KR102593390B1 - A device for controlling drone landing and a method thereof - Google Patents

Abstract

A drone landing control device 100 is provided. The drone landing control device 100 includes a communication unit 120 configured to receive captured images from the drone 200; an object detection unit 130 configured to detect a predetermined object in the received captured image; And it may include a control unit 110 configured to control the landing of the drone 200 depending on the presence or absence of the detected object.

Description

Drone landing control device and method {A DEVICE FOR CONTROLLING DRONE LANDING AND A METHOD THEREOF}

The present invention relates to a drone landing control device and method. More specifically, when a drone lands, it determines whether a pre-designated object exists at the landing point based on artificial intelligence (AI) from the captured image of a camera mounted on the drone. An invention relating to a drone landing control device and method that can implement a safer drone landing by determining and stopping the landing of the drone if a pre-designated object exists at the landing point and controlling the drone to land only when the object does not exist. am.

A drone, also commonly referred to as an unmanned aerial vehicle (UAV), is an unmanned aerial vehicle that can fly and be controlled by radio wave guidance without a pilot. It was initially mainly used for military purposes, but has various advantages such as its simplicity, speed, and economic feasibility. Due to this advantage, it has recently been used in a wide variety of fields, including logistics delivery, disaster relief, broadcasting, leisure, forest management, facility management, and sports, in addition to military use.

It is very important for these drones to land safely at the intended point to perform their mission. Since a pilot cannot board a drone, it is necessary to safely control the landing of the drone to prevent it from falling at the landing point or colliding with various objects on the ground. There is.

If the pilot lands the drone within an area that can be visually determined, the pilot can directly visually check the safety status of the landing point. However, if the landing point is outside the area that can be visually determined by the pilot, or the drone flies autonomously, In the case of flight, a new drone control method is required to land the drone more safely at the landing point.

Therefore, in order to overcome the limitations of existing drone control methods, the demand for a new drone landing control device and method to control the landing of a drone more safely is gradually increasing in the art.

The present invention was created to solve the above problems. The present invention detects a pre-designated object based on a drone captured image and implements the landing of the drone based on the results, thereby making it possible to land the drone at the landing point more safely. The purpose is to provide a drone landing control device and method that allows the drone to land.

Additionally, the purpose of the present invention is to provide a drone landing control device and method that can further improve the efficiency of object detection by setting a detection area for detecting objects within a shooting area.

In addition, the present invention enables active response to the landing point by setting the detection area variably according to the shape, location, and surrounding hazardous facilities of the landing point, and thereby improves the accuracy and efficiency of object detection. The purpose is to provide a control device and method.

Additionally, the purpose of the present invention is to provide a landing control device and method that allows a drone to land more safely at a landing point by transmitting a predetermined evacuation broadcast when the detected object is a person.

In addition, the present invention can further improve the accuracy and efficiency of object detection by variably setting the location, size, and number of detection areas based on the location and number of objects detected as the drone descends from the detection start altitude to the ground. The purpose is to provide a capable landing control device and method.

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

A drone landing control device according to an embodiment of the present invention for solving the above technical problem includes a communication unit configured to receive captured images from a drone; an object detection unit configured to detect a predetermined object in the received captured image; and a control unit configured to control landing of the drone depending on the presence or absence of the detected object.

Additionally, the object detection unit may include a detection area setting unit configured to set a detection area in the received captured image.

In addition, the object detection unit may further include an object detection unit configured to detect the predetermined object in the set detection area.

In addition, the object detection unit may be further configured to detect the pre-designated object in the set detection area based on AI learning on a floor plan image of the pre-designated object.

In addition, the object detection unit may further include an object display unit configured to display the detected object when the pre-designated object is detected.

In addition, the detection area setting unit may be further configured to set the shape and size of the detection area based on at least one of the shape and location of the drone landing site and surrounding dangerous facilities.

In addition, a drone landing control system according to another embodiment of the present invention for solving the above technical problem includes the drone landing control device; And it may include a drone that is communicatively coupled to the drone landing control device.

In addition, a drone landing control method according to an additional embodiment of the present invention for solving the above technical problem includes receiving a captured image from a drone; Detecting a pre-designated object in the received captured image; And it may include controlling the landing of the drone depending on the presence or absence of the detected object.

According to the drone landing control device and method according to an embodiment of the present invention, a pre-designated object is detected based on a drone captured image and the drone is landed based on the result, thereby landing the drone more safely at the landing point. You can do it.

