JP7199894B2 - Flying robots and surveillance systems - Google Patents

Description

TECHNICAL FIELD The present invention relates to a flying robot and a monitoring system for driving away a target moving object that has entered a specific area.

2. Description of the Related Art Conventionally, in specific areas such as plantations, vegetable fields, and orchards, fences for preventing entry have been installed over a certain range as a countermeasure against animal damage caused by animals such as deer and wild boars entering.

However, if the fence is destroyed by a rushing animal such as a deer or wild boar, the animal will enter a specific area within the fence from the destroyed place, and the specific area will be devastated and cause great damage.

At that time, most of the beasts that entered the specific area within the fence escaped outside the fence on their own, but there are quite a few animals that remain inside the fence because they do not know where they entered.

Therefore, in order to prevent the damage caused by the beast that entered the fence, it is conceivable to set a trap for the beast and capture it. For example, in Patent Document 1 below, as one method of capturing wild deer, the existence of wild deer is confirmed, the wild deer is confined within a wide-area defined area including the existence area, and a trap is attached. A method is disclosed for emitting sound from a drone toward a capture area.

JP 2018-099044 A

However, it was forbidden to slaughter the captured beasts carelessly, and it was necessary to quickly drive the beasts out of the specified area and let them escape outside the fence.

And the above-mentioned Patent Document 1 is an invention aimed at capturing wild deer. It is not a method to escape, and it is not possible to specify the place where wild deer entered and quickly escape from the fence.

Also, as a fundamental solution, it is conceivable to block the entrance where the beast entered and prevent it from entering the fence. I couldn't.

SUMMARY OF THE INVENTION An object of the present invention is to provide a flying robot and a monitoring system capable of driving a target moving object out of a specific area through an entrance.

To achieve the above object, a flying robot according to the present invention provides detection information for acquiring detection information for detecting a target moving object that has entered through an entrance located on the boundary between the inside and outside of a physically partitioned specific area. acquisition means;
estimating means for estimating the entrance through which the target moving object entered the specific area based on the activity status of the target moving object;
When the detection information acquiring means detects the target moving object, the target moving object is driven from the current position of the target moving object toward the entrance estimated by the estimating means, and the target moving object moves out of the specific area from the entrance . a flight control means for controlling flight to guide the
characterized by comprising

Further, the monitoring system according to the present invention includes a flying robot that flies in a range including a physically partitioned specific area,
detection means for detecting a target moving object that has entered from an entrance located on the boundary between the inside and outside of the specific area;
estimating means for estimating the entrance through which the target moving object entered the specific area based on the activity status of the target moving object;
When the detection means detects the target moving object, the target moving object is driven from the current position toward the entrance estimated by the estimation means, and the target moving object is driven out of the specific area from the entrance. a control means for controlling the flight of the flying robot so as to guide it to
characterized by comprising

Further, in the flying robot and monitoring system according to the present invention, the estimating means detects the animal trail of the target moving object from the image information of the specific area, and estimates the entrance based on the direction of the animal trail. may

Further, in the flying robot and the monitoring system according to the present invention, the estimating means may estimate a position where the target moving object is detected multiple times around the specific area from the image information of the specific area as the entrance.

According to the present invention, when a target moving object that has entered a specific area is detected, the entrance through which the target moving object has entered the specific area is estimated, and the target moving object is guided to the estimated entrance. It is possible to drive out of a specific area from.

1 is a schematic configuration diagram of a monitoring system according to the present invention; FIG. 1 is a schematic diagram of a monitored area including physically demarcated specific areas; FIG. 1 is a flow chart of a monitoring system according to the invention;

A mode for carrying out the present invention will be described in detail below with reference to FIGS. 1 to 3. FIG.

[Overview of the present invention]
The present invention relates to a flying robot and a monitoring system that guide a target moving object, such as a deer or a wild boar, that has entered a specific area to the place where it has entered and drive it out of the area.

In the present invention, when a flying robot (drone) patrolls a surveillance area including a physically partitioned specific area, when it is detected that the target moving object is entering the specific area, the target moving object enters the specific area. An entrance is estimated, the flight of the flying robot is controlled so as to guide the target moving object in the direction estimated to be the entrance, and the target moving object is driven out of the specified area from the entrance.

For example, if the target moving object is a wild animal such as a deer or wild boar, the entrance is estimated from the positional relationship between the detected animal path and the specific area on the map by detecting the animal path from image information or the like. Alternatively, it is estimated that a location outside the specific area and around the specific area where the plurality of target moving objects gather (or where they have gathered in the past) is the entrance.

