KR102682467B1 - Indoor and outdoor patrol robot platform and patrol robot control method - Google Patents

Abstract

The present invention relates to a method of controlling a patrol robot, and more specifically, to control a robot that autonomously patrols a given indoor/outdoor facility or space, to identify unidentified persons, track the unidentified persons when they run away, and protect wild animals. This relates to a method of controlling a patrol robot that guides it to move away from the target facility and into a safe area.

Description

Indoor and outdoor patrol robot platform and patrol robot control method {Indoor and outdoor patrol robot platform and patrol robot control method}

The present invention relates to a method of controlling a patrol robot, and more specifically, to control a robot that autonomously patrols a given indoor/outdoor facility or space, to identify unidentified persons, track the unidentified persons when they run away, and protect wild animals. This relates to a method of controlling a patrol robot that guides it to move away from the target facility and into a safe area.

Interest in security systems for various facilities, city centers, and residential areas is increasing.

In particular, major national facilities or corporate research and development facilities are seeking various measures to prevent security threats caused by human resources.

To this end, many facilities/companies employ security management personnel and use CCTV or direct human patrol to prevent unauthorized personnel from infiltrating around or inside the facility.

However, human loopholes occur in CCTV surveillance and patrolling.

For example, not only do situations arise where people cannot see due to dark environments at night or blind spots, but there are also many cases where proper security work is not carried out due to loss of concentration due to working for long hours. I'm doing it.

Meanwhile, patrolling rice fields, etc. is becoming difficult due to the decline and aging of the rural population, and as a result, damage to crops caused by wild animals frequently occurs, but no response is being taken.

Domestic Patent No. 10-2069505 discloses “Driving performance improvement device and surveillance and reconnaissance system of driving equipment.” The above patent document proposes a method for a robot to recognize obstacles in a surveillance and reconnaissance system. However, this obstacle recognition method has the limitation that it cannot be a response to the intrusion of unauthorized personnel (people) or damage to wild animals to facilities.

The problem that the present invention aims to solve is to solve the limitations of conventional facility security technology, so that a self-driving robot automatically patrols indoor/outdoor facilities, etc., and when a person is discovered, it autonomously responds and sends an alert to the manager. The goal is to provide a method of controlling a patrol robot that performs patrolling tasks.

In addition, patrolling around facilities such as rice fields (facilities subject to protection) can be used to guide wild animals away from the fields (to a safe area) when wild animals appear.

In order to solve the limitations of conventional facility security technology, the present invention includes the steps of autonomously driving along a designated path and detecting surrounding objects (hereinafter referred to as the 'patrolling step'); If an animal is detected while driving, transmitting a guidance voice and turning on a light; Transmitting animal detection information to an administrator terminal; confirming the identity of the animal; And after completing the designated route, moving to the charging dock to proceed with charging; and the step of confirming the identity of the animal is: If the identity of the animal is confirmed, switching back to the patrol step. Step; If the identity of the animal is not confirmed, transmitting an image of the unidentified animal (hereinafter referred to as 'unidentified person') to the administrator terminal to request a confirmation procedure; When an identity verification message is received from the manager terminal, switching back to the patrol stage; and, if an identity verification message is not received from the manager terminal, tracking the unidentified person.

In addition, based on the map information of the patrol area, it further includes a step of distinguishing an intersection and a road, and the step of tracking the unidentified person is: an intersection at the end based on the road where the unidentified person is located (hereinafter referred to as 'selected selecting an intersection (referred to as an 'intersection point'); And designate the selected intersection as the target point, and use a patrol robot (hereinafter referred to as 'second patrol robot') other than [the patrol robot that detected the unidentified person (hereinafter referred to as 'first patrol robot')]. Including the step of moving, the road where the unidentified person is located may be surrounded by a second patrol robot or the unidentified person may be guided to a safe area.

Then, among the roads connected to the selected intersection, [the road in the direction away from the road where the unidentified person is located] is designated as the target point, and patrol robots other than the first and second patrol robots (hereinafter referred to as 'third patrol robots') Including a step of moving the unidentified person, the expected escape route of the unidentified person may be blocked by a third patrol robot.