In addition, according to the drone landing control apparatus and method according to an embodiment of the present invention, the efficiency of object detection can be further improved by setting a detection area for detecting objects within the shooting area.

In addition, according to the drone landing control device and method according to an embodiment of the present invention, active response to the landing point is possible by variably setting the detection area according to the shape, location, and surrounding dangerous facilities of the landing point, Accordingly, the accuracy and efficiency of object detection can be further improved.

In addition, according to the drone landing control apparatus and method according to an embodiment of the present invention, when the detected object is a person, a predetermined evacuation broadcast is transmitted, so that the drone can land at the landing point more safely.

In addition, according to the drone landing control device and method according to an embodiment of the present invention, as the drone descends from the detection start altitude to the ground, the location, size, number, etc. of the detection area are changed based on the location and number of detected objects. By variably setting , the accuracy and efficiency of object detection can be further improved.

In order to more fully understand the drawings cited in the detailed description of the present invention, a brief description of each drawing is provided.
Figure 1 is a schematic block diagram of a drone landing control system 1000 according to an embodiment of the present invention.
Figure 2 is a detailed block diagram of the object detection unit 130 according to an embodiment of the present invention.
Figure 3 is a schematic flowchart of a drone landing control method (S300) according to an embodiment of the present invention.
4A and 4B are schematic diagrams illustrating a photographing area of a drone and an object detection area according to an embodiment of the present invention.
Figure 5 is a schematic diagram illustrating a detection area according to an embodiment of the present invention.
Figure 6 is a schematic diagram illustrating improvement in object detection accuracy in the detection area as the drone descends according to an embodiment of the present invention.
Figure 7 is a schematic diagram illustrating variable setting (or movement) of a detection area according to an embodiment of the present invention.

Hereinafter, embodiments according to the present invention will be described with reference to the attached drawings. When adding reference signs to components in each drawing, it should be noted that the same components are given the same reference numerals as much as possible even if they are shown in different drawings. Additionally, when describing embodiments of the present invention, if detailed descriptions of related known configurations or functions are judged to impede understanding of the embodiments of the present invention, the detailed descriptions will be omitted. In addition, embodiments of the present invention will be described below, but the technical idea of the present invention is not limited or limited thereto and may be modified and implemented in various ways by those skilled in the art.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected," but also the case where it is "indirectly connected" with another element in between. . Throughout the specification, when a part is said to “include” a certain element, this means that it may further include other elements rather than excluding other elements, unless specifically stated to the contrary. Additionally, in describing the components of the embodiment of the present invention, terms such as first, second, A, B, (a), and (b) may be used. These terms are only used to distinguish the component from other components, and the nature, sequence, or order of the component is not limited by the term.

Figure 1 is a schematic block diagram of a drone landing control system 1000 according to an embodiment of the present invention.

As shown, the drone landing control system 1000 according to an embodiment of the present invention may be composed of a drone landing control device 100, a drone 200, a communication network 300, etc. For reference, although a single drone 200 is shown as an example in FIG. 1, a plurality of drones 200 are communicatively coupled to the drone landing control device 100 through the communication network 300, so that each drone landing control is performed. It is clear that a safe landing can be implemented by means of the device 100.

When the drone 200 lands, the drone landing control device 100 determines whether a pre-designated object exists at the landing point based on artificial intelligence (AI) from the captured image of the camera 210 mounted on the drone 200. If a pre-designated object exists at the landing point, the drone can be stopped from landing and the drone can be controlled to land only if the object does not exist. To implement this, as shown in FIG. 1, the drone landing control device 100 according to an embodiment of the present invention includes a control unit 110, a communication unit 120, an object detection unit 130, and a storage unit. It may include a unit 140, etc., and depending on various implementations, the drone landing control device 100 may be referred to by another name such as ground control PC.

For reference, the components 110, 120, 130, and 140 of the drone landing control device 100 shown in FIG. 1 perform the operations, functions, etc. of the drone landing control device 100 according to an embodiment of the present invention. They are merely exemplary components for explanation, and therefore, the drone landing control device 100 according to an embodiment of the present invention includes other components (e.g., 110, 120, 130, and 140) other than the illustrated components. It is clear that it may further include a power supply unit, a notification unit, a display unit, etc.).

