WO2021177192A1 - Path generation method, path generation device, and program - Google Patents

Abstract

Provided is a path generation method for generating a path for sensing a detection target area in which rows of a sequence of detection targets are aligned in parallel, wherein a path generation process is performed to generate a full path such that a partial path passing through the detection target area crosses at least two rows.

Description

Path generation method, path generator, program

This technology relates to a path generation method, a path generation device, and a program, for example, a technique for generating a path for sensing in a field.

In recent years, due to the increase in scale of farms, automatic crop planting has come to be carried out. In automatic planting, it is very important to make sure that the crop is planted correctly or that the crop is growing as expected. If some crops are not planted and grown as expected, farm staff must also decide whether to replant. Replanting (replanting) means replanting (re-sowing, etc.) when the plant has already been planted.

On the other hand, with the development of aerial imaging technology using cameras mounted on flying objects such as drones, attempts are being made to simplify and automate the work of large-scale farms by omitting the work of manually looking around the farmland. The detection of the above-mentioned planting defects is also being automated by such aerial photography technology.
Patent Document 1 discloses a technique for imaging a field and performing remote sensing of a vegetation state.

Japanese Patent No. 5162890

When imaging for the purpose of acquiring sensing data as a sample, it is necessary to perform appropriate sampling that can accurately detect, for example, planting defects in the field.

Therefore, this disclosure proposes a technique for performing appropriate sampling.

In the path generation method according to the present technology, as a path generation method for generating a path for sensing for a detection target area in which rows, which are an array of detection target objects, are parallel, at least two partial paths passing through the detection target area are used. A path generation process is performed to generate an entire path so as to cross the row.
In the sensing for the detection target area, various data sensed for the detection target area are acquired as sensing data. For example, sensing data in a discrete area of the detection target area is acquired as a sample.
The partial path passing through the detection target area is a line segment whose at least a part is located in the detection target area. For example, a line segment from the start point to the end point on the outer edge of the detection target area and crossing at least a part of the detection target area can be considered. Further, it may be a line segment from one point on the outer edge of the detection target area to one point in the detection target area, or a line segment as a whole located in the detection target area.

In the path generation method according to the present technology described above, it is conceivable that at least a part of the partial paths in the entire path are paths that cross all the rows in the detection target area.
As a partial path that crosses all the rows in the detection target area, for example, a line segment from a start point to an end point on the outer edge of the detection target area and that crosses the detection target area can be considered.

In the path generation method according to the present technology described above, the path generation process includes a first step of setting a boundary shape surrounding the detection target area and a second step of generating the partial path within the boundary shape. A third step of extracting the intersection of the outer edge of the detection target area and the partial path generated in the second step, and a fourth step of generating the entire path by a process of connecting the intersections. It is possible to have.
The boundary shape is a frame-like shape that surrounds the detection target area. The boundary shape is used as an auxiliary line for generating a partial path in the path generation process.

In the path generation method according to the present technology described above, it is conceivable that the boundary shape is rectangular.
That is, a partial path is generated within the boundary shape having a rectangular shape.

In the path generation method according to the present technology described above, in the second step, it is conceivable to generate the partial path so that the adjacent partial paths form a predetermined path angle.
A partial path is generated based on a predetermined path angle within the boundary shape.

In the path generation method according to the present technology described above, it is conceivable to set the predetermined path angle so as to satisfy the predetermined minimum coverage rate.
The coverage rate is, for example, a ratio indicating how much area was acquired as a sample with respect to the entire area of the detection target area. The minimum coverage rate is a value that serves as a guideline for the coverage rate in the detection target area.

In the path generation method according to the present technology described above, in the fourth step, it is conceivable to connect the intersections in the adjacent partial paths to generate a connection path.
Such a connection path is a line segment connecting adjacent partial paths.

In the path generation method according to the present technology described above, it is conceivable that the partial path is formed by a line segment which is a straight line or a curved line.
That is, the partial path may be formed in a straight line or a curved line.

In the path generation method according to the present technology described above, it is conceivable that a connection path along the outer edge of the detection target area is included as one of the paths included in the entire path.
The connection path along the outer edge of the detection target area is, for example, a line segment connecting adjacent partial paths.

In the path generation method according to the present technology described above, it is conceivable that the connection path along the outer edge of the detection target area is not included as one of the paths included in the overall path.
The connection path along the outer edge of the detection target area is, for example, a line segment connecting adjacent partial paths.

In the path generation method according to the present technology described above, it is conceivable to generate the entire path so that the adjacent partial paths are connected to each other at the end.
For example, when all the adjacent partial paths in the whole path are connected at the end, the whole path does not include the connection path connecting the adjacent partial paths. In addition, only some of the adjacent partial paths in the entire path may be connected at the end.

In the path generation method according to the present technology described above, it is conceivable to generate the entire path such that the adjacent partial paths are non-parallel.
For example, an entire path in which adjacent partial paths extend in alternating directions or an overall path having a zigzag shape as a whole is generated.

In the path generation method according to the present technology described above, it is conceivable that the detection target area is a field and the wax is a line where crops are planted in the field.
The field broadly includes farmland where crops are cultivated, such as crop cultivated land, cultivated land, hydroponic cultivated land, and house cultivated land. Crop refers to, for example, a crop grown in a field for the purpose of harvesting.

In the path generation method according to the present technology described above, the entire path is the flight path of the flying object, and it is conceivable that the sensing is performed by the flying object.
The aircraft moves over the detection target area along the entire path and senses the detection target area.

The path generator according to the present technology is a path generator that generates a path for sensing for a detection target area in which rows, which are an array of detection target objects, are parallel to each other, and has at least two partial paths passing through the detection target area. A path generation process is performed to generate an entire path so as to cross the two rows.
As a result, the path generation device executes the path generation process.
The program according to the present technology is a program that causes the path generator to execute the process of the path generation method. This facilitates the realization of a path generator that executes the path generation process.

