KR102384444B1 - Domestic animal management robot and operation method thereof - Google Patents

Abstract

The livestock management robot according to an embodiment of the present application includes a recognition unit for recognizing a sheep-use object detected from a livestock image, an autonomous driving unit that travels along a rear region of the livestock image spaced apart from the sheep-use object by a predetermined distance, and the and an event controller configured to perform at least one of a first event operation for outputting a guide signal and a second event operation for touching the mole-use object, based on whether the mole-use object moves.

Description

Livestock management robot and its operation method

The present application relates to a livestock management robot and an operating method thereof.

Currently, in farms raising dairy cows, it is necessary to move the cows for milking the cows or cleaning the farm. This operation requires at least one skilled worker to move the cows and manage the cows that deviate or do not move.

Accordingly, the present application intends to provide a livestock management robot that can replace the operation of moving cows.

An object of the present application is to provide a livestock management robot capable of performing a task of moving livestock and an operating method thereof.

The livestock management robot according to an embodiment of the present application includes a recognition unit for recognizing a sheep-use object detected from a livestock image, an autonomous driving unit that travels along a rear region of the livestock image spaced apart from the sheep-use object by a predetermined distance, and the and an event controller configured to perform at least one of a first event operation for outputting a guide signal and a second event operation for touching the mole-use object, based on whether the mole-use object moves.

According to an embodiment, the recognition unit applies a camera unit for photographing the livestock image, an image extractor for extracting at least one image of interest from the livestock image, and a preset object recognition algorithm for the at least one image of interest, , an object detection unit for detecting the plurality of livestock as at least one stationary object and at least one moving object, and a distance measurement unit for measuring each separation distance between the at least one stationary object and the camera unit.

According to an embodiment, the object detection unit determines a specific object among the at least one stationary object as the mole-using object.

According to an embodiment, the event controller re-performs any one of the first and second event operations at a preset departure expected point based on the shortest moving distance to the destination according to the mole-using object.

According to an embodiment, the autonomous driving unit has an outer surface formed of an elastic material, a mobile device in which a battery and a motor for driving are disposed inside a lower side, is formed on an upper outer surface of the mobile device, and is controlled by the event controller. A touch device extending in length from the rear region to the mole-using object and an output device for outputting at least one of a manager voice, an alarm signal, a vibration, a sound of the same livestock, and an LED light according to the event control unit are included.

According to an embodiment, the moving device adjusts the lengths of the front wheel axle and the rear wheel axle according to the impact applied from the cow.

According to an embodiment, a location information collecting unit that collects robot location information through an RFID reader that communicates with RFID tags installed at predetermined distances along a movement path of the plurality of livestock, and the livestock image and the robot location information Further comprising a communication unit for providing to the manager terminal through a network, the RFID tags further include at least one tag for recognizing at least one livestock for recognizing the livestock.

According to an embodiment, the communication unit controls at least one of the recognition unit, the autonomous driving unit, and the event control unit based on manual operation information received from the manager terminal.

According to an embodiment, when a preset emergency situation object shape is detected from the at least one image of interest, the event control unit transmits an emergency message to a pre-registered terminal and travels to a pre-registered area where the manager is located to alert output a signal.

As an operation method of a livestock management robot according to an embodiment of the present application, the step of recognizing a sheep-use object from a livestock image by a recognition unit, the autonomous driving unit autonomously driving to a rear area of the livestock image spaced a certain distance from the sheep-use object and performing, by the event controller, a first event operation of outputting a guide signal in the rear region, the event controller touching the livestock corresponding to the sheep-using object based on whether the sheep-using object moves. performing a second event operation; and when the autonomous driving unit drives to a preset destination, moving to a preset charging area and automatically charging the vehicle.

According to an embodiment, the step of the location information collecting unit collecting robot location information through an RFID reader communicating with RFID tags installed at predetermined distances along the movement path of the plurality of livestock, and the communication unit collecting the livestock image and the robot The method further includes providing the location information to the manager terminal.

