JP2018069961A - Device arrangement device, device arrangement method, and device arrangement program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a flight controller and a flight control method, which enable avoidance of harm to humans and animals upon descent.SOLUTION: A flight controller includes: image acquisition means for acquiring an image of a drop target zone; object recognition means for recognizing an object image included in the image; safety determination means for determining safety indicating drop compatibility, on the basis of the object recognized by the object recognition means; and airframe control means for dropping an airframe on the basis of the safety.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a device placement apparatus, a device placement method, and a device placement program.

  Conventionally, a method for arranging devices such as unmanned ground sensors using an unmanned aerial vehicle has been proposed (Patent Document 1).

JP 2008-150028 A

  However, in the above-described technology, it is necessary for a person located at a remote place to evaluate a drop point where the device is placed based on an image from an unmanned aircraft. For this reason, it may take time to evaluate the drop point, and when the image is complicated, the drop point may not be correctly evaluated.

  The present invention is an attempt to solve such a problem, and can arrange a device by arranging the device by evaluating the arrangement position of the device quickly and accurately without human intervention when arranging the device. The purpose is to provide a device.

The first invention is an autonomous movement means for autonomously moving toward a target area which is an area in which a device is arranged;
Image acquisition means for acquiring an image of the target area;
Object recognition means for recognizing an object included in the image;
Based on the object recognized by the object recognition means, aptitude degree calculating means for calculating aptitude degree indicating the aptitude as the position where the device is disposed;
Device placement means for placing the device in the target area based on the suitability level;
Is a device placement apparatus.

  According to the configuration of the first aspect of the invention, the device placement apparatus autonomously moves toward the target area, acquires an image, recognizes an object in the image, and calculates the suitability for placing the device. Place the device. For this reason, when the device is arranged, the device can be arranged by evaluating the arrangement position of the device quickly and accurately without using a human.

  According to a second aspect of the present invention, in the configuration of the first aspect of the invention, the object recognition means refers to a device placement device that refers to feature data generated by deep learning.

According to a third aspect of the present invention, in the configuration of the first or second aspect, the aptitude degree calculating unit subdivides the target area into small areas, and calculates the aptitude degree for each small area,
The device placement unit is a device placement device configured to place the device based on the suitability degree of the small area.

  According to a fourth aspect of the present invention, in the configuration according to any one of the first to third aspects, the device placement apparatus sets the adjacent small regions as small region groups, the suitability degree of each of the small regions, and the small region group. The device placement apparatus is adapted to place the device on the basis of the suitability degree.

According to a fifth invention, in the configuration according to any one of the first to fourth inventions, the device placement apparatus includes a whole image including a peripheral portion of the target area by the image acquisition means in the vicinity of each target area. Get
The device placement apparatus is configured to place the device with reference to the suitability of the peripheral portion.

A sixth invention is a device placement method implemented by a device placement device capable of autonomous movement,
An autonomous movement step for autonomously moving toward a target area, which is an area where devices are arranged,
An image acquisition step of acquiring an image of the target area;
An object recognition step for recognizing an object contained in the image;
Based on the object recognized in the object recognition step, an aptitude degree calculating step for calculating an aptitude degree indicating aptitude as a position where the device is arranged;
A device placement step of placing the device in the target area based on the suitability level;
Is a device placement method.

The seventh invention provides a computer for controlling the device placement apparatus,
An autonomous movement means for autonomously moving toward a target area, which is an area where devices are arranged,
Image acquisition means for acquiring an image of the target area;
Object recognition means for recognizing an object included in the image;
Based on the object recognized by the object recognition means, aptitude degree calculating means for calculating aptitude degree indicating aptitude as a position where the device is arranged,
Device placement means for placing the device in the target area based on the suitability level;
It is a device arrangement program for making it function as.

  As described above, according to the present invention, a device can be placed by evaluating the placement position of the device quickly and accurately without using a person when placing the device.

It is a figure which shows the outline of an effect | action of embodiment of this invention. It is a figure which shows the outline of an effect | action of embodiment of this invention. It is a conceptual diagram which shows the coordinate of the target area | region which arrange | positions a device. It is the schematic which shows the function structure of the unmanned air vehicle of embodiment of this invention. It is a conceptual diagram which shows the coordinate of the target area | region which arrange | positions a device. It is a flowchart of an embodiment of the present invention. It is a conceptual diagram which shows the coordinate of the target area | region which arrange | positions a device. It is a conceptual diagram which shows the coordinate of the target area | region which arrange | positions a device.