The control unit 110 may be configured to comprehensively control the operations and functions of the drone landing control device 100. For example, the control unit 110 stops the landing of the drone 200 when a pre-designated object exists at the landing point where the drone 200 wants to land, and only when the pre-designated object does not exist at the landing point. The flight of the drone 200, that is, the landing flight, can be controlled so that the drone 200 lands. In other words, the control unit 110 may be configured to control the landing of the drone 200 depending on whether or not a pre-designated object is detected at the landing point.

Additionally, according to one embodiment of the present invention, the detection start altitude at which object detection begins (e.g., 20 m, etc.) may be defined in advance, and detection begins as the drone 200 descends from the upper part of the landing point. When the altitude is reached, the control unit 110 can control the operation of the object detection unit 130 to be activated.

For reference, the control unit 110 may be implemented as a processor, controller, micro-processor, micro-controller, etc. That is, the control unit 800 may operate at least two of the components included in the drone landing control device 100 in combination with each other in order to run the application program.

The communication unit 120 may implement wireless communication between the drone landing control device 100 and the drone 200. For example, the drone landing control device 100 may be configured to receive captured images from the drone 200 through the communication unit 120.

For reference, the communication unit 120 includes LAN, Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), Wireless Broadband Internet (WiBro), Radio Frequency (RF) communication, Wireless LAN, It may be configured to communicate with the drone 200 through Wi-Fi (Wireless Fidelity), NFC (Near Field Communication), Bluetooth, infrared communication, etc., and more preferably, the communication unit 120 is configured to communicate with the drone 200 through an RF communication network, Wifi communication network, LTE 4G/ It can be configured to communicate with the drone 200 through a 5G communication network, IOT communication network, etc.

For reference, considering that the drone 200 is relatively vulnerable to security, has a high risk of hacking, and there is a risk that hacked drones may be exploited for crime and terrorism, there is a need to further strengthen the security of transmitting and receiving drone data. To this end, communication between the drone 200 and the drone landing control device 100 may be KCMVP (Korea Crypto-graphical Module Validation Program) encrypted secure communication, which may be implemented based on VPN or module.

The object detection unit 130 may be configured to detect a pre-designated object in the captured image received from the communication unit 120. More specifically, the object detection unit 130 according to an embodiment of the present invention analyzes the captured image captured by the drone 200 with artificial intelligence (AI), and according to the results, a pre-designated object ( It is possible to determine whether an object exists.

A detailed block diagram of the object detection unit 130 configured to detect a pre-designated object is shown in FIG. 2. As shown in Figure 2, the object detection unit 130 according to an embodiment of the present invention includes a detection area setting unit 131, an object designation unit 132, an object detection unit 133, and an object display It may be composed of a unit 134, etc.

The detection area setting unit 131 may be configured to set a detection area for object detection in a captured image captured by the drone 200. The ground image captured by the drone 200 may include a relatively large area, and if object detection is performed on the entire area of such a large area, the object detection efficiency and accuracy will inevitably deteriorate. There is a problem.

Therefore, the detection area setting unit 131 according to an embodiment of the present invention can set a detection area for object detection in the captured image captured by the drone 200 to increase the efficiency and accuracy of object detection, where The detection area setting unit 131 may be further configured to set the shape and size of the detection area based on the shape and location of the drone landing site, nearby hazardous facilities, etc. The relationship between the photographing area and the detection area and the setting of the detection area according to the present invention will be described in more detail in FIGS. 4 and 5 below.

The object designation unit 132 may be configured to designate an object to be detected when the drone 200 lands. For example, objects defined in the present invention are dangerous objects that may collide with the drone 200 when it lands or that may interfere with the safe landing of the drone 200, such as people, dogs, cats, etc. It may include objects such as cars and motorcycles, and the object designation unit 132 according to an embodiment of the present invention can designate an object to be detected among these objects.

For reference, the object designated by the object designation unit 132 may be designated in advance by the user (or operator), or may be automatically designated based on the type, location, and surrounding hazardous facilities of the drone landing site. For example, in the case of the latter object automatic designation, if the landing point of the drone 200 is the rooftop of a city building, people and rooftop facilities (e.g., antennas, outdoor air conditioner units, flower pots, etc.) are automatically selected as objects to be detected. Alternatively, if the landing point of the drone 200 is an alpine plain area, forest resources such as trees may be automatically designated as an object to be detected.