According to an embodiment, the step of recognizing the mole-using object may include: a camera unit photographing the plurality of livestock, an image extraction unit extracting at least one image of interest from the livestock image, and an object detection unit executing an object recognition algorithm classifying and detecting at least one still object and at least one moving object from the at least one image of interest using, by a distance measuring unit measuring each separation distance between the at least one still object and the camera unit and determining, by the object detection unit, a stationary object having the largest separation distance among the at least one stationary object as the mole-using object.

The livestock management robot system according to an embodiment of the present application collects robot location information from a plurality of RFID tags installed at predetermined distances along a movement path of a plurality of livestock, and collects livestock images for the plurality of livestock. A livestock management robot that takes pictures and a manager server that receives the livestock image and the robot location information and relays it to a manager terminal, wherein the livestock management robot recognizes the sheep-use object from the livestock image through an object detection algorithm A recognition unit, an autonomous driving unit that travels along a rear region separated by a predetermined distance from the mole-using object, a first event operation of outputting a guide signal based on whether the mole-using object moves, and touching the mole-using object and an event control unit that performs at least one of the second event operations, and a communication unit that controls at least one of the recognition unit, the autonomous driving unit, and the event control unit based on manual operation information received from the manager terminal.

According to an embodiment of the present application, the livestock management robot and its operating method can reduce the manpower according to the livestock moving operation, and prevent the risk of an accident of the operator due to the attack or collision of the livestock.

1 is a block diagram of a livestock management robot according to an embodiment of the present application.
FIG. 2 is a block diagram of the recognition unit of FIG. 1 .
3 is a diagram specifically illustrating an autonomous driving unit.
4 is a block diagram of a livestock management robot according to another embodiment of the present application.
5 is a block diagram of a livestock management robot system.
6 is an operation process according to an embodiment of the livestock management robot of FIG.
7 is an operation process for the recognition unit of FIG.

Hereinafter, embodiments of the present application will be described with reference to specific embodiments and the accompanying drawings. However, the embodiments of the present application may be modified in various other forms, and the scope of the present application is not limited to the embodiments described below. In addition, the embodiments of the present application are provided in order to more completely explain the present invention to those skilled in the art. Accordingly, the shapes and sizes of elements in the drawings may be exaggerated for clearer description, and elements indicated by the same reference numerals in the drawings are the same elements.

And in order to clearly explain the present application in the drawings, parts irrelevant to the description are omitted, and the thickness is enlarged to clearly express various layers and regions, and components having the same function within the scope of the same idea are referred to as the same. It is explained using symbols. Furthermore, throughout the specification, when a part "includes" a certain element, it means that other elements may be further included, rather than excluding other elements, unless otherwise stated.

1 is a block diagram of a livestock management robot 10 according to an embodiment of the present application.

Referring to FIG. 1 , the livestock management robot 10 may include a recognition unit 100 , an autonomous driving unit 200 , and an event control unit 300 .

First, the recognition unit 100 may recognize the mole use object detected from the livestock image. Here, the shepherding object may mean an image object corresponding to one of the objects that do not move but revolve around when moving livestock for cleaning or milking the breeding space.

Specifically, the recognition unit 100 detects at least one mole-using object from a preset livestock motion in the livestock image, and selects any one specified according to a criterion set by a manager among the at least one sheep-use object. can be recognized as

Next, the autonomous driving unit 200 may autonomously drive along the rear region of the livestock image spaced apart from the mole-using object by a predetermined distance. Here, the rear region may refer to a region located apart from the in-use object in the image by a predetermined number of pixels. That is, by driving along the rear region of the livestock image, the autonomous driving unit 200 may automatically follow the rear of the livestock corresponding to the sheep-using object.

Next, the event controller 300 may perform at least one of a first event operation of outputting a guide signal and a second event operation of touching the mole-use object based on whether the mole-use object moves.

Here, the guide signal may be at least one of a manager's voice signal, an alarm signal, vibration, a sound of the same livestock, and LED lighting for moving the livestock corresponding to the driving object.

