KR102572841B1 - Mobile robot, customized autonomous driving method and program for each space size of mobile robot based on artificial intelligence - Google Patents

Abstract

The present disclosure provides an artificial intelligence-based autonomous driving method customized for each space size of a mobile robot, comprising: generating a global route to a destination by the mobile robot; determining whether the mobile robot is in a first space larger than a preset space size or a second space smaller than a preset space size according to a global route to a destination; If it is a first space, generating a first avoidance path for avoiding and traveling according to the approach state of the obstacle when the approach time and approach distance to the obstacle are set in a state in which the mobile robot detects the obstacle; and generating a second avoidance path for traveling by immediately avoiding the obstacle when the mobile robot detects the obstacle if it is the second space.

Description

Mobile robot, autonomous driving method and program customized for each space size of artificial intelligence-based mobile robot

The present disclosure relates to mobile robots. More specifically, the present disclosure relates to a mobile robot and a method and program for autonomous navigation customized for each space size of an artificial intelligence-based mobile robot.

In the era of the 4th industrial revolution, the introduction of service robots in various spaces such as hospitals, airports, and malls is increasing.

In large spaces such as hospitals, airports, and malls, there are too many dynamic obstacles for mobile robots to autonomously drive to their destinations.

It is easy to encounter other dynamic obstacles when detecting these dynamic obstacles far from the environment and avoiding them by speeding up the creation of an avoidance path.

Therefore, it is necessary to study mobile robots for efficiently avoiding obstacles and traveling quickly by quickly determining the destination by space.

Republic of Korea Patent Publication No. 10-2017-0094103

An object of the embodiments disclosed in the present disclosure is to provide a device capable of efficiently avoiding obstacles and quickly traveling to a destination.

The problems to be solved by the present disclosure are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

According to an aspect of the present disclosure for achieving the above-described technical problem, a method for autonomous driving customized for each space size of an artificial intelligence-based mobile robot is provided, wherein the mobile robot creating a global route to the destination; determining whether the mobile robot is in a first space larger than a preset space size or a second space smaller than the preset space size according to the global route to the destination; If it is the first space, generating a first avoidance path for avoiding and traveling according to the approach state of the obstacle when the approach time and approach distance to the obstacle are preset in a state in which the mobile robot detects the obstacle ; and generating a second avoidance path for traveling by immediately avoiding the obstacle when the mobile robot detects the obstacle if it is the second space.

In addition, in the determining step, a separation distance between a stationary object of map information preset in the mobile robot and the global route is calculated, and if the separation distance is within a preset reference value, the separation distance is determined as the second space, and the separation distance When the value exceeds the reference value, it may be characterized in that it is determined as the first space.

In addition, in the generating of the first avoidance path, the mobile robot stops and waits for the driving until the approach time and the approach distance are reached, and then the first avoidance path is reached when the approach time and the approach distance are reached. It can be characterized by generating.

In addition, in the generating of the second avoidance path, a virtual obstacle line is generated as soon as the mobile robot detects the obstacle, and the second avoidance path is generated so that the mobile robot does not come into contact with the virtual obstacle line. can be done with

In addition, a mobile robot according to another aspect of the present disclosure includes a sensor unit for detecting an obstacle; Memory storing global route and map information by destination; and a processor controlling an operation for driving while avoiding an obstacle for each destination, wherein the processor generates a global path to the destination, and a first space larger than a predetermined space size according to the global path to the destination. Alternatively, it is determined whether the second space is smaller than the preset space size, and if the first space is the first space, avoidance according to the approach state of the obstacle when the approach time and approach distance to the preset obstacle are reached in a state in which the obstacle is detected. and generating a first avoidance path for driving, and generating a second avoidance path for driving by immediately avoiding the obstacle at the moment when the obstacle is detected in the second space.

In addition, the processor calculates a separation distance between the object of the map information and the global path, determines the second space when the separation distance is within a predetermined reference value, and determines that the separation distance exceeds the reference value, It may be characterized in that it is determined as the first space.

In addition, the processor may be characterized in that the driving is stopped and waited until the approach time and the approach distance are reached, and then the first avoidance path is generated when the approach time and the approach distance are reached. .