[Monitoring area]
The monitoring area is one or a plurality of areas that can be monitored by the flying robot's patrol, and is an area of a predetermined range including a physically partitioned specific area. For example, in FIG. 2, the specific area S1 is a plantation surrounded by fences for preventing entry, and the monitoring area S2 is a predetermined range including the periphery of the plantation. Note that the target moving objects T are wild animals such as deer and wild boars, people, and the like, and cars and the like are excluded.

[Regarding the configuration of the monitoring system]
As shown in FIG. 1 , a monitoring system 1 is schematically configured including a detection device 2 , a flying robot 3 , a roboport 4 and a monitoring center 5 .

The detection device 2 is arranged in the monitoring area S2 or around the specific area S1, and is composed of various sensors such as a laser sensor, a microwave sensor, an ultrasonic sensor, and an image sensor. The detection device 2 detects the target moving object T that has entered the specific area S1 of the monitoring area S2, and when the target moving object T is detected, transmits a detection signal to the monitoring center 5 via a wired or wireless communication network. do. For example, when the detection device 2 is composed of a laser sensor, a detection signal is transmitted to the monitoring center 5 when the target moving object T is detected in the scanning range by scanning the specific area S1 with a laser beam at a predetermined cycle.

As shown in FIG. 1, the flying robot 3 includes detection information acquisition means 3a for acquiring detection information of a target moving object T that has entered a specific area S1, and radio waves from satellites or quasi-zenith satellites. Position information acquisition means 3b for acquiring GPS (Global Positioning System) position information, and flight control means 3c for performing flight control based on the position information and 3D map information (three-dimensional geographical information). It is a configuration including a function of calculating the self-location (latitude, longitude, altitude) using altitude information from a sensor, and a photographing function using an imaging device such as a visible light camera or an infrared camera.

The flying robot 3 is on standby at the robot port 4 in a normal state, and when receiving a flight instruction from the monitoring center 5, it autonomously flies while avoiding obstacles and patrols the monitoring area S2 according to the flight route of the patrol schedule. Then, the surrounding area is photographed for each monitoring area S2, the image is transmitted to the monitoring center 5, and when the patrol of all the monitoring areas S2 is completed, the robot returns to the ROBOPORT 4. - 特許庁
Alternatively, when the detection device 2 detects the target moving object T, the monitoring center 5 creates a flight route corresponding to the detection position, and based on the flight instructions from the monitoring center 5, the flying robot 3 follows the flight route to the target. It flies toward the detection position of the moving object T.

The patrol schedule indicates the schedule of the flight route of the monitoring area S2 that the flying robot 3 patrols. Further, the patrol schedule can be set by prioritizing the monitoring area S2 including the place where the target moving object T is likely to enter based on past image information, for example.

The flying robot 3 stores 3D map information in advance in a storage means (not shown), and when it comes time to start the patrol schedule, the flying robot 3 acquires its own position (latitude, longitude , altitude), and fly autonomously based on the 3D map information to patrol the monitoring area S2 according to the flight route.

While the flying robot 3 is waiting at the robot port or patrolling the monitoring area S2 according to the flight route, the detection device 2 detects the target moving object T that has entered the specific area S1, and a flight instruction is issued from the monitoring center 5. When there is, it autonomously or quasi-autonomously flies toward the location where the target moving object T is detected while avoiding obstacles.

The roboport 4 is a waiting place for the flying robot 3, and is equipped with equipment for receiving instructions from the monitoring center 5 and for taking off and landing of the flying robot 3. The Roboport 4 has a mechanism for accommodating the flying robot 3 to land in the port, and has a function of supplying power to the flying robot 3 in a contact or non-contact manner when the flying robot 3 is accommodated in the port. have.

The monitoring center 5 is provided, for example, in a facility operated by a security company or the like, and includes a monitoring console 11 comprising one or a plurality of computers for receiving images captured by the flying robots 3 and displaying the received images.

The monitoring console 11 is equipped with communication means 11a, storage means 11b, estimation means 11c, and flight control means 11d, as shown in FIG. It records the information received from the detection device 2 and the flying robot 3 via 11a, and displays the information on the abnormality on the display.

The storage means 11b stores 3D map information (three-dimensional geographical information) required for the flying robot 3 to patrol the surveillance area S2.