Accordingly, if a person is manually and/or automatically designated as an object to be detected by the object designation unit 132, the object detection unit 130 can detect the person from the detection area and the control unit 110 Can control the flight of the drone 200 so that it lands at the landing point only when no person is detected. Similarly, if people and cars are manually and/or automatically designated as objects to be detected by the object designation unit 132, the object detection unit 130 can detect people and cars from the detection area. The control unit 110 can control the flight of the drone 200 to land at the landing point only when people and cars are not detected.

The object detection unit 133 may be configured to detect an object pre-designated by the object designation unit 132 in the detection area set by the detection area setting unit 131. For example, to detect an object, the object detection unit 133 may analyze the detection area among the captured images received from the drone 200 based on artificial intelligence (AI), and as a result of the analysis, It can be determined whether pre-designated objects such as people, dogs, cats, or cars exist within the detection area.

Here, the drone 200 has no choice but to photograph objects on the ground from top to bottom, and in consideration of the peculiarities of such drone photography, the object detection unit 133 according to an embodiment of the present invention provides a plan view of a pre-designated object. It may be further configured to detect a pre-designated object within a detection area set based on AI learning for the image, and for this purpose, the object detection unit 133 may be additionally equipped with a deep learning-based AI learning module.

For example, video data taken from top to bottom of a person at various angles can be used as AI learning data, and for such AI learning, a MNIST (Modified National Institute of Standards and Technology) data set is built and stored in the storage unit (140). ), and this MNIST data set may include image data that captures a person from top to bottom.

In addition, the MNIST data set for AI learning that is being built is data with various angles and sizes, such as people photographed vertically, such as people standing, crouching, not crouching, lying down, and prone. can be provided, and in addition, various types of data holding various belongings (e.g., bags, hats, cell phones, etc.) can be additionally provided, thereby increasing the detection accuracy of people present at the landing site. It can be improved considerably.

As a result of this AI learning, the object detection unit 133 can detect a pre-designated object from a set detection area, and based on the detection result, the control unit 110 can control the landing of the drone 200 at the landing point. You can.

Object display unit 134 may be configured to display detected objects. The object display unit 134 may display text such as 'person' or 'dog' visually (for example, processing a bounding box on the detected object) and/or audibly (for example, 'Person detected' or 'Dog detected' can be displayed by sound) and can be configured to display the detected object, and in the case of visual display, through a display unit (not shown) provided in the object display unit 134. Detected objects may be displayed visually.

Accordingly, the drone operator can monitor the landing control of the drone 200 while checking the object displayed on the display unit, and accordingly, if the object detection by the object detection unit 133 is partially inaccurate or has an error. In some cases (for example, when detecting a black rock as a dog, etc.), landing control of the drone 200 may be implemented directly while checking the screen of the display unit.

For reference, display units for visually displaying detected objects include Liquid Crystal Display (LCD), Thin Film Transistor-Liquid Crystal Display (TFT LCD), and Organic Light-Emitting Diode (Organic Light-Emitting Diode). Diode, OLED), flexible display, 3D display, etc.

In this regard, according to a further embodiment of the present invention, information about the detected object may be additionally transmitted to the user terminal (not shown). For example, if the drone operator is absent from the drone landing control device 100, information about the detected object may be transmitted to a user terminal owned by the drone operator, and this user terminal may use the drone 200 directly or It is configured to control the drone 200 through the landing control device 100, and accordingly, the safe landing of the drone 200 at the landing point can be monitored and controlled while looking at the screen of the user terminal.

The storage unit 140 may store arbitrary data, signals, etc. for landing of the drone 200. For example, the storage unit 140 includes identification information about the drone 200, flight information, landing information, image data captured by the drone 200, information about the detection area, information about the detection start altitude, and detection information. Information about objects, training image data for object detection, etc. may be stored. In particular, you can set the detection area, detection start altitude (e.g., 20 m, etc.), detection object type (e.g., person, dog, cat, car, etc.), and detection end altitude (e.g., 3 m, etc.). A setting file may be created and stored in the storage unit 140, and upon setting the detection area by the detection area setting unit 131 and pre-specifying the object by the object designation unit 132. Information stored in the storage unit 140 may be utilized.