It is explanatory drawing of the sensing system of embodiment of this technique. It is explanatory drawing of the row and the imaging range in a field. It is a block diagram of the hardware composition of the information processing apparatus of embodiment. It is a flowchart which shows the work procedure for the field management of embodiment. It is explanatory drawing which shows an example of sampling in a field. It is explanatory drawing which shows an example of sampling of embodiment. It is explanatory drawing of the coverage rate in sampling. It is 1st explanatory diagram of the path generation of embodiment. It is a 2nd explanatory diagram of the path generation of an embodiment. It is a 3rd explanatory diagram of the path generation of an embodiment. It is a 4th explanatory diagram of the path generation of an embodiment. FIG. 5 is a fifth explanatory diagram of path generation according to the embodiment. FIG. 6 is a sixth explanatory diagram of path generation according to the embodiment. It is a flowchart of the path generation process of embodiment. It is explanatory drawing which shows the modification of the sampling of embodiment.

Hereinafter, embodiments will be described in the following order.
<1. Sensing system configuration>
<2. Information processing device configuration>
<3. Work procedure>
<4. Sampling>
<4-1. Path>
<4-2. Coverage rate>
<5. Path generation flow>
<6. Path generation process of the embodiment>
<7. Summary and transformation examples>

<1. Sensing system configuration>
First, the sensing system of the embodiment will be described.
FIG. 1 shows an information processing device 1 as a path generating device constituting a sensing system, and an imaging device 220 mounted on a small flying object 200 such as a drone. Also shown is a tractor 270 for planting.

The aircraft body 200 can move over the field 300 by, for example, radio control by the operator, autopilot, or the like.
An imaging device 220 is set in the flying object 200 so as to image, for example, below. When the flying object 200 moves over the field 300 by a predetermined route, the image pickup device 220 periodically performs, for example, still image imaging.

When the flying object 200 flies at a relatively low altitude (for example, at an altitude of about 10 m to 20 m), one captured image captures a part of the field 300.
By taking still images at short intervals, it is possible to perform stitch processing on each of the captured images to obtain a composite image showing the entire field. However, in the case of the sensing of the present embodiment, such a thing is not always necessary, and for example, a discrete region in the field 300 may be imaged as a sensing sample to obtain a plurality of image data.
Hereinafter, the image data as a still image for each captured image is also referred to as a “sample”.

The image pickup device 220 mounted on the flying object 200 includes a visible light image sensor (an image sensor that captures visible light of R (red), G (green), and B (blue)), NIR (Near Infra Red: near infrared region). ) Cameras for image imaging, Multi Spectrum Cameras that capture images in multiple wavelength bands, Hyper Spectrum cameras, Fourier Transform Infrared Spectroscopy (FTIR), infrared sensors, etc. are assumed. Will be done. Of course, a plurality of types of cameras (sensors) may be mounted on the flying object 200.
As the multi-spectrum camera, for example, one that captures an NIR image and an R (red) image, and one that can calculate an NDVI (Normalized Difference Vegetation Index) from the obtained image is also assumed to be used. NDVI is a vegetation index that indicates plant-likeness, and can be used as an index that indicates the distribution status and activity of vegetation.
NDVI can be obtained from the R image and the NIR image. That is, the value of NDVI is
NDVI = (NIR-R) / (NIR + R)
Is required as.

Tag information is added to the image obtained by being captured by the image pickup device 220. The tag information includes imaging date and time information, position information (latitude / longitude information) as GPS (Global Positioning System) data, flight altitude information of the flying object 200 at the time of imaging, and imaging device information (individual identification information and model of the camera). Information, etc.), information on each image data (information on image size, wavelength, imaging parameters, etc.), etc. are included.

The information processing device 1 of the present embodiment can perform a process of setting a flight path in the field 300 of the flying object 200.
Specifically, the information processing device 1 generates a path in the field 300 based on the shape of the field 300, information input by the user, and the like, and sets it as a flight path of the flying object 200.
The information processing device 1 remotely controls the flight of the flight body 200 by transmitting a flight plan including the set flight path to the flight body 200. The aircraft 200 moves over the field 300 based on the flight path included in the received flight plan.

The information processing device 1 may be capable of acquiring image data and tag information captured by the image pickup device 220 mounted on the flying object 200.
For example, image data and tag information are transferred by wireless communication or network communication between the image pickup device 220 and the information processing device 1. As the network, for example, the Internet, a home network, a LAN (Local Area Network), a satellite communication network, and various other networks are assumed.
Alternatively, the image data or tag information may be passed to the information processing device 1 in such a manner that the recording medium (for example, a memory card or the like) mounted on the image pickup device 220 is read by the information processing device 1.

Further, the information processing device 1 can generate analysis information for the field 300 as a measurement target by using the acquired image data and tag information, and can perform a process of presenting the analysis result as an image to the user. It may be set to.
Specifically, the crops copied to the image data are counted, and based on the count, for example, the number of crops, the number of crops per unit area, the germination rate, the predicted yield, the crop ratio, etc. are used for the management of the field 300. It is conceivable to generate information that can be generated and present it to the user.
Furthermore, the information processing device 1 may also perform execution control of displaying the position of each sample (image data) on the map image including the field 300 and displaying information on the replant.

The information processing device 1 is realized as, for example, a PC (personal computer), an FPGA (field-programmable gate array), or a terminal device such as a smartphone or tablet.
Although the information processing device 1 is separate from the image pickup device 220 in FIG. 1, for example, an arithmetic unit (microcomputer or the like) serving as the information processing device 1 may be provided in the unit including the image pickup device 220. ..
Further, as the information processing device 1, an information processing device that sets the flight path of the flying object 200 and an information processing device that generates analysis information for the field 300 as a measurement target may be provided separately. ..

In the case of the sensing system of the present embodiment, for example, a portion to be "row" in the field 300 is automatically determined on the image data, and a path for sensing with respect to the field 300 is generated based on the determination.
The wax is a line on which a crop is planted, and for example, a ridge formed for seed planting in a field 300 is also a kind of wax. In addition, the line formed when seeds are sown on flat ground is also low, not limited to the one in which the soil is raised like a ridge. For example, the planting line formed when sowing seeds with a tractor (sowing machine) is called a row.

FIG. 2 schematically shows the wax 301 formed in the field 300. For example, as the days rise from sowing, the seeds germinate, and the line where the leaves of the crop are lined up becomes Row 301 as shown in the figure.
In the case of the present embodiment, the field 300 is imaged by the imaging device 220 while the flying object 200 is moving, but each part is imaged at an appropriate timing, for example, as in the imaging ranges SP1 and SP2 in the figure. Become. For example, the image data of the imaging range SP1 and the image data of the imaging range SP2 are obtained by the imaging device 220 as a one-frame still image captured image, and these are taken into, for example, the information processing device 1 as a sample.
In this figure, an example is shown in which the flying object 200 moves along the flight path F inclined with respect to the row. Generally, the flying object 200 may fly in parallel with the row, but as described later, in the present embodiment, the flying object 200 flies in a flight path including a path inclined with respect to the row.