In addition, when the unidentified person escapes from the nth patrol robot (n is one or more numbers from 1 to 3), a patrol robot other than the nth patrol robot is selected along the expected escape route predicted based on the escape direction of the unidentified person. Moving step (hereinafter referred to as 'chase step'); and repeating the chase step until the unidentified person stops or a stop command is received from the administrator terminal.

In addition, the patrol step includes: taking a predetermined comparison image while driving; Comparing the comparison image taken in the previously performed patrol phase (hereinafter referred to as the 'first comparison image') with the comparison image taken in the current patrol phase (hereinafter referred to as the 'second comparison image'). step; When a difference is detected between the first comparison image and the second comparison image, transmitting the first comparison image and the second comparison image to an administrator terminal to request a confirmation procedure; When a difference confirmation message is received from the manager terminal, switching back to the patrol stage; And if a difference confirmation message is not received from the manager terminal, searching the road where the difference is located, the step of searching the road where the difference is located is: an intersection at the end of the road where the difference is located. selecting (hereinafter referred to as 'selected intersection'); A step of re-driving and patrolling the road where the difference is located without missing anything; And designate the selected intersection as the target point, and move the patrol robot (hereinafter referred to as 'second patrol robot') other than [the patrol robot that detected the difference (hereinafter referred to as 'first patrol robot')]. Including, the road where the difference is located may be surrounded by a second patrol robot.

Additionally, a patrol robot controlled according to the above patrol robot control method and a patrol robot system (platform) including the patrol robot are provided.

According to an embodiment of the present invention, a robot system that automatically patrols certain indoor/outdoor facilities can be configured.

Additionally, when an unidentified person is detected, it is possible to prevent the unidentified person from escaping.

In addition, in order to complement the limitations of the object (human/animal) recognition algorithm (AI), the images captured by the patrol robot are compared to derive differences, and when differences are found, the disguised unidentified person is identified by intensively patrolling the area. You can find it.

In addition, when wild animals are detected, the wild animals can be guided in a direction (safe area) away from protected facilities (paddy fields, crops, private houses, etc.).

1 is an operation flowchart of a patrol robot control method according to an embodiment of the present invention.
Figure 2 is a diagram showing a state in which patrol robots are controlled according to a patrol robot control method according to an embodiment of the present invention.

Hereinafter, various embodiments of this document are described with reference to the attached drawings. However, this is not intended to limit the technology described in this document to specific embodiments, and should be understood to include various modifications, equivalents, and/or alternatives to the embodiments of this document. . In connection with the description of the drawings, similar reference numbers may be used for similar components.

In this document, expressions such as "comprises" or "may include" refer to the presence of the corresponding feature (e.g., a numerical value, function, operation, or component, etc.) and do not exclude the presence of additional features. No.

In this document, expressions such as “A or B,” “at least one of A or/and B,” or “one or more of A or/and B” may include all possible combinations of the items listed together. . For example, “A or B,” “at least one of A and B,” or “at least one of A or B” (1) includes at least one A, (2) includes at least one B, or (3) it may refer to all cases including both at least one A and at least one B.

Expressions such as “first,” “second,” etc. used in this document may modify various components regardless of order and/or importance, and are only used to distinguish one component from another component. The components are not limited. For example, a first user device and a second user device may represent different user devices regardless of order or importance. For example, a first component may be renamed a second component without departing from the scope of rights described in this document, and similarly, the second component may also be renamed to the first component.

Of course, various modifications can be made by those skilled in the art without departing from the gist of the present invention as claimed in the claims of the present invention, and these modifications do not reflect the technical idea or the technical spirit of the present invention. It should not be understood in isolation from the perspective.

In the terms used in this specification, singular expressions should be understood to include plural expressions, unless clearly interpreted differently from the context, and terms such as “including” refer to the described features, numbers, steps, operations, and components. , it means the existence of parts or a combination thereof, but should be understood as not excluding the possibility of the presence or addition of one or more other features, numbers, step operation components, parts, or combinations thereof.