For reference, as known to those skilled in the art, the storage unit 140 includes a hard disk drive (HDD), read only memory (ROM), random access memory (RAM), electrically erasable and programmable read only memory (EEPROM), and flash. Various types of storage devices capable of inputting and outputting information, such as flash memory, Compact Flash (CF) card, Secure Digital (SD) card, Smart Media (SM) card, Multimedia (MMC) card, or Memory Stick. It may be implemented as, and may be provided inside the drone landing control device 100 as shown in FIG. 1 or may be provided in a separate external device. Alternatively, the storage unit 140 may be replaced with web storage that performs a storage function on the Internet.

The drone 200 may include any unmanned aerial vehicle capable of flight and control by radio wave guidance, and may be configured to fly autonomously or to fly manually by a drone controller on the ground (or a user terminal connected thereto). . For example, in the case of autonomous flight, the drone 200 may be configured to fly autonomously toward a landing point based on GPS, and may be communicatively coupled with the drone landing control device 100 through the communication network 300 to capture images upon landing. The image can be transmitted to the drone landing control device 100, and the safe landing of the drone 200 at the landing point can be controlled by detecting a pre-designated object in the detection area by the drone landing control device 100.

The drone 200 according to an embodiment of the present invention may be equipped with a camera 210 mounted toward the ground and configured to photograph the ground, and the MC (220, Mission Computer) is configured to shoot the ground from the camera 210. It is possible to receive captured video and transmit it to the drone landing control device 100 on the ground. For reference, the camera 210 mounted on the drone 200 may be implemented as a thermal image and high-definition (e.g., 20 million pixel or higher) camera, and accordingly transmits high-resolution captured images to the drone landing control device 100. By transmitting to , the accuracy and speed of detecting objects such as people can be further improved.

In addition, the drone 200 may be further equipped with sensors such as a GPS sensor, an altitude sensor, and an acceleration sensor, a driving unit such as a motor, a lighting unit, a storage unit, an Electronic Speed Control (ESC), and a Flight Controller (FC). However, the description of other components that are less directly related to the landing control of the drone 200 will be omitted in this paragraph.

Figure 3 is a schematic flowchart of a drone landing control method (S300) according to an embodiment of the present invention.

First, the communication unit 120 can receive captured images from the drone 200 (S310). The control unit 110 can control the communication unit 120 to receive all captured images captured by the camera 210 mounted on the drone 200 from takeoff to landing of the drone 200, or the drone 200 ) may control the communication unit 120 to receive captured images of the drone 200 from the time it reaches a preset altitude (e.g., 20 m, etc.), and an interface that allows the user to select such image reception mode. may be additionally provided and provided.

As the drone 200 descends from the top of the landing point, the control unit 110 may determine whether the drone 200 has reached a preset detection start altitude (for example, 20 m, etc.) (S320) . As a result of the determination, if the drone 200 has not yet reached the detection start altitude (for example, the drone is located 30 m above the ground), the control unit 110 controls the drone (200) until the drone 200 reaches the detection start altitude. 200) can be controlled to fly further downward.

As a result of the determination, if it is determined that the drone 200 has reached the detection start altitude, the control unit 110 may activate the operation of the object detection unit 130, and accordingly, the detection area setting unit 131 may set the detection area to a predefined level. The detection area (or range) can be activated or the detection area can be set in the captured image captured by the camera 210 of the drone 200 (S330).

Compared to images captured by a drone, the detection area can be set to have a smaller area, and thus the accuracy and efficiency of object detection can be further improved. Schematic diagrams illustrating the photographing area 11 of the drone 200 and the detection area 12 for detecting objects according to an embodiment of the present invention are shown in FIGS. 4 and 5.

FIGS. 4A and 4B are schematic diagrams illustrating the drone's imaging area 11 and the object detection area 12 according to an embodiment of the present invention, and FIG. 5 is a detection area according to an embodiment of the present invention. This is a schematic diagram to explain.

As shown in FIG. 4A, the drone 200 can photograph the ground at a predetermined angle using the camera 210, and the image captured in this way can be defined as the photographing area 11. The detection area 12 defined for object detection is defined at an angle smaller than the angle of the shooting area 11 based on the camera 210, that is, compared with the shooting area 11 in four directions, up, down, left, and right. So it can be defined as a smaller area.

Objects can be detected in all of the shooting area 11 captured by the camera 210 installed on the drone 200, but the detection area set by the detection area setting unit 131 by setting a specific area ( In 12), the efficiency and accuracy of object detection can be significantly improved by intensively detecting pre-specified objects. In other words, object detection is not performed for areas other than the detection area 12 in the shooting area 11, and therefore the detection area 12 where it is necessary to confirm the presence of an object at the landing point of the drone is optimally selected. Setting is the most important thing.