  Embodiments of the present invention will be described with reference to the drawings. Note that description of configurations that can be appropriately implemented by those skilled in the art will be omitted, and only the basic configuration of the present invention will be described.

<First embodiment>
The unmanned aerial vehicle 1 (hereinafter referred to as “unmanned aircraft 1”) shown in FIGS. 1 and 2 is a so-called drone or multicopter capable of autonomous flight. The drone 1 is an example of a device placement device. The drone 1 has a main body 3 and four rotating blades 5. Each rotary blade 5 is connected to a motor. The main body 3 includes a computer that controls each part of the drone 1, an autonomous flight control device, a wireless communication device, a positioning device using GPS (Global Positioning System), an inertial sensor such as an acceleration sensor and a gyro sensor, a magnetic sensor, an atmospheric pressure Sensors, batteries, etc. are arranged. A camera 7 is connected to the main body 3. The camera 7 is a digital camera that captures a digital image, a video camera, or the like. The drone 1 can communicate with the base station 50 (see FIG. 4) by a communication device.

  Each of the areas P1 to P9 in the large area 100 is an example of a target area in which devices are arranged. The drone 1 autonomously moves toward the areas P1 to P9. The drone 1 has coordinate data of the areas P1 to P9. FIG. 3 conceptually shows an example of coordinate data. The drone 1 is configured to work in sequence, for example, heading first to the region 1 and then heading to the region 2 after the work in the region 1 is completed. In the areas P1 to P9, when there is a position having the aptitude for arranging the device, the device S is arranged in each of the areas P1 to P9 as shown in FIG. The device S is, for example, a sensor or measurement device for investigating the environment, and is a vibration sensor that measures vibration, a radiation sensor that measures radiation, a wind speed sensor, a humidity sensor, a temperature sensor, or the like. For example, by arranging the device S in a specific village, the state of the ground in the area can be observed.

  A region 200 shown in the large region 100 has mountains 202, 204, and 206 as a background, and a lake 208 in the front. A road 210 is disposed between the mountain 206 and the lake 208, and an automobile 214 is passing therethrough. A village 212 made up of a plurality of houses is formed between the foot of the mountain 206 and the road 210. When the drone 1 places a plurality of devices S in the large area 100 in the region 200, the drone avoids the vicinity of the village 212 and the road 210, and depending on the type of the device S, avoids the vicinity of the lake 208. It is supposed to be. Furthermore, the drone 1 is configured to select and arrange a place where the device S can exhibit the most performance in the region 1 when the device S is arranged.

  4, the drone 1 includes a CPU (Central Processing Unit) 10, a storage unit 12, a wireless communication unit 14, a GPS (Global Positioning System) unit 16, an inertial sensor unit 18, an image processing unit 20, and drive control. Section 22, device arrangement control section 24, and power supply section 26.

  The wireless communication unit 14 enables the drone 1 to communicate with the base station 50. The base station 50 is an example of a management device that monitors and manages take-off and landing of the drone 1 and appropriately gives instructions regarding flight.

  The drone 1 can measure the position of the airframe by the GPS unit 16 and the inertial sensor unit 18. The GPS unit 16 basically receives radio waves from three or more GPS satellites and measures the position of the unmanned air vehicle 1. The inertial sensor unit 18 measures the position of the drone 1 by accumulating the movement of the drone 1 from the starting point using, for example, an acceleration sensor and a gyro sensor.

  By the image processing unit 20, the drone 1 can acquire an image by operating the camera 7.

  The drone 1 controls the rotation of each rotary blade 5 by the drive control unit 22.

  The device placement control unit 24 causes the drone 1 to place devices mounted on the machine body in the areas P1 to P9.

  The power supply unit 26 is, for example, a replaceable rechargeable battery, and supplies power to each unit of the drone 1.

  The storage unit 12 stores various data and programs necessary for unmanned flight such as data indicating a flight plan for autonomous flight from a starting point to a target position, and the following programs.

  The storage unit 12 stores an image acquisition program, an object recognition program, an aptitude degree calculation program, and a device arrangement program.

  The CPU 10 and the image acquisition program are an example of image acquisition means for operating the image processing unit 20 and acquiring an image. The drone 1 acquires images of the areas P1 to P9 by the image acquisition program. For example, when the drone 1 determines that the target region has reached the vicinity of the region P1, the drone 1 acquires an image of the entire region P1. For example, when the distance on the horizontal plane is within 100 m (meters) from the center coordinate of the region P1, the drone 1 determines that the vicinity of the region P1 has been reached and starts acquiring an image of the region P1. When the drone 1 approaches the area P1, as shown in FIG. 5, the area P1 is divided into small areas a to p, and image acquisition focusing on the small areas a to p is performed.