The estimating means 11c estimates an entrance P through which the target moving object T has entered the specific area S1 based on image information of the specific area S1 (including its periphery). For example, as shown in FIG. 2, if the target moving object T is a deer, trace information of the movement of the deer T (so-called animal trail) is detected from the visible light image or the infrared image of the specific area S1, and the detected trace is detected. The entrance P through which the deer T has entered the specific region S1 is estimated from the extension of the inclination on the information image. This is because deer, etc., which have entered once, tend to repeatedly enter and exit from the entrance, and as a result, traces of movement often remain. Known techniques can be used to find traces of movement of deer or the like on the image. For example, an image taken during a patrol flight can be compared with the same past image, and a place where a linear component of a predetermined length or more is obtained can be determined as an animal trail. Alternatively, an infrared image may be used to determine that a place having a different temperature from the surroundings and being on a straight line is an animal trail.

Alternatively, from the image information of the specific area S1, the entrance P is estimated as the position where the deer T is detected multiple times outside the specific area S1 and around the specific area. This is because when the entrance to the specific area S1 is small, or when multiple deer try to enter at the same time, one or more deer stay around the entrance outside the specific area. It is based on the knowledge that there is a tendency, and it is possible to estimate a place where multiple deer are staying around the entrance outside a specific area or where deer have been detected multiple times in the past as an entrance.

The flight control means 11d issues flight instructions (flight route instructions, target position and speed instructions, takeoff instructions, return instructions, climb instructions, descent instructions, etc.) to direct the flying robot 3 to an arbitrary location at the discretion of the observer. It controls the flight of the flying robot 3.

The flight control means 11d confirms an abnormality in the specific area S1 of the monitoring area S2 based on the detection signal of the detection device 2 in a state where monitoring of the monitoring area S2 is started, and outputs a signal giving a flight instruction, target moving object information, obstacles, and so on. Various information such as object information is transmitted to the flying robot 3 .

Although FIG. 1 illustrates that the monitoring console 11 of the monitoring center 5 includes communication means 11a, storage means 11b, estimation means 11c, and flight control means 11d, the configuration is not limited to this. For example, a flight control device functioning as communication means 11a, storage means 11b, estimation means 11c, and flight control means 11d may be provided separately from the monitoring console 11. FIG. In this case, the flight control device is installed, for example, at a predetermined location or in the vicinity of the monitoring area S2 or the specific area S1, and communicates with the detection device 2, the flying robot 3, and the monitoring center 5 via a wired or wireless communication network. Send and receive various information.

Further, the flying robot 3 may be configured to have the function of the detection device 2 for detecting the target moving object T that has entered the specific area S1. In this case, the flying robot 3 stores the pattern image of the target moving object T in advance in the storage means, and the flying robot 3 captures an image of the monitoring area S2 including the specific area S1 from the sky with an imaging device. Comparison between the information and a reference image (an image in which the target moving object T does not exist in the specific region S1), or comparison of image information before and after the flying robot 3 captures the monitoring region S2 including the specific region S1 from the sky, that is, the pattern A target moving object T that has entered the specific area S1 is detected by matching.

[Operation of the monitoring system]
Next, as an operation of the monitoring system 1 configured as described above, an entry handling process for driving away the target moving object T that has entered the specific area S1 from the entrance P to the outside of the specific area S1 will be described with reference to the flowchart of FIG. do.

At the start time of the patrol schedule, the flying robot 3 starts patrolling the monitoring area S2 according to the flight route of the patrol schedule while avoiding obstacles (ST1).

When the flying robot 3 starts patrolling the monitoring area S2, it determines whether or not the target moving object T has entered the specific area S1 of the monitoring area S2 (ST2). This determination is made based on whether or not the detection device 2 has detected the target moving object T that has entered the specific area S1. Alternatively, when the flying robot 3 arrives at the monitoring area S2 of the patrol destination and takes an image, the flying robot 3 detects the target moving object T that has entered the specific area S1 within the monitoring area S2 by pattern matching of the image. It depends on whether or not

When it is determined that the target moving object T has entered the specific region S1 of the monitoring region S2 (ST2-Yes), the estimating means 11c estimates the entrance P at which the target moving object T has entered the specific region S1 (ST3 ). For example, when the specific area S1 is an area surrounded by fences and the target moving object T is a wild animal such as a deer or a wild boar, the estimating means 11c uses the following methods (1) and (2) to determine the activity status of the wild animal. Estimate entrance P.

(1) Detect a narrow path that is naturally created by the traffic of wild animals, a so-called "animal trail", from image information obtained by imaging the specific area S1, and estimate the entrance P from the positional relationship between the detected animal trail and the fence. .