Although FIG. 2 shows a linear row 301, the row 301 is not always a straight line. The linear wax 301 may bend at the edge of the field 300 or the like. Further, depending on the shape of the field 300, the path at the time of sowing, obstacles, etc., the row 301 may be partially curved, or the row 301 may be formed in a spiral shape or a concentric circle shape, for example.

The tractor 270 plants seeds and the like (seed, seedlings, strains, etc.) in the field 300. The tractor 270 automatically travels and plants based on, for example, planting instruction data provided by the information processing device 1. As a result, the above-mentioned wax 301 is formed in the field 300.
A computer device is mounted on the tractor 270, and the computer device controls traveling and planting operations based on instruction data. For example, a computer device imports a data file containing planting instruction data (rate map data including the number of planting instructions per unit area for each location), and the tractor 270 travels and plants based on the instruction data. To control.
Further, the actual results of the planting work by the tractor 270 (the actual number of plantings per unit area for each location) may be provided to the information processing apparatus 1 as actual data.

<2. Information processing device configuration>
In the above sensing system, an information processing device 1 that performs a process of generating a path in a field 300 will be described.

FIG. 3 shows the hardware configuration of the information processing device 1. The information processing device 1 includes a CPU (Central Processing Unit) 51, a ROM (Read Only Memory) 52, and a RAM (Random Access Memory) 53.
The CPU 51 executes various processes according to the program stored in the ROM 52 or the program loaded from the storage unit 59 into the RAM 53. The RAM 53 also appropriately stores data and the like necessary for the CPU 51 to execute various processes.
The CPU 51, ROM 52, and RAM 53 are connected to each other via the bus 54. An input / output interface 55 is also connected to the bus 54.

The input / output interface 55 can be connected to a display unit 56 composed of a liquid crystal panel or an organic EL (Electroluminescence) panel, an input unit 57 composed of a keyboard, a mouse, etc., a speaker 58, a storage unit 59, a communication unit 60, and the like.

The display unit 56 may be integrated with the information processing device 1 or may be a separate device.
The display unit 56 displays captured images, various calculation results, and the like on the display screen based on the instructions of the CPU 51. Further, the display unit 56 displays various operation menus, icons, messages, etc., that is, as a GUI (Graphical User Interface) based on the instruction of the CPU 51.

The input unit 57 means an input device used by a user who uses the information processing device 1.
For example, as the input unit 57, various controls and operation devices such as a keyboard, mouse, keys, dial, touch panel, touch pad, and remote controller are assumed.
The user's operation is detected by the input unit 57, and the signal corresponding to the input operation is interpreted by the CPU 51.

The storage unit 59 is composed of a storage medium such as an HDD (Hard Disk Drive) or a solid-state memory. The storage unit 59 stores, for example, detection data received from the macro measurement unit or the micro measurement unit, analysis results, and various other information. The storage unit 59 is also used for storing program data for analysis processing and the like.

The communication unit 60 performs communication processing via a network including the Internet and communication with devices in various peripheral parts.
The communication unit 60 may be, for example, a communication device that communicates with the flying object 200, the image pickup device 220, or the tractor 270.

A drive 61 is also connected to the input / output interface 55 as needed, and a storage device 62 such as a memory card is attached to the input / output interface 55 to write or read data.
For example, the computer program read from the storage device 62 is installed in the storage unit 59 as needed, and the data processed by the CPU 51 is stored. Of course, the drive 61 may be a recording / playback drive for a removable storage medium such as a magnetic disk, an optical disk, or a magneto-optical disk. These magnetic disks, optical disks, magneto-optical disks, and the like are also aspects of the storage device 62.

The information processing device 1 of the embodiment is not limited to the information processing device (computer device) 1 having a hardware configuration as shown in FIG. 3 being configured as a single unit, and a plurality of computer devices are systematized. It may be configured. The plurality of computer devices may be systematized by a LAN or the like, or may be arranged in a remote place by a VPN (Virtual Private Network) or the like using the Internet or the like. The plurality of computer devices may include computer devices available by cloud computing services.
Further, the information processing device 1 of FIG. 3 can be realized as a stationary type, a notebook type or the like personal computer, or a mobile terminal such as a tablet terminal or a smartphone. Further, electronic devices such as a measuring device, a television device, a monitoring device, an imaging device, and an equipment management device having a function as the information processing device 1 can also be equipped with the information processing device 1 of the present embodiment.

For example, the information processing device 1 having such a hardware configuration has a calculation function by the CPU 51, a storage function by the ROM 52, the RAM 53, and the storage unit 59, a data acquisition function by the communication unit 60 and the drive 61, and an output function by the display unit 56 and the like. By having and installed software functioning, various functional configurations are provided.

The information processing device 1 is provided with the path generation unit 2 shown in FIG. 3 as a function.
The processing function of the path generation unit 2 is realized by software started by the CPU 51.
The program constituting the software is downloaded from the network or read from the storage device 62 (for example, a removable storage medium) and installed in the information processing device 1 of FIG. Alternatively, the program may be stored in advance in the storage unit 59 or the like. Then, when the program is started in the CPU 51, the function of the path generation unit 2 is exhibited.
Further, the calculation progress and the storage of the result of the function of the path generation unit 2 are realized by using, for example, the storage area of the RAM 53 or the storage area of the storage unit 59.

The path generation unit 2 generates a path for sensing in the field 300. For example, the path generation unit 2 acquires information necessary for generating a path, and performs various calculations based on the acquired information.
Further, the path generation unit 2 sets the generated path as the flight path of the flying object 200 in the field 300, and outputs a flight plan including the flight path.

<3. Work procedure>
FIG. 4 shows an example of a work procedure for farm management.
As step S1, the flight of the flying object 200 on the field 300 is performed, and images are taken by the image pickup device 220 during the flight, for example, at regular time intervals. As a result, a large number of image data as samples are obtained.

As step S2, the information processing device 1 acquires image data. For example, the information processing device 1 captures image data captured during flight by wired or wireless communication from the image pickup device 220 or delivery of a storage device 62 such as a memory card. The information processing device 1 stores image data in, for example, a storage unit 59.