Additionally, the method according to the present invention can be implemented as computer-readable code (program) on a computer-readable recording medium. Computer-readable recording media may include all types of recording devices that store data that can be read by a computer system. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage devices, and can also be implemented in the form of a carrier wave (e.g., transmitted via the Internet). Includes. Additionally, the computer-readable recording medium may store computer-readable code that can be executed in a distributed manner by a distributed computer system connected to a network.

A plurality of patrol robots (three or more) may be distributed outdoors and/or indoors and configured to drive autonomously. LiDAR sensors, obstacle avoidance functions, etc. for this purpose are the domain of known technology, so detailed descriptions will be omitted.

The patrol robot may be equipped with a speaker or/and display for transmitting a predetermined guidance message to the animal, and may be equipped with an LED that emits predetermined lighting.

First, in standby mode (charging mode), the battery can be charged using a wireless charger provided at the patrol robot's waiting area (charging dock). Afterwards, the autonomous driving route, patrol time, etc. can be input by manipulating the patrol robot server or patrol robot terminal.

Afterwards, in autonomous driving mode, it patrols along the designated path and automatically avoids obstacles identified along the path. However, if the obstacle is judged to be a person, a guidance voice is sent to guide the person so that it recognizes the robot and does not collide with the robot. , In addition, identity verification procedures can be carried out.

If the set patrol time has passed or the battery is low, the patrol robot can be moved back to the waiting area and switched to charging (standby) mode. When the battery is forcibly switched to charging mode, the driving command (autonomous driving path, patrol time, etc.) held by the low-battery patrol robot is sent to other patrol robots located in the waiting area of the patrol robot that has not yet received a driving command. You can automatically switch to patrol mode by sending the message.

Each step described below can be performed in a form where a plurality of patrol robots are controlled centrally by a server or/and the patrol robots operate independently.

The term 'animal' used in the description of the present invention means 'a living organism that moves dynamically as one of the two branches of the biological world (plants/animals). Including people.’ The term ‘wild animal’ used in contrast is ‘stray animals, birds, water animals, etc. that can cause damage to facilities (paddy fields, private houses, etc.). It can mean ‘not including people.’

According to an embodiment of the present invention, a step of autonomously driving along a designated path and detecting surrounding objects (hereinafter referred to as a 'patrolling step'); When an animal (including a person) is detected while driving, transmitting a guidance voice and turning on a light; Transmitting animal detection information to an administrator terminal; confirming the identity of the animal; And after completing the designated route, moving to the charging dock to proceed with charging; and the step of confirming the identity of the animal is: If the identity of the animal is confirmed, switching back to the patrol step. Step; If the identity of the animal is not confirmed, transmitting an image of the unidentified animal (hereinafter referred to as 'unidentified person') to the administrator terminal to request a confirmation procedure; When an identity verification message is received from the manager terminal, switching back to the patrol stage; and, if an identity verification message is not received from the manager terminal, tracking the unidentified person.

If it is confirmed that the unidentified person is a wild animal rather than a human, the patrol robot may track the wild animal and operate in a guiding direction to return it to a safe area.

“Step of transmitting animal detection information to the manager terminal,” “Step of transmitting an image of an unidentified person to the manager terminal to request a confirmation procedure,” and “If an identification message is received from the manager terminal, go back to the patrol step.” The “conversion step” may transmit data via a predetermined server.

The step of confirming the animal's identity may be performed by one or more of ID card authentication, face recognition, fingerprint recognition, iris recognition, voice recognition, and password authentication. Naturally, wild animals cannot pass the above-mentioned identification steps, so they may be judged as unidentified.

If it is determined that the person is unidentified, an alarm sound can be generated through the speaker to encourage personnel such as managers and security guards located nearby to quickly find the location. At this time, the light can also flash to combine visual and auditory information to help managers and other personnel quickly find the location of the unidentified person. This operation may function to make the wild animal feel threatened by the patrol robot if the unidentified person is a wild animal.