Additionally, although not explicitly shown in FIG. 3, a step of the user (i.e., the operator) confirming the detection area 12 set by the detection area setting unit 131 may be additionally implemented. For example, the detection area 12 set by the detection area setting unit 131 can be confirmed by the drone operator through a display provided in the drone landing control device 100 or through a user terminal (not shown), In response, the operator may check the set detection area 12 or request adjustment of the location, size, etc. of the detection area 12.

In addition, as shown in FIG. 5, the detection area 12 set by the detection area setting unit 131 is circular (see (a) of FIG. 5), square (see (b) of FIG. 5), and polygonal. It can be set to have various shapes, such as (see (c) in FIG. 5), where the detection area setting unit 131 according to an embodiment of the present invention is based on the shape, location, and surrounding dangerous facilities of the drone landing site, etc. The shape and size of the detection area 12 can be set.

For example, as shown by way of example in (a) of FIG. 5, when the landing point is a circular roundabout, the detection area setting unit 131 sets a shooting area captured by the camera 210 of the drone 200. In (11), it is detected that the landing point of the drone is a circular roundabout, and the detection area 12 for object detection can be set to circular accordingly.

Here, more preferably, the center point of the set detection area 12 may correspond or substantially correspond to the landing point of the drone 200, and for this purpose, the landing point of the drone 200 in setting the detection area 12 GPS information about the point may be utilized.

In this way, the detection area 12 is most optimally set and applied variably based on the type, location, and surrounding dangerous facilities of the drone landing site, so that the detection efficiency and accuracy for pre-designated objects existing within the detection area 12 are applied. can be further improved.

There may be various ways to set the detection area 12 of various shapes. For example, in the case of the circular detection area 12 shown in (a) of Figure 5, the center point and radius (or diameter) are defined. In the case of the rectangular detection area 12 shown in (b) of FIG. 5, the coordinates of one vertex and the coordinates of the diagonal vertices can be defined, respectively, and the polygonal detection area shown in (c) of FIG. 5 ( In case 12), the coordinates of each vertex constituting the polygon can be defined.

When the detection area 12 suitable for the landing point is optimally set, the object detection unit 133 can detect a pre-designated object in the set detection area 12 (S340). More specifically, the object detection unit 133 may be configured to detect an object pre-designated by the object designation unit 132 in the detection area set by the detection area setting unit 131. For example, to detect an object, the object detection unit 133 may analyze the detection area among the captured images received from the drone 200 based on artificial intelligence (AI), and as a result of the analysis, the object detection unit 133 may analyze the detection area within the detection area. It can be determined whether pre-designated objects such as people, dogs, cats, cars, etc. exist. Here, the object detection unit 133 according to an embodiment of the present invention may be further configured to detect a pre-designated object within a detection area set based on AI learning for the floor plan image of the pre-designated object. It's like a bar.

As a result of this AI learning, the object detection unit 133 can detect a pre-designated object from a set detection area, and based on the detection result, the control unit 110 can control the landing of the drone 200 at the landing point. You can.

That is, the control unit 110 may determine whether an object is detected in the detection area 12 by the object detection unit 133 (S350). If an object, for example, a person, is detected in the detection area 12, the control unit 110 may control the flight of the drone 200 to stop the landing of the drone 200 (S360) and the object display unit ( 134) can display the detected object (S370). The detected object is displayed visually and/or audibly by the object display unit 134 so that the operator can recognize the detected object.

Because it is dangerous to land a drone when an object exists at the landing point, if the detected object is a person, the control unit 110 sends an evacuation announcement through the speaker of the drone 200 to evacuate the detected person to a safe area. (not shown), etc., can be controlled to output (S380), and in some cases, the voice of a person detected at the landing point can be received through the microphone (not shown) of the drone 200.

If the object is not detected, or if the object is not detected because a person moves to a safe area through an evacuation announcement, the control unit 110 may land the drone at the landing point (S390).

Here, just as the detection start altitude (e.g., 20 m, etc.) for starting object detection for the landing point is preset, according to a further embodiment of the present invention, the detection end altitude for ending object detection for the landing point is It can also be preset. For example, if the detection end altitude is set to 3m, object detection begins when the drone 200 enters an altitude of 20m, and if no object is detected, the drone 200 is descended in sequence, and finally, even at an altitude of 3m. If there is no object detection, the control unit 110 may control the flight of the drone 200 so that the drone 200 lands at the landing point.