As step S3, the information processing device 1 performs an analysis process based on the image data. For example, stand counting using each image data is performed. A specific processing example will be described later as a preparatory processing.

As step S4, the information processing device 1 presents the status of the field 300 and various information to the user (for example, the staff of the farm) on the UI screen in order to determine the replant, and responds to the user's operation. Specifically, it performs a process of presenting information for considering whether or not the user replants, which area to replant, and accepting the designation of the area where the replant is assumed.

When a replant is performed as a result of input / output on the UI screen, the process proceeds from step S5 to step S6, and the information processing apparatus 1 generates an instruction file including a ROI (Region of Interest) for the replant. I do. This instruction file is provided to the tractor 270, and the ROI is information indicating the area to be replanted.
Then, in step S7, the information processing device 1 exports the instruction file.
Upon receipt of the instruction file by the tractor 270, the actual replanting is then performed.

In the above procedure, the information processing device 1 of the present embodiment generates a path of the flight path of the flying object 200, particularly at the time of sample acquisition in step S1.

<4. Sampling>
Hereinafter, sampling in the field 300 will be described with reference to FIGS. 5 to 7. In the present embodiment, acquisition of image data captured in a plurality of discrete regions in the field 300 as a sample is referred to as sampling.

<4-1. Path>
FIG. 5 is an explanatory diagram for explaining an example of sampling for the field 300, and shows the field 300 as the detection target area 310 and the outer peripheral area 320 surrounding the outer periphery of the detection target area 310.
In the detection target area 310, a plurality of rows, which are an array of detection target objects, are arranged in parallel. Although the wax is not shown in FIG. 5, it is assumed that the plurality of waxes extend in the row direction Dr indicated by the double-headed arrows. In the following, the direction in which the wax extends is referred to as the row direction Dr, and the direction in which the waxes are parallel is referred to as the row arrangement direction Dj.
The outer peripheral area 320 is a buffer area that separates the detection target area 310 from the outside, and is not provided with a detection target or a wax.

The entire path P1 is set in the detection target area 310 shown in FIG. The entire path P1 is a path for sampling, and a sample is acquired while the aircraft body 200 travels in the flight path of the entire path P1.
The flying object 200 starts flying from a point located outside the detection target area 310, flies over the detection target area 310, and then ends the flight at a point located outside the detection target area 310. The acquisition of the sample may be performed not only during the flight of the aircraft body 200 through the entire path P1, but also at regular intervals from the start to the end of the flight.

In the detection target area 310, the flying object 200 has one end of the entire path P1 as the start point P1s and the other end of the entire path P1 as the end point P1e, and is along a path such as a partial path included in the entire path P1 from the start point P1s to the end point P1e. To fly.

The partial path is a line segment that passes through the detection target area 310.
That is, the partial path is a line segment whose at least a part is located in the detection target area 310, for example, a line segment from the start point to the end point on the outer edge of the detection target area 310, and at least a part of the detection target area 310. It is a path that crosses a part. The partial path from the start point on the outer edge of the detection target area 310 to the end point and crossing at least a part of the detection target area 310 is, for example, on one side forming the square, assuming a rectangular detection target area. It is a line segment from the starting point to the ending point on the other side.
As a partial path from the start point to the end point on the outer edge of the detection target area 310, for example, assuming a rectangular detection target area, the same one side is on the same side from the start point on one side via the inside of the detection target area 310. A line segment reaching the end point of the above, and a line segment from the starting point on one side to the end point equal to the starting point via the detection target area 310 are also included.
Further, if at least a part of the partial path is a line segment located in the detection target area 310, the line segment from one point on the outer edge of the detection target area 310 to one point in the detection target area 310 or the whole is detected. It may be a line segment located in the target area 310.

The entire path P1 shown in FIG. 5 includes a plurality of parallel paths Pp (Pp1 to Pp10) and a plurality of connection paths Pc.
The parallel path Pp is a line segment from the start point to the end point on the outer edge of the detection target area 310, and is a partial path parallel to the row direction Dr.
The connection path Pc is a line segment along the outer edge of the detection target area 310, and is a path that connects the ends of adjacent partial paths.

The entire path P1 as a whole has a grid-like shape that divides the detection target area 310 into strip-shaped sections surrounded by adjacent parallel paths Pp and Pp and connection paths Pc located between them. ..
The aircraft body 200 flies in the flight path of the entire path P1 while sequentially following consecutive partial paths such as a parallel path Pp1, a connection path Pc continuous with the parallel path Pp1, and a parallel path Pp2 continuous with the connection path Pc. ..

In FIG. 5, each region of the detection target area 310 imaged while the flying object 200 flies through the entire path P1 is represented as an image plane IF. The image plane IF corresponds to the imaging range SP described with reference to FIG.
The image plane IF has a rectangular shape and has a vertical width lv extending in the extending direction of the path and a horizontal width hl extending in the direction orthogonal to the extending direction of the path (see FIG. 7). The image plane IF is positioned so that the path passes through the center of the width of the image plane IF.

In the sampling example of FIG. 5, while the flying object 200 traverses the detection target area 310 along each parallel path Pp of the entire path P1, the imaging data of the image plane IF is acquired as a sample. The plurality of sampled image plane IFs are located on each parallel path Pp.

Sampling in the field 300 is carried out, for example, in order to detect a defect in the planting of the crop planted in the field 300. Planting defects include, for example, defects that occur at random locations in the field 300 due to meteorological phenomena, and local defects that occur in specific waxes, for example. Local defects include defective agricultural equipment, such as defective planting of specific waxes due to clogging of nozzles in tractors used for planting crops.
In the sampling example of FIG. 5, since the sample is acquired along the parallel path Pp parallel to the row, the imaging data of the row located in the vicinity of the parallel path Pp is acquired as a sample. On the other hand, the imaging data of the row not located near the parallel path Pp (for example, the row located between the image plane IF1 on the parallel path Pp1 and the image plane IF2 on the parallel path Pp2) cannot be acquired as a sample. , There is a possibility that a local defect occurring in a row not located near the parallel path Pp cannot be detected.