In addition, as soon as the obstacle is detected, a virtual obstacle line may be generated, and the second avoidance path may be generated so as not to come into contact with the virtual obstacle line.

In addition, the virtual obstacle line may be disposed in front of the obstacle on the global path.

In addition to this, a computer program stored in a computer readable recording medium for execution to implement the present disclosure may be further provided.

In addition to this, a computer readable recording medium recording a computer program for executing a method for implementing the present disclosure may be further provided.

According to the above-mentioned problem solving means of the present disclosure, it is possible to efficiently avoid obstacles and quickly travel to a destination.

The effects of the present disclosure 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 below.

1 is a diagram showing the configuration of a mobile robot according to the present disclosure.
FIG. 2 is a diagram showing map information of a large building stored in the memory of FIG. 1 as an example.
3 is a flowchart illustrating a method for autonomous driving customized for each space size of a mobile robot according to the present disclosure.
FIG. 4 is a diagram illustrating a process of determining spaces in the processor of FIG. 1 as an example.
FIG. 5 is a diagram illustrating an avoidance driving process when the processor of FIG. 1 determines that the space is a first space, as an example.
6 and 7 are diagrams illustrating an example of an avoidance driving process when the processor of FIG. 1 determines that the space is a second space.
8 is a diagram illustrating a process of storing optimal avoidance path data in the memory of FIG. 1 as an example.
9 is a flowchart illustrating an autonomous driving method customized for each space size of an artificial intelligence-based mobile robot according to the present disclosure.
10 to 12 are diagrams illustrating a safe driving process when the processor of FIG. 1 determines the first space and the second space as an example.

Like reference numbers designate like elements throughout this disclosure. The present disclosure does not describe all elements of the embodiments, and general content or overlapping content between the embodiments in the technical field to which the present disclosure belongs is omitted. The term 'unit, module, member, or block' used in the specification may be implemented as software or hardware, and according to embodiments, a plurality of 'units, modules, members, or blocks' may be implemented as one component, It is also possible that one 'part, module, member, block' includes a plurality of components.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case of being directly connected but also the case of being indirectly connected, and indirect connection includes being connected through a wireless communication network. do.

In addition, when a certain component is said to "include", this means that it may further include other components without excluding other components unless otherwise stated.

Throughout the specification, when a member is said to be located “on” another member, this includes not only a case where a member is in contact with another member, but also a case where another member exists between the two members.

Terms such as first and second are used to distinguish one component from another, and the components are not limited by the aforementioned terms.

Expressions in the singular number include plural expressions unless the context clearly dictates otherwise.

In each step, the identification code is used for convenience of description, and the identification code does not explain the order of each step, and each step may be performed in a different order from the specified order unless a specific order is clearly described in context. there is.

Hereinafter, the working principle and embodiments of the present disclosure will be described with reference to the accompanying drawings.

In this specification, the control unit of the mobile robot according to the present disclosure includes all various devices capable of providing results to users by performing calculation processing. For example, the control unit according to the present disclosure may include both a computer and a server, or may be in any one form.

Here, the computer may include, for example, a laptop, a desktop, a laptop, a tablet PC, a slate PC, a single board computer, and the like with a web browser.

A server device is a server that processes information by communicating with an external device, and may include an application server, a computing server, a database server, a file server, a mail server, a proxy server, and a web server.

A portable terminal is, for example, a wireless communication device that ensures portability and mobility, and includes a Personal Communication System (PCS), Global System for Mobile communications (GSM), Personal Digital Cellular (PDC), Personal Handyphone System (PHS), and PDA (Personal Handyphone System). Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet) terminal, smart phone All kinds of handheld-based wireless communication devices such as watches, rings, bracelets, anklets, necklaces, glasses, contact lenses, or wearable devices such as head-mounted-devices (HMDs) can include

The mobile robot according to the present disclosure generates a global path to the destination, determines whether it is a first space larger than a preset space size or a second space smaller than the preset space size according to the global path to the destination, and determines whether the first space is a second space smaller than the preset space size. When the preset approach time and approach distance to the obstacle are reached in the state where the obstacle is detected, a first avoidance path is created according to the approach state of the obstacle and the first avoidance path is created, and in the second space, the moment the obstacle is detected, It may be provided to generate a second avoidance path for traveling while immediately avoiding the obstacle.