Specifically, after performing the first event operation when located in the rear region of the mole-using object, the event controller 300 may determine whether to perform the second event operation based on whether the mole-using object moves.

For example, the event controller 300 may first perform the first event operation when it is located in the rear region of the mole-using object. In this case, when there is no movement of the mole use object for a predetermined time, the event controller 300 may perform a second event operation. Meanwhile, when the motion of the mole-using object occurs, the event controller 300 may not perform the second event operation.

Also, the event controller 300 may sequentially perform first and second event operations when located in the rear region of the mole-using object. In this case, the event controller 300 may change the operation order of the first and second events.

In addition, the event control unit 300 re-performs any one of the first and second event operations for each pre-collected departure point based on a preset shortest moving distance to the destination according to the mall use object, thereby It is possible to prevent the escape of an object in advance.

In addition, when a preset emergency situation object shape is detected from the livestock image, the event control unit 300 transmits an emergency message to a pre-registered terminal and at the same time drives to a pre-registered area where the manager is located and outputs an alarm signal. can For example, the preset emergency situation object shape may include a fire-shaped image object, a collapsed livestock-shaped image object, an image object in which a wild animal attacks livestock, and the like.

According to an embodiment of the present application, the livestock management robot 10 recognizes a sheep-use object that prevents or does not move in the operation of moving the livestock through the recognition unit 100 , and operates the autonomous driving unit 200 . Through this, it is possible to autonomously drive along the rear area of the driving object. In this case, the livestock management robot 10 may perform at least one event operation of outputting a guide signal and a touch operation to encourage movement of the sheep-using object through the event controller 300 . Accordingly, the livestock management robot 10 can reduce the additional manpower for moving the livestock, and prevent the risk of an accident of the operator due to the attack or collision of the livestock.

Hereinafter, in the present application, an example of a plurality of livestock will be described as a cow for better understanding, and the recognition unit 100 will be described in more detail with reference to FIG. 2 .

FIG. 2 is a block diagram of the recognition unit 100 of FIG. 1 .

1 and 2 , the recognition unit 100 may include a camera unit 110 , an image extraction unit 120 , an object detection unit 130 , and a distance measurement unit 140 .

First, the camera unit 110 may capture livestock images for the plurality of livestock 1_1 to 1_N.

Next, the image extraction unit 120 may extract at least one image of interest from the livestock image captured through the camera unit 110 in order to reduce the computational load. Here, the at least one image of interest may be at least one frame including an object for use.

Next, the object detector 130 may detect at least one still object and at least one moving object from at least one image of interest by using an object recognition algorithm.

Here, the still object may be an image object having a motionless cow shape, and the moving object may be an image object having a moving cow shape.

Specifically, the object detection unit 130 detects a plurality of objects from at least one image of interest, and classifies the plurality of objects into a static object corresponding to a preset still object shape and a moving object corresponding to a preset moving object shape among the plurality of objects. can be detected.

The object recognition algorithm may be at least one of a SIFT (Scale Invariant Feature Transform) algorithm, a HOG (Histogram of Oriented Gradient) algorithm, a Haar feature algorithm, a Ferns algorithm, a Local Binary Pattern (LBP) algorithm, and a Modified Census Transform (MCT) algorithm. can

Next, the distance measuring unit 140 may be disposed adjacent to the camera unit 110 , and may measure each separation distance between the at least one still object and the camera unit 110 .

The object detection unit 130 according to an embodiment may determine an object having the largest separation distance among at least one stationary object as the mole-using object. In this case, the object detection unit 130 may detect the rear area of the sheep-using object from the livestock image in real time and transmit it to the autonomous driving unit 200 .

This rear region includes location information to be driven, and in this case, the separation distance may be proportional to the movement distance between the driving object and the destination.

Also, the object detection unit 130 may determine a specific object determined according to a criterion preset by an administrator among the at least one stationary object as the mole use object.

3 is a diagram specifically illustrating the autonomous driving unit 200 .

Referring to FIG. 3 , the autonomous driving unit 200 may include a mobile device 210 , a touch device 220 , and an output device 230 .