Further, in the sampling example of FIG. 5, the sample is not acquired in the connection path Pc included in the entire path P1. Even when the imaging data is acquired in the connection path Pc, the data near the boundary of the detection target area 310 is not suitable for the analysis material in the detection target area 310, so the imaging data acquired in the connection path Pc is statistical. It is preferable not to use it as an analytical sample.
Therefore, it can be said that the connection path Pc is an extra path for sampling, but the flying object 200 has no choice but to move on the connection path Pc when moving from the parallel path Pp to the next adjacent parallel path Pp. It can be said that sampling using the entire path P1 is not efficient in that an extra path is included due to the convenience of movement.

Subsequently, an example of sampling according to the present embodiment will be described with reference to FIG. The field 300 shown in FIG. 6 is the same as the field 300 shown in FIG. 5, and in the detection target area 310 shown in FIG. 6, a plurality of waxes extend in the row direction Dr and are parallel to the row arrangement direction Dj.

The entire path P2 is set in the detection target area 310 shown in FIG. The aircraft body 200 flies along a partial path included in the entire pass P2 from the start point to the end point, with one end of the entire pass P2 as the start point P2s and the other end of the entire pass P2 as the end point P2e.

The entire path P2 includes a plurality of inclined paths Pd (Pd1 to Pd8) as partial paths passing through the detection target area.
The inclined path Pd is a line segment from the start point to the end point on the outer edge of the detection target area 310, and is inclined with respect to the row direction Dr. The ramp path Pd crosses at least two rows.
Further, each inclined path Pd shown in FIG. 6 is a line segment that crosses the detection target area 310, and is provided so as to cross all the rows arranged in the row arrangement direction Dj. Further, the inclined path Pd is formed by a straight line segment.

In the inclined path Pd shown in FIG. 6, adjacent inclined paths Pd and Pd are connected to each other at an end portion. Further, the adjacent inclined paths Pd and Pd are inclined in different directions and are made non-parallel.
The entire path P2 composed of such a plurality of inclined paths Pd has a zigzag shape as a whole.
The aircraft body 200 follows a continuous partial path in order, such as the inclined path Pd1, the next inclined path Pd2 connected to the inclined path Pd1, and the next inclined path Pd3 connected to the inclined path Pd2, and the entire path P2. Fly the flight path of.

Note that the two adjacent inclined paths Pd and Pd are positioned so as to form a predetermined inclined path angle. The details of the tilt path angle will be described later.

In the sampling example of FIG. 6, while the flying object 200 traverses the detection target area 310 along each inclined path Pd of the entire path P2, the imaging data of the image plane IF is acquired as a sample. The plurality of image plane IFs are evenly spaced on each inclined path Pd.

Since the sloping pass Pd crosses at least two waxes, one sloping pass can provide a sample covering a plurality of waxes. As a result, for example, the possibility of detecting a local defect occurring in a specific row is increased, and the sampling accuracy can be improved.
Further, since the inclined path Pd shown in FIG. 6 crosses all the rows arranged in the row arrangement direction Dj, the probability that all the rows are to be sampled is increased.

Further, in the sampling example of FIG. 6, the connection path Pc is not included in the entire path P2. Therefore, efficient sampling can be performed by using the entire path P2 that does not include an extra path.

The sampling example shown in FIG. 6 is assumed to be carried out under the following conditions.
・ Field area including outer circumference: 40 hectares ・ Flight altitude: 15m
-Angle of view of the image surface: 50 degrees in the horizontal direction and 30 degrees in the vertical direction
-Width of image surface: width 14m, height 8m
-Image surface overlap rate: width direction variable (indefinite), vertical direction -20%
-Distance between samples: 56 m in width direction, 1.6 m in length direction,
・ Overall path length: 4621m
・ Inclined path angle: 14 degrees ・ Coverage rate: 14%

The length of the entire path P2 in the sampling example of FIG. 6 is 4621 m, but the length of the overall path P1 in the sampling example of FIG. 5 targeting the same detection target area 310 is 6225 m. That is, the whole pass P2 is a pass having a length shorter than that of the whole pass P1, and sampling for the same range can be efficiently performed.
Further, when the entire paths P1 and P2 are used as the flight paths of the flying object 200 as in the examples of FIGS. 5 and 6, the flight distance is shortened as the length of the paths becomes shorter. Therefore, by shortening the flight distance by using the entire pass P2, it is possible to suppress the consumption of the charging power of the fuel and the battery used for the flight.

<4-2. Coverage rate>
The path used for sampling the detection target area 310 as described above is set based on a desired coverage rate according to the purpose of sampling. Therefore, the coverage rate in sampling will be described below with reference to FIG. 7.

The coverage rate in the present embodiment is a ratio indicating how much area is acquired as the image plane IF with respect to the area of the detection target area 310.

The coverage rate in sampling using the parallel path Pp will be described with reference to FIG. 7A.
FIG. 7A shows a part of two parallel paths Pp1 and Pp2 adjacent to each other with the track interval TI in between.

In sampling using the parallel path Pp shown in FIGS. 5 and 7A, the ratio of the image plane IF in the unit grid UG can be obtained as an intuitive index corresponding to the coverage rate. Hereinafter, this ratio is referred to as a coverage rate in the unit grid UG.
The unit grid UG is a rectangular area including only one image plane IF, and is a vertical width consisting of the vertical width lv of the image plane IF and the interval from the image plane IF to the next image plane IF in the extension direction of the path. It is defined by Lv, the width lh of the image plane IF, and the width Lh consisting of the interval from the image plane IF to the next image plane IF in the arrangement direction of the parallel paths Pp.
In the embodiment in which the flying object 200 performs sampling, the vertical width Lv is a value obtained by multiplying the speed of the flying object 200 by the sampling interval time, and the next image surface IF is obtained from the acquisition of the image surface IF. It corresponds to the distance traveled in the time until acquisition. Further, the width Lh corresponds to the length of the track interval TI.

The coverage rate in the unit grid UG is calculated based on the coverage rate in the vertical width direction and the coverage rate in the horizontal width direction of the unit grid UG.