  The CPU 10 and the object recognition program are examples of object recognition means. The drone 1 refers to feature data generated by deep learning (deep learning) by the object recognition program, and can recognize the feature by identifying the feature of the object included in the acquired image. It has become. Note that deep learning is machine learning of a neural network having a multilayer structure, and the field of image recognition is one of the promising fields of use. The object recognition program recognizes terrain irregularities as objects.

  The drone 1 recognizes that, for example, the small areas a, b, c, d, e, f, i, j, m, n, o, and p are terrain with severe irregularities. Then, for example, it is recognized that the small areas g and h are relatively flat but have a lot of plants, and the small areas k and l are very flat and dry.

  The CPU 10 and the aptitude degree calculation program are examples of aptitude degree calculation means. The aptitude degree calculation program is a program for calculating an aptitude degree indicating aptitude as a position where the device is arranged based on the object recognized by the object recognition program. The aptitude degree calculation program has aptitude degree definition data for each type of device. The aptitude degree definition data defines aptitude degrees according to the environment in which the device functions effectively. For example, the suitability is low when it is close to a house, close to a road on which an automobile passes, or when a device cannot be stably placed on a rough terrain. A device that is not suitable for submersion is a puddle, and if the device is submerged, its suitability is low. On the other hand, when the land is flat and dry, the suitability is high. For example, the aptitude is evaluated in five stages from low to high. The lowest aptitude is aptitude 1, and aptitude 5 is the highest aptitude. The aptitude degree definition data includes, for example, the aptitude level 1 for the vicinity of a house and a puddle, the aptitude level 2 for a road area, the aptitude level 3 for rugged terrain, and the aptitude level 5 for a flat and dry place. It is data associated with aptitude. The drone 1 refers to the aptitude degree definition data of the aptitude degree calculation program to determine the aptitude degree of the position where the device is arranged.

  For example, the small areas k and l are very flat and dry and have the highest suitability and 5 suitability because the device is likely to function effectively. The small areas g and h are relatively flat but have a lot of plants, and the suitability is 3. The small areas a, b, c, d, e, f, i, j, m, n, o, and p are terrain with severe irregularities, and therefore the suitability is 1.

  The CPU 10 and the device arrangement program are examples of machine control means. The device placement program is a program for placing devices based on suitability. In the drone 1, the small areas k and l have aptitude 5, the small areas g and h have aptitude 3, and the small areas a, b, c, d, e, f, i, j, m, Based on the fact that n, o, and p have aptitude 1, the small areas k and l having high aptitude are selected, and the device is placed in either the subarea k or l.

  Hereinafter, the operation of the drone 1 will be described with reference to a flowchart. As shown in FIG. 6, the drone 1 autonomously flies from the starting point toward the target area, determines whether or not the target area, for example, the vicinity of the area P1, has been reached (step ST1), If it is determined that the vicinity of the area has been reached, the camera 7 is operated to start image acquisition (step ST2). Subsequently, the drone 1 determines whether or not it has reached the sky above the area where the device is initially placed, for example, the sky above the area P1 (step ST3). An image is acquired by focusing on each small region a and the like, and the suitability degree for each small region is calculated based on the image (step ST4). Subsequently, it is determined whether or not there is a small area having an aptitude degree equal to or higher than a reference for arranging the device, that is, whether or not there is an aptitude point (step ST5). ST6) If there is no suitable point, no device is placed (step ST7). Subsequently, when the drone 1 determines that the operations for all the regions P1 to P9 have been completed (step ST8), it stops image acquisition and returns to the base station (step ST9). On the other hand, when it is determined that the operations for all the areas have not been completed (step ST8), the drone 1 moves to the next area (step ST10).

<Second Embodiment>
In the second embodiment, the drone 1 arranges devices based on the suitability degree of each small area and the suitability degree of the small area group by setting the adjacent small areas as small area groups by the suitability degree calculation program. It is like that.