Such a mobile robot can efficiently avoid obstacles and travel quickly to a destination.

Hereinafter, the mobile robot will be examined in detail.

1 is a diagram showing the configuration of a mobile robot according to the present disclosure. FIG. 2 is a diagram showing map information of a large building stored in the memory of FIG. 1 as an example.

Referring to FIG. 1 , the mobile robot 100 may include a sensor unit 110, a control unit 120, and a driving unit 130.

The sensor unit 110 may include at least one of an ultrasonic sensor, a radar sensor, and a lidar sensor. The sensor unit 110 may detect an obstacle or an unexpected situation in front of the mobile robot 100 . Here, the obstacle may refer to any element that hinders the movement of the mobile robot 100 while driving. For example, the obstacle may be a stationary object, a mobile object, or a mobile animal located in a space where the mobile robot 100 is automatically driven. For example, the fixed objects may be walls, furniture, stalls, pillars, stairs, etc., the mobile objects may be other mobile robots performing the same task, and the mobile animals may be people, cats, dogs, and the like.

The control unit 120 performs the above-described operation using a memory 121 storing data for an algorithm or a program reproducing the algorithm for controlling the operation of the components in the present device, and the data stored in the memory 121. It may be implemented with at least one processor 122 that does. Here, the memory 121 and the processor 122 may be implemented as separate chips. Also, the memory 121 and the processor 122 may be implemented as a single chip.

The memory 121 may store data supporting various functions of the device and programs for operation of the control unit, may store input/output data, and may store a plurality of application programs (application programs or applications running in the device). (application)), data and instructions for the operation of the device can be stored. At least some of these application programs may be downloaded from an external server through wireless communication.

The memory 121 is a flash memory type, a hard disk type, a solid state disk type, a silicon disk drive type, and a multimedia card micro type. micro type), card type memory (eg SD or XD memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable EEPROM (EEPROM) It may include a storage medium of at least one type of a programmable read-only memory (PROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, and an optical disk. In addition, the memory 211 is separated from the apparatus, but may be a database connected by wire or wirelessly.

The memory 121 may store global routes for each destination. The global route for each destination may be a route for the mobile robot 100 to arrive at a specific destination in a space such as a hospital, an airport, or a mall.

The memory 121 may store map information A. In this case, as shown in FIG. 2 , the map information A may be a grid map. The grid map may be generated in a form for 2D or 3D autonomous navigation through sensing of the mobile robot 100 or by using various simultaneous localization and mapping (SLAM) techniques in advance. For example, the grid map may be generated as a 2D map in the form of an image file such as png or pgm, or as a 3D map through point cloud data including 3-axis spatial information. The first space S1 of the map information A may be a wide space, and the second space S2 of the map information A may be a narrow space.

The processor 122 may control an operation for driving while avoiding an obstacle for each destination.

For example, the processor 122 may generate a global path to the destination and determine whether the global path to the destination is a first space larger than the preset space size or a second space smaller than the preset space size. Here, the processor 122 may calculate a separation distance between the object of map information and the global path. The processor 122 may determine the second space if the separation distance is within a preset reference value, and determine the first space if the separation distance exceeds the reference value. In this case, the first space may be larger than the second space.

In the first space, the processor 122 may generate a first avoidance path for avoiding and traveling according to the approach state of the obstacle when the preset approach time and approach distance to the obstacle are reached in a state in which the obstacle is detected. . For example, if the processor 122 is in the first space, the processor 122 stops driving and waits until the preset approach time and approach distance to the obstacle are reached in a state where the obstacle is detected, and then the preset access time and approach to the obstacle is reached. When the distance is reached, a first avoidance path may be created.

If the processor 122 is in the second space, upon detecting an obstacle, the processor 122 may immediately avoid the obstacle and create a second avoidance path for driving. Here, the processor 122 may generate a virtual obstacle line upon detecting an obstacle, and generate a second avoidance path so as not to come into contact with the virtual obstacle line. At this time, the virtual obstacle line may be placed in front of the obstacle on the global path.