In addition, the patrol robot moves a predetermined short distance (approximately 30 to 50 cm) in the direction closer to the wild animal, and then repeats the movement of returning to the original position, thereby encouraging the wild animal to run away from the patrol robot. You can.

The step of detecting the surrounding objects may be performed by a predetermined sensor. The sensor may be provided on the patrol robot itself, or may be placed on one or more sides of the facility where the patrol robot is operated. Sensors may include one or more of surveillance cameras, infrared sensors, thermal imaging cameras, vibration sensors, and sound sensors. The sensors detect the cries of wild animals when the animal passes through the infrared sensor, when a heat change is detected in a specific area, when irregular vibration is detected, when irregular or sounds above a certain level are detected, and when the sound of a wild animal is detected. When this occurs, it can be determined that there is an object or animal nearby.

When an object is detected, lights located in the corresponding area of the facility can be turned on. By turning on the lights, you can more accurately identify the face of an animal (especially a human) in that location, and prevent the animal from being detected by cameras or other sensors because it is obscured by surrounding terrain.

In addition, based on the map information of the patrol area, it further includes a step of distinguishing an intersection and a road, and the step of tracking the unidentified person is: an intersection at the end based on the road where the unidentified person is located (hereinafter referred to as 'selected selecting an intersection (referred to as an 'intersection point'); And designate the selected intersection as the target point, and use a patrol robot (hereinafter referred to as 'second patrol robot') other than [the patrol robot that detected the unidentified person (hereinafter referred to as 'first patrol robot')]. Including the step of moving, the road where the unidentified person is located may be surrounded by a second patrol robot or the unidentified person may be guided to a safe area.

The road (corridor) refers to a surface on which a patrol robot can run in a facility, etc., and may be divided by walls, obstacles, shelves, crops, etc.

The intersection refers to a point/place where two or more roads intersect.

The selected intersection may mean the intersection itself or a road in the direction of movement of the unidentified person adjacent to the intersection.

In the map information, areas that must be protected from being stepped on by patrol robots in facilities such as rice fields may be designated to be recognized as no-driving areas. Accordingly, vacant areas where crops, etc. are not planted, may be designated as intersections and roads.

When the unidentified person is located at an intersection, it is assumed that the unidentified person is located simultaneously in all corridors connected to the intersection, and a plurality of patrol robots can be driven to block all ends of the corridors.

By applying a predetermined path creation algorithm, a patrol robot can be placed in advance on a possible path according to the direction in which the unidentified person moves.

Meanwhile, if the unidentified person is a wild animal, the wild animal can be encouraged to move in that direction by only opening the road leading to the safe area (by not deploying a patrol robot).

Figure 2a shows a state in which the first patrol robot has discovered an unidentified person and the second patrol robot is moving toward the selected intersection.

Figure 2b shows a state in which the second patrol robot has arrived at the selected intersection.

Below, we will explain in detail how to 'surround' the escape route when the unidentified person is a person.

Then, among the roads connected to the selected intersection, [the road in the direction away from the road where the unidentified person is located] is designated as the target point, and patrol robots other than the first and second patrol robots (hereinafter referred to as 'third patrol robots') Including a step of moving the unidentified person, the expected escape route of the unidentified person may be blocked by a third patrol robot.

If the unidentified person flees toward the lower right road in FIG. 2b, the third patrol robot may move toward intersections of the expected escape route and block the expected escape route.

In addition, when the unidentified person escapes from the nth patrol robot (n is one or more numbers from 1 to 3), a patrol robot other than the nth patrol robot is selected along the expected escape route predicted based on the escape direction of the unidentified person. Moving step (hereinafter referred to as 'chase step'); and repeating the chase step until the unidentified person stops or a stop command is received from the administrator terminal.

Figure 2c shows a state in which an unidentified person escapes from the nth patrol robot (third patrol robot) and places the first and second patrol robots on the expected escape route.