The coverage rate in the vertical width direction and the coverage rate in the horizontal width direction are calculated based on the overlap rate of the image plane IFs in each direction.
The overlap rate of the image plane IF is a ratio indicating how much the image plane IF overlaps with the adjacent image plane IF. When the vertical width lv of the image surface IF is 100%, for example, when the overlap rate in the vertical width direction is 10%, the vertical width lv of the image surface IF and the adjacent image surface IF overlap by 10%. show.
In sampling for the purpose of taking samples discretely, the duplication rate is set to a negative value for convenience. The overlap rate of negative values indicates a state in which the image plane IFs are separated from each other. For example, when the overlap rate of the image plane IF in the vertical width direction is -300%, the image plane IF is separated from the adjacent image plane IF in the vertical width direction by a distance of 300% of the vertical width lv. Show that you are doing. That is, it is shown that there is a distance of three vertical widths of the image surface IF from the image surface IF to the next image surface IF.

The coverage rate in the vertical width direction (Front Coverage Rate) and the coverage rate in the horizontal width direction (Side Coverage Rate) are calculated by the respective formulas shown in Equation 1 below. The overlap rate in Equation 1 is an integer less than 100.

For example, when the overlap rate of the image plane IF in the vertical width direction is -300%, 25% is calculated as the coverage rate in the alignment direction.

As shown in the following formula, the coverage rate (Unit Coverage Rate) of the unit as a whole is the coverage rate in the vertical width direction (Front Coverage Rate) of the image surface IF and the coverage rate (Side Coverage) in the horizontal direction of the image surface IF. It is calculated by multiplying Rate).
(Unit Coverage Rate) = (Front Coverage Rate) × (Side Coverage Rate)
For example, when the coverage rate in the vertical width direction and the coverage rate in the horizontal width direction are 25% each, 6.25% is calculated as the coverage rate of the entire unit unit.

When sampling is performed using the entire path P1 including the parallel path Pp, the desired coverage rate is received from the user, and the length of the track interval TI is determined so that, for example, a unit unit that realizes the coverage rate is formed. It is conceivable to generate a parallel path Pp.

Subsequently, with reference to FIG. 7B, the coverage rate in sampling using the inclined path Pd will be described.
FIG. 7B shows adjacent inclined paths Pd1 and Pd2. The inclined paths Pd1 and Pd2 are paths from the start point to the end point on the outer edge of the boundary shape Bs, which will be described later. The inclined paths Pd1 and Pd2 are connected on one side forming the outer edge of the boundary shape Bs, and the distance between the inclined path Pd1 and the inclined path Pd2 is maximized on the side facing the one side. In FIG. 7B, the maximum distance between the inclined pass Pd1 and the inclined path Pd2 is represented by the maximum spacing Tmax, and the length from one side of the boundary shape Bs to the opposite side is represented by the distance D.

In sampling using the inclined paths Pd shown in FIGS. 6 and 7B, since the intervals between the adjacent inclined paths Pd1 and Pd2 are not constant, the overlap rate of the adjacent image plane IFs in the width direction is also not constant. Therefore, when sampling using the inclined path Pd, it is conceivable to use the value of the minimum coverage rate Rmin, which is the minimum coverage rate in the adjacent inclined paths Pd1 and Pd2, instead of the coverage rate in the unit grid UG. ..

The minimum coverage rate Rmin is a ratio calculated by the formula shown in Equation 2 below using the maximum interval Tmax and the width lh of the image plane IF. The Image Width in the formula represents the width lh.

The minimum coverage rate Rmin is used to set the tilted path angle α formed by the adjacent tilted paths Pd1 and Pd2.
The inclination path angle α is calculated by the formula shown in Equation 3 below using the maximum interval Tmax and the distance D from one end to the other end of the boundary shape.

When sampling is performed using the entire path P2 including the inclined path Pd, the desired minimum coverage rate Rmin is received from the user, and the inclined path Pd is generated based on the inclined path angle α that realizes the minimum coverage rate Rmin. ..

The minimum coverage rate Rmin is the minimum coverage rate, but when there are a plurality of image surface IFs on the inclined path Pd, the coverage rate is increased by the number of image surface IFs. This ratio can be obtained from the speed sp of the flying object 200 and the shutter interval si, and further, the maximum interval Tmax can be obtained and the inclination path angle α can be obtained.
Letting such a coverage rate be the coverage rate Ra, it can be expressed by the following equation I.
Formula I: n × IF / (Tmax × D) ≧ Ra
That is, assuming that the number of image surface IFs located on the inclined path Pd1 and the inclined path Pd2 is n and the area of the image surface IF is IF, the total area of all the image surface IFs on the inclined path Pd is defined as the maximum interval Tmax. The coverage rate Ra can be obtained by dividing by the area of the rectangle consisting of the distance D (maximum interval Tmax × distance D). Since the image plane IFs may overlap at the folded apex portion and the calculated coverage rate may be larger than the actual coverage rate Ra, “≧” is used in the formula I.
Further, assuming that the speed sp (per second) of the flying object 200 and the shutter interval si (second) of the imaging device 220, the aircraft exists on two inclined paths Pd (that is, 2 × inclined path Pd1) according to the following equation II. The number n of image plane IFs can be represented.
Equation II: n = 2 × √ {(Tmax / 2) 2 + D2)} ÷ sp ÷ si
That is, the value obtained by dividing the length of one inclined path Pd1 by the speed sp and the shutter interval si is multiplied by 2.
By substituting this n into the above equation I, a conditional equation of the maximum interval Tmax that satisfies the coverage rate Ra can be obtained. By calculating the path angle α that satisfies the coverage rate Ra, the inclination of the inclination path Pd can be determined.

<5. Path generation flow>
The flow of path generation in the present embodiment will be described with reference to FIGS. 8 to 13. In the following, the flow of generating the entire path P3 including the inclined path Pd will be described with respect to the detection target area 410 which is the field 400.

FIG. 8 shows a detection target area 410 to be sampled.
In generating the path, first, the shape of the detection target area 410 is specified. Specifically, at least the outer edge (boundary line) that defines the shape of the detection target area 410 is specified.
When specifying the detection target area 410, the wax direction Dr and the row arrangement direction Dj provided in the detection target area 410 may also be specified. In the detection target area 410 shown in FIG. 8, it is assumed that the wax extends in the row direction Dr in the drawing and is parallel to the row arrangement direction Dj.