The driving unit 130 may drive the mobile robot 100 to travel to the destination based on the first avoidance path and the second avoidance path generated through the processor 122 . For example, the driving unit 130 may include a battery, an electronic component supplied with power from the battery, and a wheel or drive belt driven by receiving power from the battery and the battery in order to provide power necessary for driving the mobile robot 100. It may include mechanical driving parts such as

3 is a flowchart illustrating a method for autonomous driving customized for each space size of a mobile robot according to the present disclosure. FIG. 4 is a diagram illustrating a process of determining spaces in the processor of FIG. 1 as an example.

FIG. 5 is a diagram illustrating an avoidance driving process when the processor of FIG. 1 determines that the space is a first space, as an example. 6 and 7 are diagrams illustrating an example of an avoidance driving process when the processor of FIG. 1 determines that the space is a second space.

3 to 7, the autonomous driving method customized for each space size of a mobile robot includes a global route generation step (S310), a determination step (S330), a first avoidance route generation step (S350a), and a second avoidance route generation step. (S350b) and driving step (S370).

In the global path generating step, the mobile robot 100 may generate a global path MR to the destination (S310).

In the determining step, it may be determined whether the mobile robot 100 is in a first space S1 larger than a preset space size or a second space S2 smaller than a preset space size according to the global route MR to the destination. Yes (S330). The mobile robot 100 may determine whether it is the first space S1 or the second space S2 based on the sensor unit 110 including at least one of an ultrasonic sensor, a radar sensor, and a LiDAR sensor. Here, the processor 122 may calculate the separation distance d1 between the object P of the map information A and the global path MR, as shown in FIG. 4 . The processor 122 may determine the second space S2 when the separation distance d1 is within a preset reference value. The processor 122 may determine the first space S1 when the separation distance d1 exceeds the reference value. In this case, the first space S1 may be larger than the second space S2.

In the first avoidance path generation step, if the first space S1 is through the processor 122, as shown in FIG. t2) and the approach distance d2, a first avoidance path R1 for avoiding and driving may be created according to the approaching state of the obstacle D (S350a). For example, if the processor 122 is in the first space S1, in a state in which the obstacle D is detected, driving is performed until the preset approach time t2 and approach distance d2 with the obstacle D are reached. After stopping and waiting, the first avoidance path R1 may be created when the approach time t2 and the approach distance d2 with the preset obstacle D are reached. At this time, the length and width of the sensing range C can be determined by the user. When an obstacle D is detected in the detection range C, the mobile robot 100 determines the detected position, travels to the vicinity of the detected position without avoiding, and the approach time and distance to the obstacle D are determined. The driving stops and waits until the approach time (t2) and approach distance (d2) with the preset obstacle (D) are reached, and then the approach time (t2) and approach distance (d2) with the preset obstacle (D) are reached. , a first avoidance path R1 for traveling while avoiding the obstacle D may be created.

In the second avoidance path generating step, if the second space S2 is through the processor 122, as shown in FIG. 6, the moment the obstacle D is detected, the obstacle D is immediately avoided and driven. A second avoidance path R2 may be created (S350b). At this time, the length and width of the sensing range C can be determined by the user. Since the second space S2 is a narrow space, when the mobile robot 100 is stopped, mobile objects such as other mobile robots that occupy the space and perform the same task, and mobile animals such as people, cats, and dogs move. can interfere with Accordingly, in such a narrow space, the mobile robot 100 can quickly detect the obstacle D and create a second avoidance path R2 for avoiding the obstacle D in advance.

In addition, as shown in FIG. 7 , the processor 122 generates a virtual obstacle line L1 at the moment when the obstacle D is detected, and the second avoidance path R2 so as not to come into contact with the virtual obstacle line L1. ) can also be created. Here, the virtual obstacle line L1 may be created in a T-shape and disposed in front of the obstacle D on the global path MR. In this case, the processor 122 may generate a second avoidance path R2 surrounding the obstacle D in an elliptical shape by reflecting the generated virtual obstacle line L1 to the local costmap.