By calculating the distance that each patrol robot moves to each selected intersection, the patrol robot whose sum of the distances is the minimum can be moved to the selected intersection. In Figure 2 c, the first patrol robot blocks the intersection selected on the left, and the second patrol robot, which was relatively on the left in Figure 2 b, blocks the intersection selected below, and among the second patrol robots, Figure 2 The second patrol robot, which was relatively on the right in b, is blocking the intersection selected on the right.

In addition, the patrol step includes: taking a predetermined comparison image while driving; Comparing the comparison image taken in the previously performed patrol phase (hereinafter referred to as the 'first comparison image') with the comparison image taken in the current patrol phase (hereinafter referred to as the 'second comparison image'). step; When a difference is detected between the first comparison image and the second comparison image, transmitting the first comparison image and the second comparison image to an administrator terminal to request a confirmation procedure; When a difference confirmation message is received from the manager terminal, switching back to the patrol stage; And if a difference confirmation message is not received from the manager terminal, searching the road where the difference is located, the step of searching the road where the difference is located is: an intersection at the end of the road where the difference is located. selecting (hereinafter referred to as 'selected intersection'); A step of re-driving and patrolling the road where the difference is located without missing anything; And designate the selected intersection as the target point, and move the patrol robot (hereinafter referred to as 'second patrol robot') other than [the patrol robot that detected the difference (hereinafter referred to as 'first patrol robot')]. Including, the road where the difference is located may be surrounded by a second patrol robot.

The step of comparing the first comparison image and the second comparison image may be performed by applying a machine learning/deep learning algorithm. Since image processing and image similarity determination technology are areas of known technology, detailed descriptions will be omitted in the description of the present invention.

Through image comparison as above, it is possible to detect unidentified objects that are not recognized as animals through camouflage, etc., and even if the unidentified objects are not detected, the state in which the unidentified individuals have affected the facility (moving or damaging objects, etc.) By determining whether an unidentified person has invaded, you can indirectly check.

In the step of transmitting the first comparison image and the second comparison image to the administrator terminal and requesting a confirmation procedure, a still image at the time when the difference between the first comparison image and the second comparison image is derived is displayed on one screen of the administrator terminal. can do.

S100: Step of autonomous driving along a designated path and detecting surrounding objects (hereinafter referred to as ‘patrolling step’)
S110: Step of shooting a certain comparison image while driving
S120: Comparison image captured in the previously performed patrol phase (hereinafter referred to as 'first comparison image') and comparison image captured in the current patrol phase (hereinafter referred to as 'second comparison image') steps to compare
S130: When a difference is detected in the first comparison image and the second comparison image, sending the first comparison image and the second comparison image to the administrator terminal to request a confirmation procedure.
S131: When a difference confirmation message is received from the administrator terminal, switching back to the patrol stage
S132: If the difference confirmation message is not received from the administrator terminal, searching the road where the difference is located
S1321: Step of selecting the end intersection (hereinafter referred to as 'selected intersection') based on the road where the difference is located
S1322: Step of re-driving and patrolling the road where the difference is located without missing anything
S1323: Designate the selected intersection as the target point and move the patrol robot (hereinafter referred to as 'second patrol robot') other than [the patrol robot that detected the above difference (hereinafter referred to as 'first patrol robot')] step of ordering
S200: When an animal is detected while driving, a voice guidance is transmitted and the light is turned on.
S300: Step of transmitting animal detection information to the administrator terminal
S400: Step to confirm the identity of the animal
S410: When the animal's identity is confirmed, switching back to the patrol stage
S420: Step of transmitting an image of an unidentified animal (hereinafter referred to as 'unidentified person') to the administrator terminal to request a confirmation procedure
S430: When an identity verification message is received from the administrator terminal, switching back to the patrol stage
S440: If an identity verification message is not received from the administrator terminal, tracking the unidentified person.
S441: Step of selecting the end intersection (hereinafter referred to as ‘selected intersection’) based on the road where the unidentified person is located.
S442: Designate the selected intersection as the target point, and use patrol robots (hereinafter referred to as 'second patrol robots') other than [the patrol robot that detected the unidentified person above (hereinafter referred to as 'first patrol robots')]. moving steps
S500: After completing the designated route, move to the charging dock and proceed with charging.
S600: Among the roads connected to the selected intersection, [the road in the direction away from the road where the unidentified person is located] is designated as the target point, and patrol robots other than the 1st and 2nd patrol robots (hereinafter referred to as '3rd patrol robots') are designated as the target point. step of moving)
S700: When an unidentified person escapes from the nth patrol robot (n is one or more numbers from 1 to 3), move a patrol robot other than the nth patrol robot to the expected escape route predicted based on the escape direction of the unidentified person. Ordering stage (hereinafter referred to as ‘chasing stage’)
S800: Repeating the chase step until the unidentified person stops or a stop command is received from the administrator terminal.
R: road
U: Unidentified person
P1: 1st patrol robot
P2: 2nd patrol robot
P3: Third patrol robot
C: intersection
S: selected intersection