First, the mobile device 210 may have an outer surface made of an elastic material so as to withstand a collision with a cow and a hind paw. In this case, the mobile device 210 may have a battery and a motor for driving along the rear region disposed inside the lower portion.

According to an embodiment, the moving device 210 may be configured to adjust the length L1 of the front wheel shaft and the rear wheel shaft according to the impact applied from the cow. Specifically, the moving device 210 may adjust the length L1 of the front wheel shaft and the rear wheel shaft by using a shock sensor and a pump, a cylinder, a valve, etc. for controlling the length L1 of the front wheel shaft and the rear wheel shaft. That is, the moving device 210 increases the length L1 of the front and rear axles through a pump, a cylinder, a valve, etc. based on the amount of impact detected by the shock sensor, so that the livestock management robot 10 can prevent the cow's hind paws. This can prevent falling accidents.

According to another embodiment, the moving device 210 may be formed to adjust the interaxial length L2 between the front wheel shaft and the rear wheel shaft according to the impact applied from the cow. That is, the moving device 210 may include a pump, a cylinder, a valve, etc. for controlling the interaxial length L2 between the shock sensor and the front and rear axles.

Also, the mobile device 210 may autonomously drive along the mole-using object based on the rear region of the livestock image received from the object detection unit 130 . For example, the mobile device 210 may follow the cow corresponding to the sheep-using object by automatically determining the driving direction with the rear region of the sheep-using object as the front.

Next, the touch device 220 is disposed on the upper outer surface of the mobile device 210 and can extend from the rear region to the body of the cow corresponding to the sheep-using object under the control of the event controller 300 . there is.

Next, the output device 230 includes a recorder, a speaker, and an LED for outputting at least one of a manager's voice, an alarm signal, vibration, and LED lighting for guiding the movement of the cow under the control of the event controller 300 . can do.

4 is a block diagram of the livestock management robot 11 according to another embodiment of the present application.

Referring to FIG. 4 , the livestock management robot 11 may include a recognition unit 100 , an autonomous driving unit 200 , an event control unit 300 , a location information collection unit 400 , and a communication unit 500 . Hereinafter, redundant descriptions of the recognition unit 100 , the autonomous driving unit 200 , and the event control unit 300 having the same reference numbers described with reference to FIGS. 1 to 3 will be omitted.

First, the location information collecting unit 400 may collect robot location information through an RFID reader that communicates with a plurality of RFID tags installed at predetermined locations along a movement path of a plurality of livestock. Here, the RFID tags may further include at least one livestock recognition tag for recognizing at least one livestock. In this case, the RFID reader may be disposed in the location information collecting unit 400 to collect location information corresponding to different identification information transmitted from a plurality of RFID tags.

Next, the communication unit 500 may provide the livestock image captured through the camera unit 110 and the robot position information collected through the location information collection unit 400 to the manager terminal 20 through the network in real time. .

The communication unit 500 according to an embodiment may control at least one of the recognition unit 100 , the autonomous driving unit 200 , and the event control unit 300 based on manual operation information received from the manager terminal 20 . .

Here, the manual operation information includes object selection information for arbitrarily selecting an object to be driven by a manager within the livestock image, driving operation information for manually operating the speed, direction, and stop of the autonomous driving unit 200, the first and event action information for manipulating the number of repetitions or order of the second event action.

5 is a block diagram of the livestock management robot system 1000 .

Referring to FIG. 5 , the livestock management robot system 1000 may include a livestock management robot 11 and a manager server 21 .

First, the livestock management robot 11 may collect robot location information by using the RFID reader 420 that communicates with the plurality of RFID tags 410_1 to 410_N. Here, the plurality of RFID tags 410_1 to 410_N may be installed at regular distances on a wall or a stanchion located along a movement path of a plurality of livestock.

Also, the livestock management robot 11 may use the camera 111 to capture livestock images for a plurality of livestock.

Next, the manager server 21 may receive the livestock image and robot location information from the livestock management robot 11 through the network and back it up to the server.