  This will be described with reference to FIG. For example, if the small regions g and k are small region groups, the average suitability is 4. Similarly, the average suitability of the small region group composed of the regions h and l is 4, and the average suitability of the small region group composed of the small regions k and o is 3, which is composed of the small regions l and p. The average suitability of the small area group is 3, and the average suitability of the small area group consisting of the small areas k and l is 5. When calculated in this way, the average aptitude of the boundary between the small areas k and l is 5 and the highest value, but the drone 1 has a small average aptitude of 3 among the boundaries of the small areas k and l. It is determined that the suitability is higher on the boundary side with the small regions g and h side having aptitude degree 4 than on the border side with the regions o and p side. That is, in addition to the suitability degree for each small area, the suitability degree of the small area group composed of the adjacent small areas can be used to determine a more appropriate position for arranging the device.

<Third embodiment>
In the third embodiment, the drone 1 is executed by an image acquisition program.
In the vicinity of each target area P1, etc., an entire image including the peripheral part of the target area P1, etc. is acquired, and the device is arranged with reference to the suitability degree of the peripheral part.

  This will be described with reference to FIG. In the stage where the drone 1 is based on the suitability of the small area and the small area group, the boundary line between the small areas k and l and the vicinity of the boundary line with the small areas g and h are most appropriate. to decide. Here, for example, it is assumed that the aptitude degree in the vicinity of the small region 1 is 4 in the peripheral region P1out. On the other hand, the suitability degree of the small region j adjacent to the small region k in the region P1 is 1. The average suitability of the small region group consisting of the small regions j and k is 3, and the average suitability of the region composed of the small region 1 and the peripheral region P1out in the vicinity of the small region 1 is 4.5. The drone 1 determines that the most appropriate device arrangement position is not the boundary between the small areas k and l but the part moved in the direction of the arrow X1. Thereby, it is possible to determine an even more appropriate position for arranging the device.

In the first to third embodiments described above, the storage unit 12 of the unmanned air vehicle 1 has been described as storing an image acquisition program, an object recognition program, an aptitude degree calculation program, and a device arrangement program. The recognition program and / or the suitability calculation program is stored in a storage unit in the server of the base station 50, and the result of processing in the base station 50 is transmitted to the drone 1, thereby controlling the drone 1. May be. In this case, the storage unit 12 of the drone 1 only needs to store a processing program for processing an instruction from the base station 50.

The device placement apparatus and the device placement method of the present invention are not limited to the above-described embodiment, and various modifications can be made without departing from the scope of the present invention.

1 Unmanned flying vehicle (unmanned aircraft)
10 CPU
14 wireless communication unit 16 GPS unit 18 inertial sensor unit 20 image processing unit 22 drive control unit 24 device arrangement control unit 26 power supply unit

Claims (7)

An autonomous movement means for autonomously moving toward a target area, which is an area where devices are arranged,
Image acquisition means for acquiring an image of the target area;
Object recognition means for recognizing an object included in the image;
Based on the object recognized by the object recognition means, aptitude degree calculating means for calculating aptitude degree indicating the aptitude as the position where the device is disposed;
Device placement means for placing the device in the target area based on the suitability level;
A device placement apparatus.   The device placement apparatus according to claim 1, wherein the object recognition unit refers to feature data generated by deep learning. The aptitude degree calculating means subdivides the target area into small areas, calculates the aptitude degree for each small area,
The device placement apparatus according to claim 1, wherein the device placement unit places the device based on the suitability degree of the small area.   The device placement apparatus arranges the devices based on the suitability degree of each small area and the suitability degree of the small area group, with the small areas adjacent to each other being a small area group. The device placement apparatus according to any one of claims 1 to 3. In the vicinity of each target area, the device arrangement apparatus acquires an entire image including a peripheral portion of the target area by the image acquisition unit,
The device placement apparatus according to any one of claims 1 to 4, wherein the device is placed with reference to the suitability degree of the peripheral portion. A device placement method implemented by a device placement apparatus capable of autonomous movement,
An autonomous movement step for autonomously moving toward a target area, which is an area where devices are arranged,
An image acquisition step of acquiring an image of the target area;
An object recognition step for recognizing an object contained in the image;
Based on the object recognized in the object recognition step, an aptitude degree calculating step for calculating an aptitude degree indicating aptitude as a position where the device is arranged;
A device placement step of placing the device in the target area based on the suitability level;
Device placement method including: A computer that controls the device placement device,
An autonomous movement means for autonomously moving toward a target area, which is an area where devices are arranged,
Image acquisition means for acquiring an image of the target area;
Object recognition means for recognizing an object included in the image;
Based on the object recognized by the object recognition means, aptitude degree calculating means for calculating aptitude degree indicating aptitude as a position where the device is arranged,
Device placement means for placing the device in the target area based on the suitability level;
Program to function as.