Claims (5)

A patrol stage that autonomously drives along a designated path and detects surrounding objects;
If an animal is detected while driving, transmitting a guidance voice;
Transmitting animal detection information to an administrator terminal;
confirming the identity of the animal; and
After completing the designated route, moving to a charging dock and charging, wherein the animal includes a person,
The step of confirming the identity of the animal includes, when the identity of the animal is confirmed, switching back to the patrol step,
The step of confirming the identity of the animal is, if the identity of the animal is not confirmed,
Requesting a verification process by transmitting an image of an unidentified animal, which is an unidentified animal, to an administrator terminal;
When an identity verification message is received from the manager terminal, switching back to the patrol stage; and
If the identity verification message is not received from the manager terminal, the step of tracking the unidentified person is performed,
The patrol steps are:
Taking a predetermined comparison image while driving;
Comparing the comparison image taken in the previously performed patrol phase (hereinafter referred to as the 'first comparison image') with the comparison image taken in the current patrol phase (hereinafter referred to as the 'second comparison image'). step;
When a difference is detected between the first comparison image and the second comparison image, transmitting the first comparison image and the second comparison image to an administrator terminal to request a confirmation procedure;
When a difference confirmation message is received from the manager terminal, switching back to the patrol stage; and
If a difference confirmation message is not received from the manager terminal, searching the road where the difference is located,
The steps to search the road where the difference is located are:
selecting an end intersection (hereinafter referred to as 'first selected intersection') based on the road where the difference is located;
Re-driving and patrolling the road where the difference is located without missing a beat; and
A step of designating the first selected intersection as a target point and moving a second patrol robot other than the first patrol robot that detected the difference,
A patrol robot control method further comprising surrounding the road where the difference is located with the second patrol robot. In claim 1,
Further comprising: distinguishing intersections and roads based on map information of the patrol area,
The steps for tracking the unidentified person are:
selecting an end intersection (hereinafter referred to as 'second selected intersection') based on the road where the unidentified person is located; and
The second selected intersection is designated as the target point, and patrol robots other than [the patrol robot that detected the unidentified person (hereinafter referred to as 'third patrol robot')] (hereinafter referred to as 'fourth patrol robot') Including the step of moving,
A patrol robot control method that surrounds the road where the unidentified person is located with a fourth patrol robot or guides the unidentified person to a safe area. In claim 2,
Among the roads connected to the selected intersection, the road in the direction away from the road where the unidentified person is located is designated as the target point, and patrol robots other than the 3rd patrol robot and the 4th patrol robot (hereinafter referred to as the '5th patrol robot') Including the step of moving,
A patrol robot control method that blocks the unidentified person's expected escape route with a fifth patrol robot. In claim 3,
When the unidentified person escapes from the third to fifth patrol robots, moving patrol robots other than the third to fifth patrol robots to an expected escape route predicted based on the escape direction of the unidentified person (hereinafter, ' (referred to as the ‘chase phase’); and
A patrol robot control method including; repeating the chase step until the unidentified person stops or a stop command is received from an administrator terminal. delete