if the open node does not re-check whether each open node is open, it enters each open node and performs search driving in that open node again. Therefore, the additional search of the mobile robot is very frequent and the search of open nodes is redundant.

On the other hand, as shown in Figure 11b, each driving node defined as an open node in the topology map is divided on the grid map to recheck whether the open node is open, and by closing some open nodes, the number of open nodes is significantly reduced in the first instance. You can do it. In addition, by simultaneously checking the unsearched area on the secondary grid map, it is possible to check whether there are missing unsearched areas due to the reduced open nodes, thereby improving reliability.

The mobile robot 100 according to the present invention is not limited to the configuration and method of the embodiments described above, but all or part of each embodiment is optional so that various modifications can be made. It may be composed of a combination of .

Likewise, although operations are depicted in the drawings in a particular order, this should not be construed to mean that those operations must be performed in the specific order or sequential order shown or that all of the depicted operations must be performed to obtain desirable results. . In certain cases, multitasking and parallel processing may be advantageous.

Meanwhile, the control method of the mobile robot 100 according to an embodiment of the present invention can be implemented as processor-readable code on a processor-readable recording medium. Processor-readable recording media includes all types of recording devices that store data that can be read by a processor. It also includes those implemented in the form of carrier waves, such as transmission through the Internet. Additionally, the processor-readable recording medium is distributed in a computer system connected to a network, so that the processor-readable code can be stored and executed in a distributed manner.

In addition, although preferred embodiments of the present invention have been shown and described above, the present invention is not limited to the specific embodiments described above, and the technical field to which the invention pertains without departing from the gist of the present invention as claimed in the claims. Of course, various modifications can be made by those skilled in the art, and these modifications should not be understood individually from the technical idea or perspective of the present invention.

Mobile robot: 100
Body: 110
Storage: 305
Video acquisition department: 320
Input: 325
Suction unit: 330
Control: 350
Running part: 360
Sensor unit: 370

Claims (12)

a traveling unit that moves the main body within the traveling area;
a detection unit that acquires distance detection information regarding the distance to an object outside the main body; and
Generate a grid map for the driving area from the distance detection information, search for open nodes for a plurality of driving nodes on the driving path from the LiDAR information among the distance detection information, and select each open node from the grid map. A control unit that re-determines whether to open a device and generates a topology map;
Mobile robot including.
According to claim 1,
The sensor unit is a mobile robot, characterized in that it includes a lidar sensor that irradiates a laser to an object outside the main body and calculates the distance detection information by reflected light.
According to claim 2,
A mobile robot, wherein the control unit searches for open space by executing the ray casting for each of a plurality of traveling nodes, and sets the traveling node where the open space exists as the open node.
According to claim 3,
The control unit closes the open node when the open node has an open space facing the direction in which the open space of the open node exists based on the grid map.
According to claim 3,
The control unit, for each open node, divides the area around the open node from the grid map, determines whether an unexplored area exists around the open node, and determines whether to close the open node. robot.
According to claim 5,
The control unit divides the driving area into a plurality of sub-areas and generates the grid map and the topology map for each sub-area,
The sub-area is a mobile robot characterized in that the mobile robot is divided into an area when the mobile robot travels the travel area for a predetermined time or a predetermined distance.
According to claim 5,
The control unit determines whether an unsearched area for the sub-area exists from the grid map when it is re-determined whether the open node is open for the sub-area.
According to clause 6 or 7,
The mobile robot is characterized in that, when the unsearched area exists, the control unit moves to the travel node around the unsearched area and performs a search drive to see if an open node exists in the unsearched area.
According to claim 8,
The mobile robot is characterized in that the control unit simultaneously performs node-based search and the grid map-based search while traveling along the travel node.
According to claim 3,
The control unit determines that the mobile robot can travel by ray casting from the travel node and determines a space in which the mobile robot has not previously driven as the open space.
According to claim 10,
The mobile robot is characterized in that the control unit performs 360-degree ray casting around the traveling node.
According to claim 1,
A mobile robot, wherein the sensor unit further includes an obstacle detection sensor for generating the grid map.