In the driving step (S270), after avoiding the first avoidance path R1 and the second avoidance path R2 generated through the processor 122, the driver 130 drives the global path MR to the destination. It is possible to drive the mobile robot 100 through.

Meanwhile, in the memory 121 , image information for each space, image information for each obstacle type, and avoidance path information may be linked and stored.

8 is a diagram illustrating a process of storing optimal avoidance path data in the memory of FIG. 1 as an example.

Referring to FIG. 9 , the server 10 electrically connected to the control unit 120 converts input values of spatial image data ID1, image data ID2 for each obstacle type, and avoidance path data ID3, which are meta data, to an avoidance path. It is possible to output the result value of the optimal avoidance path data OD recommended by learning based on the model M.

In this case, the spatial image data ID1 may include each spatial size and each spatial coordinate. In addition, the image data ID2 for each type of obstacle may include stationary object data ID2-1, mobile object data ID2-2, and mobile animal data ID2-3. For example, fixed object data (ID2-1) may be walls, furniture, stalls, pillars, stairs, etc., mobile object data (ID2-2) may be other mobile robots performing the same task, and mobile animal data. (ID2-3) may be a person, cat, dog, etc. Also, the avoidance path data ID3 may be data for avoiding obstacles in various avoidance paths while the mobile robot 100 is driving.

The avoidance path model M may be constructed to learn spatial image data ID1 included in the input data, image data ID2 for each obstacle type, and avoidance path data ID3 through correlation. The avoidance path model (M) can be constructed and reinforced-learned as a training data set using a CNN algorithm or an RNN algorithm on spatial image data (ID1), image data (ID2) for each obstacle type, and avoidance path data (ID3).

The server 10 stores the optimal avoidance path data OD in the space determined based on the spatial image data ID1, the image data ID2 for each obstacle type, and the avoidance path data ID3 in the memory of the controller 120. (121). The memory 121 may store the received optimal avoidance path data OD in the corresponding space. At this time, the memory 211 may store optimal avoidance path data according to the corresponding obstacle type in the corresponding space.

When an image including a space and an obstacle type is received from the camera 140, the processor 122 may extract space information and obstacle type information from the image based on artificial intelligence. Here, the processor 122 may extract space information and obstacle type information using an object detection model. In this case, spatial information may include spatial size and spatial coordinates, and obstacle type information may include fixed objects, movable objects, and movable animals.

Here, the camera 140 may be provided as a video camera using an image sensor. The camera 140 may process an image frame such as a still image or a moving image acquired through an image sensor. Meanwhile, when there are a plurality of cameras 140, they may be arranged to form a matrix structure. A plurality of image information having various angles or focal points may be input through image cameras constituting such a matrix structure. In addition, the video cameras may be arranged in a stereo structure to acquire left and right images for realizing a 3D stereoscopic image. Also, the camera 140 may be provided as an AI camera. The AI camera can fine-tune the image recognized through a wide-angle sensor that mimics the human retina with a brain neural network algorithm. The AI camera can adjust shutter speed, light exposure, saturation, color depth, dynamic range, and contrast. In addition, the AI camera can output the captured image as a high-quality image.

9 is a flowchart illustrating an autonomous driving method customized for each space size of an artificial intelligence-based mobile robot according to the present disclosure.

First, the autonomous driving method customized for each space size of the artificial intelligence-based mobile robot of FIG. It can be performed including generating an avoidance path (S350a), generating a second avoidance path (S350b), and driving (S370). Since the function of each step and the organic connection between them have been described above in the method of FIG. 3, they will be omitted below.

Referring to FIG. 9 , the autonomous driving method customized for each space size of an artificial intelligence-based mobile robot includes an image acquisition step (S341), an obstacle type determination step (S342, S343, and S345), a notification step (S344), and a first output step. (S346) and a second output step (S347).

In the image acquisition step, the type image of the obstacle in the first space S1 or the second space S2 obtained from the camera 140 may be received through the processor 122 (S341).