Figure 6 is a schematic diagram illustrating improvement in object detection accuracy in the detection area as the drone descends according to an embodiment of the present invention.

As already described, the detection start altitude of the drone for object detection (e.g., 20 m, etc.) may be set in advance, and when the drone 200 detects that the detection start altitude has been reached, the control unit 110 detects the object. The operation of the detection unit 130 can be activated.

Then, as the drone 200 sequentially descends from the detection start altitude, the object detection unit 133 detects a pre-designated object within the detection area 12. As shown in FIG. 6, the drone detects as it descends. The pixel size of area 12 may remain the same. For example, if the length of one side of the rectangular detection area is maintained at 300 pixels regardless of the drone altitude, the actual terrain distance contained in the detection area 12 may gradually decrease as the drone 200 descends, that is, the drone As this descends, the actual detection area decreases and the clarity of the image increases. As a result, the object becomes larger and clearer, eliminating unnecessary landing stops that may occur due to object detection errors and improving the accuracy of object detection. It can be.

In this regard, according to a further embodiment of the present invention, the pixel size of the detection area 12 may be implemented to change, for example, increase, as the drone 200 descends. For example, at the detection start altitude (e.g., 20 m, etc.), the object was detected to be completely contained within the detection area 12, but as the drone 200 descends and the distance to the object increases, the detection area 12 If the object is not completely contained within (e.g., the object straddles the border of the detection area 12, etc.), the detection area setting unit 131 determines whether the detected object is detected within the detection area 12. It can be judged that

In addition, in Figure 6, a configuration in which the shape of the detection area 12 remains the same as the drone 200 descends is described as an example, but according to an additional embodiment of the present invention, the drone 200 descends with more precision. The detection area setting unit 131 may be further configured to change the shape of the detection area 12 based on the observed shape, location, and surrounding hazardous facilities of the landing site. For example, at an altitude of 20 m, the detection area 12 may be set to a square shape, and as the drone 200 descends, the detection area 12 may be set to a circular shape at an altitude of 10 m.

Figure 7 is a schematic diagram illustrating variable setting (or movement) of a detection area according to an embodiment of the present invention. As shown, when the object 20 is recognized in the detection area 12 by the object detection unit 133, the detection area setting unit 131 tracks the recognized object as the drone 200 descends. ), the detection area 12 can be set, that is, the position of the detection area 12 can be moved so that the recognized object is placed in the center.

As shown in (b) of FIG. 7, when the object 20 is detected at the corner of the detection area 12 and the drone 200 continues to descend while the detection area 12 is fixed, the object 20 A problem may occur where the sensor leaves the detection area 12. To solve this, as shown in (c) of FIG. 7, the accuracy of object detection can also be significantly improved by variably moving the position of the detection area 12 to track the recognized object.

In addition, although not explicitly shown in FIG. 7, when a plurality of objects spaced apart from each other are recognized at the detection start altitude (e.g., 20 m, etc.), as the drone 200 descends, the detection area setting unit 131 may be further configured to set a plurality of detection areas 12 corresponding to each of the plurality of recognized objects. Object detection for the plurality of detection areas 12 may be implemented individually or independently, and accordingly, even if a plurality of objects exist at the landing point where the drone 200 wants to land, all of the plurality of objects may be missed. By accurately detecting drones, drone landings can be implemented more safely.

As described above, according to the drone landing control device and method according to an embodiment of the present invention, a pre-designated object is detected based on the drone captured image and the drone is landed based on the result, thereby making the drone safer. can land at the landing site.

In addition, according to the drone landing control apparatus and method according to an embodiment of the present invention, the efficiency of object detection can be further improved by setting a detection area for detecting objects within the shooting area.

In addition, according to the drone landing control device and method according to an embodiment of the present invention, active response to the landing point is possible by variably setting the detection area according to the shape, location, and surrounding dangerous facilities of the landing point, Accordingly, the accuracy and efficiency of object detection can be further improved.

In addition, according to the drone landing control apparatus and method according to an embodiment of the present invention, when the detected object is a person, a predetermined evacuation broadcast is transmitted, so that the drone can land at the landing point more safely.