FIG. 9 shows a state in which the detection target area 410 is surrounded by the boundary shape Bs.
When the detection target area 410 is specified, the boundary shape Bs is set according to the shape of the specified detection target area 410. The boundary shape Bs is the smallest shape that surrounds the detection target area 410, and has, for example, the smallest polygonal shape that surrounds the detection target area 410. The boundary shape Bs shown in FIG. 9 has a rectangular shape including a short side Ss extending in the row direction Dr and a long side Sl extending in the row arrangement direction Dj.
The boundary shape Bs is a frame composed of auxiliary lines used for convenience when generating the inclined path Pd. The shape of the boundary shape Bs is not limited to the polygonal shape, and may be another shape such as a shape including an arc-shaped line segment as long as it is a shape that is convenient for generating the inclined path Pd.

FIG. 10 shows a state in which the inclined path Pd is generated in the boundary shape Bs. The inclined path Pd extends in the direction extending from one short side Ss of the boundary shape Bs to the other short side Ss. In the following, the direction in which the inclined path Pd crosses the boundary shape Bs in this way is referred to as the path crossing direction Dt. In FIG. 10, the path crossing direction Dt is the same direction as the row arrangement direction Dj.
When the boundary shape Bs surrounding the detection target area 410 is set, a process of generating an inclined path Pd in the boundary shape Bs is performed. The inclined path Pd is generated based on a predetermined inclined path angle α calculated based on the path crossing direction Dt and the above-mentioned minimum coverage rate Rmin.
Specifically, for example, from the midpoint M on one short side Ss of the boundary shape Bs, inclined paths Pd4 and Pd5 having a predetermined inclined path angle α are generated based on the path crossing direction Dt. Subsequently, from the intersection of each of the inclined paths Pd4 and Pd5 with the other short side Ss, the inclined paths Pd3 and Pd6 forming a predetermined inclined path angle α with each of the inclined paths Pd4 and Pd5 are generated. For example, by adding the inclined paths Pd one after another within the frame of the boundary shape Bs in this way, a plurality of inclined paths Pd are generated from the long side Sl of the boundary shape Bs to the other long side Sl.
The procedure for generating the inclined path Pd described above is an example, and of course, the inclined path Pd may be generated in the boundary shape Bs by another procedure.

FIG. 11 shows a state in which the inclined path Pd is superimposed on the detection target area 410.
When the inclined path Pd is generated in the boundary shape Bs, the generated inclined path Pd is superposed on the detection target area 410. By applying the inclined path Pd to the detection target area 410, the intersection Ip of the outer edge (boundary line) of the detection target area 410 and the inclined path Pd can be obtained.

FIG. 12 shows a state in which the intersection Ip is extracted as the passing point Wp of the flying object 200.
When the intersection Ip of the detection target area 410 and the inclined path Pd is obtained, the obtained intersection Ip is extracted as the passing point Wp to be passed by the flying object 200.
The passing point Wp is a point that defines the flight path of the flying object 200 that senses the detection target area 410. The passing point Wp can be expressed as GPS (Global Positioning System) position information indicating a specific point of the field 400, for example.

FIG. 13 shows a state in which the entire path P3 is generated.
When the passing point Wp is extracted on the outer edge of the detection target area 410, the passing point Wp is connected to generate the entire path P3.
Specifically, for example, by generating a line segment connecting the passing points Wp and Wp in the path crossing direction Dt, an inclined path Pd as a partial path from the start point to the end point on the outer edge of the detection target area 410 is obtained. Depending on the shape of the detection target area 410, an inclined path Pd in which a part of the intermediate portion of the path is located outside the detection target area 410 may be generated, for example, an inclined path Pd1 or an inclined path Pd2.
Among the inclined paths Pd thus obtained, there are some inclined paths Pd5 and Pd6 that are adjacent to each other but whose ends are not connected to each other, such as inclined paths Pd5 and Pd6. In this case, a connection path Pc is generated as a line segment connecting the passing points Wp and Wp which are the ends of the adjacent inclined paths Pd and Pd, and the adjacent inclined paths Pd and Pd are connected by the connecting path Pc.
By connecting the passing points Wp, a plurality of inclined paths Pd and connection paths Pc are obtained, and these paths generate an uninterrupted overall path P3.

The entire path P3 generated as described above is output as the flight path of the flying object 200, and sampling is performed for the detection target area 410 while the flying object 200 flies through the entire path P3.

<6. Path generation process of the embodiment>
Subsequently, the path generation process performed by the information processing apparatus 1 will be described with reference to FIG.
The process shown in FIG. 14 is realized by the CPU 51 having the function of the path generation unit 2 shown in FIG.

In step S101, the CPU 51 receives from the user the designation of the detection target area 410 to be sampled.

In step S102, the CPU 51 accepts a selection from the user regarding an input method for inputting the shape of the specified detection target area 410. The shape of the detection target area 410 can be loaded from the data file or manually input by the user.

When the user chooses to load the shape of the detection target area 410 from the data file, the CPU 51 proceeds from step S102 to step S103. In step S103, the CPU 51 loads the data file of the detection target area 410 specified by the user from the storage area and specifies the shape of the detection target area 410.

When the user chooses to manually input the shape of the detection target area 410, the CPU 51 proceeds from step S102 to step S104. In step S104, the CPU 51 receives an input drawn by the user or the like and specifies the shape of the detection target area 410.

The CPU 51 identifies the shape of the detection target area 410 in step S103 or step S104, and then proceeds to the process in step S105. In step S105, the CPU 51 calculates the minimum shape surrounding the detection target area 410, and sets the calculated shape as the boundary shape Bs.

In step S106, the CPU 51 receives an input of parameter information used for path generation from the user. The parameter information used for path generation includes, for example, the flight altitude of the flying object 200, the minimum coverage rate Rmin, and the path crossing direction Dt.
The path crossing direction Dt is set to be different from the low direction Dr in the detection target area 410. This causes the ramp path Pd generated in subsequent steps to cross at least two rows.
Further, instead of having the user input the parameter information, the information processing apparatus 1 may appropriately set each value of the parameter information. By automating the input of parameter information, it is possible to save the trouble of input operation by the user and improve the convenience.

In step S107, the CPU 51 calculates the inclined path angle α based on the minimum coverage rate Rmin and the path crossing direction Dt acquired in step S106.

In step S108, the CPU 51 generates an inclined path Pd within the boundary shape Bs. The CPU 51 generates the inclined path Pd so that the adjacent inclined paths Pd and Pd form the inclined path angle α set in S107.
For example, as described with reference to FIG. 10, the CPU 51 generates an inclined path Pd starting from the midpoint M of the boundary shape Bs, and based on the inclined path angle α, further next from the generated inclined path Pd. The inclined path Pd is continuously generated.