(2) Based on image information obtained by imaging the specific area S1, the entrance P is estimated to be near a place where a plurality of wild animals gather (or have gathered in the past) outside the fence of the specific area S1.

When the entrance P is estimated by the estimation means 11c (ST3), the flying robot 3 drives the target moving object T to the estimated entrance P and drives it out of the specific area S1 from the entrance P. Guided flight (ST4). For example, if the target moving object T is a deer and the specified area S1 is a plantation surrounded by fences for preventing entry as shown in FIG. Or guide the deer T to go to the entrance P by emitting a guidance sound (sound of a deer whistle) from outside the entrance prevention fence to guide the deer T to go to the entrance P to prevent entry from the entrance P drive out of the fence. When the target moving object T is detected, the control is switched to manual operation from the monitoring center 5, and the position of the entrance P estimated to be the target moving object T is displayed on the monitor of the monitoring center 5 as a symbol. The controller may drive the target moving object T from the entrance P by manual control while referring to the symbol.

Then, it is determined whether or not the patrol of the surveillance area S2 by the flying robot 3 has been completed (ST5). to end the process.

On the other hand, if it is determined that the patrol of the monitoring area S2 by the flying robot 3 has not been completed (ST5-No), the flying robot 3 flies to the next monitoring area S2 according to the flight route of the patrol schedule, and moves to the next monitoring area S2. The patrol is continued, the process returns to ST2, and the determination process described above is repeated until the flying robot 3 completes the patrol of the monitoring area S2.

By the way, the above-described operation of FIG. 3 is the processing for dealing with the entry of the target moving object T during patrol of the monitoring area S2. When the target moving object T is detected, the flying robot 3 flies to the place where the target moving object T is detected according to the flight instruction from the monitoring center 5 and performs the entry coping process.

As described above, according to the present embodiment, when the flying robot 3 patrolls the monitoring area S2, which includes the specific area S1 physically partitioned by a fence or the like, for example, when an object such as a deer or a wild boar moves. When it is detected that the object T is entering the specific area S1, the flight control is performed so as to guide the target moving object T in the direction assumed to be the entrance P. As a result, it is possible to guide the target moving object T that has entered the specific area S1 to the entrance P and drive it out of the specific area S1 from the entrance P. FIG.

Although the best mode of the flying robot and monitoring system according to the present invention has been described above, the present invention is not limited by the description and drawings according to this mode. In other words, it goes without saying that other forms, embodiments, operation techniques, etc. made by persons skilled in the art based on this form are all included in the scope of the present invention.

1 monitoring system 2 detection device 3 flying robot 3a detection information acquisition means 3b position information acquisition means 3c flight control means 4 roboport 5 surveillance center 11 monitoring console 11a communication means 11b storage means 11c estimation means 11d flight control means S1 specific area S2 monitoring area T Target moving object P Entrance

Claims (6)

detection information acquisition means for acquiring detection information of a target moving object that has entered through an entrance located on the boundary between the inside and outside of a physically partitioned specific area;
estimating means for estimating the entrance through which the target moving object entered the specific area based on the activity status of the target moving object;
When the detection information acquiring means detects the target moving object, the target moving object is driven from the current position of the target moving object toward the entrance estimated by the estimating means, and the target moving object moves out of the specific area from the entrance . a flight control means for controlling flight to guide the
A flying robot comprising: A flying robot that flies in a range that includes a physically partitioned specific area;
detection means for detecting a target moving object that has entered from an entrance located on the boundary between the inside and outside of the specific area;
estimating means for estimating the entrance through which the target moving object entered the specific area based on the activity status of the target moving object;
When the detection means detects the target moving object, the target moving object is driven from the current position toward the entrance estimated by the estimation means, and the target moving object is driven out of the specific area from the entrance. a control means for controlling the flight of the flying robot so as to guide it to
A surveillance system comprising: 2. The estimating means according to claim 1, wherein said estimating means detects trace information of movement of said target moving object from image information of said specific area, and estimates said entrance based on the direction of said trace information. flying robot. 2. The flying robot according to claim 1, wherein the estimating means estimates, from the image information of the specific area, a position where the target moving object is detected multiple times outside and around the specific area as the entrance. 3. The estimating means according to claim 2, wherein said estimating means detects trace information of movement of said target moving object from image information of said specific area, and estimates said entrance based on the direction of said trace information. Monitoring system. 3. The monitoring system according to claim 2, wherein the estimating means estimates a position where the target moving object is detected multiple times outside and around the specific area from the image information of the specific area as the entrance.

Images