In addition, according to the drone landing control device and method according to an embodiment of the present invention, as the drone descends from the detection start altitude to the ground, the location, size, number, etc. of the detection area are changed based on the location and number of detected objects. By variably setting , the accuracy and efficiency of object detection can be further improved.

Meanwhile, various embodiments described in this specification may be implemented by hardware, middleware, microcode, software, and/or a combination thereof. For example, various embodiments may include one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), and field programmable gate arrays (FPGAs). ), processors, controllers, microcontrollers, microprocessors, other electronic units designed to perform the functions presented herein, or a combination thereof.

Additionally, for example, various embodiments may be encoded or embodied in a computer-readable medium containing instructions. Instructions contained or encoded in a computer-readable medium may cause a programmable processor or other processor to perform a method when the instructions are executed, for example. Computer-readable media includes computer storage media, which may be any available media that can be accessed by a computer. For example, such computer-readable media may include RAM, ROM, EEPROM, CD-ROM, or other optical disk storage media, magnetic disk storage media, or other magnetic storage devices.

Such hardware, software, firmware, etc. may be implemented within the same device or within individual devices to support the various operations and functions described herein. Additionally, components, units, modules, components, etc. described as “~” in the present invention may be implemented together or individually as separate but interoperable logic devices. The description of different features for modules, units, etc. is intended to highlight different functional embodiments and does not necessarily imply that they must be realized by individual hardware or software components. Rather, functionality associated with one or more modules or units may be performed by separate hardware or software components or may be integrated within common or separate hardware or software components.

Although operations are shown in the drawings in a particular order, it should not be understood that these operations are performed in the particular order shown, or in sequential order, or that all depicted operations need to be performed to achieve the desired results. . In some environments, multitasking and parallel processing can be advantageous. Moreover, the distinction of various components in the above-described embodiments should not be construed as requiring such a distinction in all embodiments, and the described components may generally be integrated together into a single software product or packaged into multiple software products. It must be understood that it can be done.

As described above, the optimal embodiments are disclosed in the drawings and specifications. Although specific terms are used here, they are used only for the purpose of explaining the present invention and are not used to limit the meaning or scope of the present invention described in the claims. Therefore, those skilled in the art will understand that various modifications and other equivalent embodiments are possible therefrom. Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the attached patent claims.

11: Shooting area
12: Detection area
20: object
100: Drone landing control device
110: control unit
120: Department of Communications
130: Object detection unit
131: Detection area setting unit
132: Object designation unit
133: Object detection unit
134: Object display unit
140: storage unit
200: Drone
300: communication network
1000: Drone landing control system

Claims (8)

As a drone landing control device 100,
A communication unit 120 configured to receive captured images from the drone 200;
a detection area setting unit 131 configured to set one or more detection areas in the received captured image having an area smaller than the received captured image; and an object detection unit 130 including an object detection unit 133 configured to detect a predetermined object in one or more detection areas. and
A control unit configured to activate the operation of the object detection unit 130 when the drone 200 reaches a predefined detection start altitude and control the landing of the drone 200 depending on the presence or absence of the detected object. Contains (110),
The detection area setting unit 131 sets the shape and size of the detection area based on at least one of the shape and location of the drone landing place and the surrounding dangerous facilities, and as the drone 200 descends, the detection area changes. further configured to keep the pixel size the same,
Drone landing control device (100). delete delete According to claim 1,
The object detection unit 133 is further configured to detect the pre-designated object in the set detection area based on AI learning on the floor plan image of the pre-designated object.
Drone landing control device (100). According to claim 1,
The object detection unit 130 further includes an object display unit 134 configured to display the detected object when the pre-designated object is detected.
Drone landing control device (100). delete As a drone landing control system 1000,
Drone landing control device (100) according to claim 1; and
Comprising a drone (200) communicatively coupled to the drone landing control device (100),
Drone landing control system (1000). As a drone landing control method,
Receiving captured images from a drone;
Setting one or more detection areas in the received captured image having an area smaller than the received captured image, and detecting a pre-designated object in the one or more set detection areas; and
When the drone reaches a predefined detection start altitude, activating a detection operation of the object and controlling landing of the drone depending on the presence or absence of the detected object,
The shape and size of the detection area are set based on at least one of the shape, location, and surrounding dangerous facilities of the drone landing site, but the pixel size of the detection area remains the same as the drone descends.
How to control drone landing.

Images