In the obstacle type determination step, the type of obstacle in the image may be determined based on artificial intelligence through the processor 122 (S342, S343, S345). The processor 122 may determine whether the type of obstacle is a fixed object or a movable object.

Here, the processor 122 may generate a first avoidance path (S350a) or a second avoidance path (S350b) if it is a fixed object (YES in S342). In this case, the fixed objects may be walls, furniture, stalls, pillars, stairs, and the like.

In the obstacle type determination step, if the object is not a fixed object through the processor 122 (No in S342), it may be determined whether the object is a movable object (S343).

In the notification step, if it is a mobile object (Yes in S343) through the processor 122, the expected movement direction and expected movement path may be notified to other mobile robots performing the same task through the communication unit 150 (S344). Here, the mobile object may be another mobile robot performing the same task, and since the other mobile robot does not have a significant change in the moving direction, the mobile robot detects the moving direction of the other mobile robot using a camera and lidar sensor, , based on the detected movement direction, it can move in a direction that does not overlap with other mobile robots. The mobile robot may transmit coordinates of an expected movement direction and an estimated movement path to other mobile robots through the communication unit 150 to notify them. The processor 122 may generate a first avoidance path (S350a) or a second avoidance path (S350b) when notifying the expected movement direction and expected movement path (S344).

Here, the communication unit 150 includes a WiFi module and a wireless broadband module, as well as a global system for mobile communication (GSM), code division multiple access (CDMA), wideband code division multiple access (WCDMA), and UMTS ( A wireless communication module supporting various wireless communication schemes such as universal mobile telecommunications system (TDMA), Time Division Multiple Access (TDMA), Long Term Evolution (LTE), 4G, 5G, and 6G may be included.

In the obstacle type determination step, if it is not a mobile object through the processor 122 (No in S343), it may be determined as a mobile animal. Since the mobile animal (human, pet) may change its movement direction instantaneously, it is necessary to notify the mobile animal of the expected movement direction and expected movement path before or simultaneously with the movement to the avoidance path.

In the first output step, if the mobile animal is a human (YES in S345) through the processor 122, the expected movement direction and expected movement path may be output through the output unit 160 as voice to inform the human (S346). . For example, the output unit 160 may be any means for outputting voice, such as a voice speaker. When the processor 122 outputs the expected movement direction and expected movement path as voice (S346), it may generate a first avoidance path (S350a) or a second avoidance path (S350b).

In the second output step, if the mobile animal is not a human (No in S345) through the processor 122, a warning sound may be output through the output unit 160 to inform the pet (dog, cat, etc.) (S347). . For example, the output unit 160 may be any means for outputting a warning sound such as a buzzer. At this time, the buzzer is provided on the left and right sides of the mobile robot, and outputs a warning sound in a direction opposite to the expected movement path so that the pet can avoid it. That is, when the mobile robot evades forward by 45 degrees to the left, the buzzer provided on the right outputs a warning sound, so that the pet can further run away to the right. When the processor 122 outputs a warning sound in a direction opposite to the expected movement path (S347), it may generate a first avoidance path (S350a) or a second avoidance path (S350b).

10 to 12 are diagrams illustrating a safe driving process when the processor of FIG. 1 determines the first space and the second space as an example.

10 to 12, the present disclosure provides coast boundary lines CL11 and CL12 (from grid lines DL11 and DL12 (DL21 and DL22) of map information A so that the mobile robot 100 can move safely. By giving cost values up to CL21 and CL22, it is possible to prevent the mobile robot 100 from moving near the wall. At this time, the inside between the coast boundary lines CL11 and CL12 (CL21 and CL22) may be a safe space in which the mobile robot 100 can move.

Based on the cost value, the processor 122 determines a first space S1, which is a wide space, and a second space S2, which is a narrow space, of the map information A, and the mobile robot 100 determines the first space The driving of the mobile robot 100 can be controlled to prevent movement near the wall of (S1) or the wall of the second space (S).