In addition, the manager server 21 may perform security authentication for the manager terminal 20 based on manager information received in response to a request for authentication information from the manager terminal 20 . In this case, the manager server 21 may transmit the livestock image captured through the camera unit 110 and the robot location information collected through the location information collecting unit 400 to the manager terminal 20 on which the security authentication is completed.

The livestock management robot 11 according to the embodiment determines whether to change the sheep-use object based on manual operation information received from the manager terminal 20 through the manager server 21, and determines whether the sheep-use object is changed. Accordingly, the driving direction can be changed and the driving speed can be increased for a certain period of time.

At this time, the livestock management robot 11 performs at least one of a first event operation of outputting a guide signal and a second event operation of touching the livestock body based on whether or not the sheep-using object is moved, which is changed according to the manual operation information. action can be performed.

5, the manager terminal 20 is described as remotely controlling the livestock management robot 11 through the manager server 21, but is not limited thereto, and as shown in FIG. 4, the manager terminal 20 It is also possible to directly control the livestock management robot 11 .

6 is an operation process according to an embodiment of the livestock management robot 11 of FIG.

Referring to FIGS. 1 and 6 , first, in step S110 , the recognition unit 100 may recognize a sheep-using object from the livestock image.

Then, in step S120 , the autonomous driving unit 200 may autonomously drive to the rear region within the livestock image spaced a predetermined distance from the mole-using object recognized through the recognition unit 100 .

At this time, in step S130, the location information collecting unit 400 receives the robot location information through the RFID reader 420 that communicates with the RFID tags 410_1 to 410_N installed at predetermined distances along the movement path of the plurality of livestock. can be collected

Then, in step S140 , the communication unit 500 may provide the livestock image and robot location information captured through the camera unit 110 to the manager terminal 20 or the manager server 21 .

In this case, in step S150 , the event controller 300 may perform a first event operation of outputting a guide signal in the direction of the mole-using object in the rear region.

Also, in step S160 , the event controller 300 may perform a second event operation of touching the livestock corresponding to the sheep-using object based on whether the sheep-using object moves in the livestock image. For example, when the motion of the mole-using object does not occur, the event controller 300 may perform a second event operation.

Thereafter, in step S170 , when driving to a preset destination is completed, the autonomous driving unit 200 may travel to a preset charging area and automatically charge the vehicle.

FIG. 7 is an operation process for the recognition unit 100 of FIG. 2 .

Referring to FIGS. 2 and 7 , first, in step S210 , the camera unit 110 may capture livestock images for a plurality of livestock 1_1 to 1_N.

Then, in step S220 , the image extraction unit 120 may extract at least one image of interest from the livestock image captured through the camera unit 110 . Here, the at least one image of interest may be at least one frame including an object for use.

In this case, in step S230 , the object detection unit 130 may classify and detect the plurality of livestock into at least one stationary object and a moving object by applying a preset object recognition algorithm to at least one image of interest.

Then, in step S240 , the distance measurement unit 140 may measure each separation distance between the at least one still object detected by the object detection unit 130 and the camera unit 110 .

Thereafter, in step S250 , the object detection unit 130 may determine a stationary object having the largest separation distance among at least one stationary object as an object for driving.

Although the present application has been described with reference to an embodiment shown in the drawings, this is merely exemplary, and those of ordinary skill in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Accordingly, the true technical protection scope of the present application should be determined by the technical spirit of the appended claims.

10, 11, 12: Livestock management robot
100: recognition unit
200: autonomous driving unit
300: event control unit
1000: livestock management robot system

Claims (13)

a recognition unit for recognizing a mole-using object detected from a livestock image;
an autonomous driving unit that runs along a rear region of the livestock image spaced a certain distance from the driving object; and
and an event controller performing a first event operation of outputting a guide signal in the rear region and performing a second event operation of touching a livestock corresponding to the sheep-use object based on whether the sheep-use object moves A livestock management robot. According to claim 1,
The recognition unit includes: a camera unit for photographing the livestock image;
an image extraction unit for extracting at least one image of interest from the livestock image;
an object detection unit configured to detect at least one still object and at least one moving object from the at least one image of interest by using a preset object recognition algorithm; and
A livestock management robot comprising a distance measuring unit for measuring each separation distance between the at least one stationary object and the camera unit. 3. The method of claim 2,
The object detection unit determines a specific object among the at least one stationary object as the sheep-use object, the livestock management robot. According to claim 1,
The event controller re-performs any one of the first and second event operations at a preset departure expected point based on the shortest moving distance to the destination according to the mole-using object. According to claim 1,
The autonomous driving unit has an outer surface formed of an elastic material, and a moving device in which a battery and a motor for driving are disposed inside the lower side;
a touch device formed on an upper outer surface of the mobile device and extending from the rear region to the mole-using object under the control of the event controller; and
An output device for outputting at least one of a manager's voice, an alarm signal, a vibration, and an LED light according to the event control unit, the livestock management robot. 6. The method of claim 5,
The moving device is a livestock management robot that adjusts the length of the front axle and the rear axle according to the impact applied from the cow. According to claim 1,
a location information collecting unit for collecting robot location information through an RFID reader that communicates with RFID tags installed at predetermined distances along a movement path of a plurality of livestock and one or more object recognition-capable tags; and
Further comprising a communication unit for providing the livestock image and the robot position information to the manager terminal through a network,
The RFID tags further include at least one livestock recognition tag for recognizing at least one livestock. 8. The method of claim 7,
The communication unit, based on the manual operation information received from the manager terminal, to control at least one of the recognition unit, the autonomous driving unit, and the event control unit, livestock management robot. 3. The method of claim 2,
The event control unit transmits an emergency message to a pre-registered terminal when a preset emergency object shape is detected from the at least one image of interest, and outputs an alert signal by driving to a pre-registered area where the manager is located, Livestock management robot. A method of operating a livestock management robot, comprising:
Recognizing, by a recognition unit, a sheepish object from the livestock image;
autonomously driving, by an autonomous driving unit, to a rear region of the livestock image spaced a certain distance from the driving object; and
performing, by the event controller, a first event operation for outputting a guide signal from the rear region;
performing, by the event controller, a second event operation of touching a livestock corresponding to the mole-using object based on whether the mole-using object moves; and
When the autonomous driving unit travels to a preset destination, moving to a preset charging area and automatically charging the car, the operating method of the livestock management robot. 11. The method of claim 10,
collecting, by a location information collecting unit, robot location information through an RFID reader communicating with RFID tags installed at predetermined distances along a movement path of a plurality of livestock; and
The method of operation of the livestock management robot further comprising the step of a communication unit providing the livestock image and the robot position information to a manager terminal. 11. The method of claim 10,
The step of recognizing the object for use may include: photographing a plurality of livestock by a camera unit;
extracting, by an image extraction unit, at least one image of interest from the livestock image;
detecting, by an object detection unit, at least one still object and at least one moving object from the at least one image of interest using an object recognition algorithm;
measuring, by a distance measuring unit, each separation distance between the at least one still object and the camera unit; and
and determining, by the object detection unit, a stationary object having the largest separation distance among the at least one stationary object as the sheep-use object. a livestock management robot that collects robot location information from a plurality of RFID tags installed at predetermined distances along a movement path of a plurality of livestock and captures livestock images for the plurality of livestock; and
and a manager server that receives the livestock image and the robot location information and relays it to a manager terminal,
The livestock management robot,
a recognition unit for recognizing a mole-using object from the livestock image through an object detection algorithm;
an autonomous driving unit that drives along a rear area spaced apart from the driving object by a predetermined distance;
an event controller performing a first event operation of outputting a guide signal from the rear region and performing a second event operation of touching a livestock corresponding to the sheep-use object based on whether the sheep-use object moves; and
Based on the manual operation information received from the manager terminal, comprising a communication unit for controlling at least one of the recognition unit, the autonomous driving unit, and the event control unit, livestock management robot system.

Images