At this time, the first cost value is a distance between PFk points (PFk1 to PFkn), which are a set of points on the coast boundary lines (CL11, CL21), and PGk points (PGk1 to PGkn), which are a set of points perpendicular to the grid lines (DL11, DL21). can be derived. And, the second coast value obtains the equation of a straight line using the PFk points (PFk1 to PFkn) and the PGk points (PGk1 to PGkn), and based on the equation of the straight line, on the coast boundary lines (CL12, CL22) having the same slope It can be calculated as a distance between PFj points (PFj1 to PFjn), which are a set of points, and PGj points (PGj1 to PGjn), which are a set of points perpendicular to the grid lines (DL12, DL22) at the PFj points (PFj1 to PFjn). In this case, a safe space may be provided between the PFk points PFk1 to PFkn and the PFj points PFj1 to PFjn.

Here, the distance between the PFk points PFk1 to PFkn and the PFj points PFj1 to PFjn may be defined as d. At this time, if d is smaller than the user-specified value dth, it may be set as a narrow second space (S2), and if it is greater than the user-specified value dth, it may be set as a wide space (first space S1).

Therefore, the autonomous driving method and program customized for each space size of a mobile robot and an artificial intelligence-based mobile robot according to the present disclosure can efficiently avoid obstacles and drive quickly to a destination.

100: mobile robot 110: sensor unit
120: control unit 121: memory
122: processor 130: driving unit
140: camera 150: sound output unit

Claims (10)

In the autonomous driving method customized for each space size of an artificial intelligence-based mobile robot,
generating, by the mobile robot, a global route to a destination;
determining whether the mobile robot is in a first space larger than a preset space size or a second space smaller than the preset space size according to the global route to the destination;
In the first space, when the approach time required to collide with a preset obstacle and the approach distance required to collide with a preset obstacle in a state in which the mobile robot detects an obstacle based on a lidar sensor are reached, the approach state of the obstacle generating a first avoidance path for avoiding and traveling according to; and
If it is the second space, generating a second avoidance path for traveling by immediately avoiding the obstacle when the mobile robot detects the obstacle based on the lidar sensor,
The judgment step is
A separation distance between a stationary object of map information preset in the mobile robot and the global path is calculated, and if the separation distance is within a preset reference value, it is determined as the second space, and if the separation distance exceeds the reference value, Determined as the first space,
The first avoidance path generating step,
The mobile robot does not create the first avoidance path until the approaching time and the approaching distance are reached based on the lidar sensor, and waits while stopping the driving, and then waits for the approaching time based on the lidar sensor. and generating the first avoidance path when the approach distance is reached;
In the step of generating the second avoidance path,
The moment the mobile robot detects the obstacle, it creates a virtual obstacle line,
characterized in that the second avoidance path is generated so that the mobile robot does not come into contact with the virtual obstacle line. delete delete delete A program stored in a computer-readable recording medium in order to execute a customized autonomous driving method for each space size of the artificial intelligence-based mobile robot of claim 1 in combination with a computer, which is hardware. a sensor unit that detects an obstacle;
Memory storing global route and map information by destination; and
A processor controlling an operation for driving while avoiding an obstacle for each destination,
the processor,
create a global route to the destination,
determining whether a first space larger than a preset space size or a second space smaller than the preset space size is determined according to the global path to the destination;
In the first space, when the approach time required to collide with a preset obstacle and the approach distance required to collide with a preset obstacle in a state where the obstacle is detected based on the lidar sensor of the sensor unit are reached, the approach state of the obstacle generating a first avoidance path for driving by avoiding according to
If it is the second space, the moment the obstacle is detected based on the lidar sensor, a second avoidance path for driving by immediately avoiding the obstacle is created,
calculating a separation distance between a stationary object of the map information and the global path;
If the separation distance is within a preset reference value, it is determined as the second space,
When the separation distance exceeds the reference value, it is determined as the first space,
The first avoidance route is not generated until the approach time and the approach distance are reached based on the lidar sensor, and the driving is stopped and waited. generating the first avoidance path when
The moment the obstacle is detected based on the lidar sensor, a virtual obstacle line is created,
Characterized in that the second avoidance path is generated so as not to come into contact with the virtual obstacle line, the mobile robot. delete delete delete According to claim 6,
The mobile robot, characterized in that the virtual obstacle line is disposed in front of the obstacle on the global path.

Images