In step S109, the CPU 51 applies the generated inclined path Pd to the detection target area 410, and extracts the intersection Ip where the outer edge (boundary line) of the detection target area 410 and the inclined path Pd intersect.

In step S110, the CPU 51 sets the extracted intersection Ip as a passing point Wp that the flying object 200 should pass in order to fly along the inclined path Pd.
When the entire path P3 is not used as a route for a moving body such as the flying object 200, the process proceeds to step S111 without performing the process for obtaining the passing point Wp.

In step S111, the CPU 51 performs a process of generating an entire path. The CPU 51 generates the entire path P3 by, for example, a process of connecting the extracted passing points Wp. If the process for obtaining the passing point Wp is not performed in step S110, the intersection Ip is connected.

In step S112, the CPU 51 presents the generated overall path P3 to the user together with the parameter information. The entire path P3 is presented to the user as the flight path of the aircraft 200.

In step S113, the CPU 51 accepts a user confirmation result as to whether or not the generated overall path P3 and parameter information are appropriate as a flight plan.

When the user's confirmation result that the flight plan is not appropriate is accepted in step S113, the CPU 51 returns to step S106 and accepts the input of parameter information from the user again.

When the user's confirmation result that the flight plan is appropriate is received in step S113, the CPU 51 proceeds to step S114. In step S114, the CPU 51 sets the entire path P3 as a flight path and saves it as a flight plan together with the parameter information.

In step S115, the CPU 51 performs a process of outputting the saved flight plan. For example, it is a process of converting a flight plan into a format that can be read by the control unit of the flight body 200, or transmitting the flight plan to the flight body 200 by the communication unit 60.

By performing the above processing, a path to be used as a flight path of the flying object 200 is generated.
The above processing example is an example, and other processing examples can be considered.

<7. Summary and transformation examples>
According to the above embodiment, the following effects can be obtained.
The path generation method of the embodiment is a path generation method for generating a path for sensing for a detection target area (detection target areas 310 and 410) in which rows, which are an array of detection target objects, are parallel, in the detection target area. A path generation process is performed to generate an entire path (overall paths P2, P3) so that the passing partial path (inclined path Pd) crosses at least two rows.
As shown in FIGS. 6 and 13, the path generation process generates the entire paths P2 and P3 including the inclined paths Pd that cross at least two rows. Since the inclined pass Pd crosses at least two rows, sensing data covering a plurality of rows can be acquired as a sample by one inclined pass Pd. As a result, the probability that each row will be sampled increases, and the sampling accuracy can be improved.
For example, as described with reference to FIG. 6, by increasing the probability that each row will be the target of sensing, it is possible to improve the accuracy of detecting a local defect that has occurred in a specific row. By improving the detection accuracy of local defects, it becomes difficult to underestimate or overestimate local defects in statistical analysis. Therefore, for example, when the detection target area is a field and a crop planting defect occurs in a specific wax, it is possible to appropriately detect the planting defect of the specific wax and take appropriate measures such as replanting. That is, it is possible to avoid disadvantages such as a decrease in yield due to oversight of poor planting. Alternatively, unnecessary replanting based on false estimates of the extent of poor planting can be avoided and cost increases can be avoided.
Further, as described by comparing FIG. 5 and FIG. 6, by using the entire path P2 including the inclined path Pd, sampling can be efficiently performed with a path having a short length. For example, when the entire path P2 is used as a movement path for a moving body such as a flying object 200 or a vehicle, the shorter the path length, the shorter the moving distance. Therefore, by shortening the travel distance by using the entire path P2 including the inclined path Pd, it is possible to suppress the consumption of fuel and electric power used as a power source for the movement. Further, it becomes possible to perform sampling for a wider range of detection target areas with the same amount of fuel or electric power.
In addition to planting with a sowing machine, there is also a method of planting seeds at random in the field called "sowing", in which case the above "wax" is not formed. Since it is possible to similarly generate a path for sensing in order to obtain the germination rate and the like even in a field where wax is not formed, this method can be applied even when there is no wax. For example, by using the whole path including the inclined path as a partial path, it is possible to increase the probability that the field is randomly detected and each point in the field is sampled. Further, by using the whole path including the inclined path as a partial path, sampling can be efficiently performed with a short path even in a field where wax is not formed.

In the embodiment, as shown in FIGS. 6 and 13, at least a part of the partial paths (inclined paths Pd) in the entire paths (overall paths P2 and P3) are all in the detection target area (detection target areas 310 and 410). I gave an example of a path that crosses the row of.
By crossing all the rows in the detection target area with at least a part of the inclined path Pd, all the rows have the same probability of being the target of sensing, and the sampling accuracy can be further improved.

In the embodiment, the path generation process performs the first step (step S105) of setting the boundary shape Bs surrounding the detection target area (detection target area 410) and the partial path (inclined path Pd) within the boundary shape Bs. The second step (step S108) to generate, the third step (step S109) to extract the intersection Ip of the outer edge of the detection target area and the partial path generated in the second step, and the intersection (set as the passing point Wp). An example has been given with a fourth step (step S111) of generating the entire path P3 by the process of connecting the intersections Ip) (see FIG. 14).
By using the boundary shape Bs, the inclined path Pd can be generated regardless of the shape of the detection target area 410. Therefore, the path generation process can be performed corresponding to the detection target area having various shapes.
Further, by connecting the intersection Ip, which is a point on the outer edge of the detection target area 410, an inclined path Pd from the start point to the end point on the outer edge of the detection target area 410 can be obtained. Therefore, it is possible to generate an inclined path Pd according to the shape of the detection target area 410.
Further, in the fourth step, when a whole path is generated without providing an inclined path (for example, inclined paths Pd1 and Pd2 in FIG. 13) in which a part of the intermediate portion of the path is located outside the detection target area 410. For example, as shown in FIG. 6, it is possible to generate the entire path P3 in which all the inclined paths Pd are within the range of the detection target area 310. For example, when the entire path P3 is used as a movement path (flight path) for the flying object 200 or the like, the moving object does not cover the area outside the detection target area 310 by keeping the moving path of the moving body within the detection target area 310. You won't have to move as needed.
It is possible to apply this method to generate an entire path even in a field where wax is not formed (detection target area).

In the embodiment, an example in which the boundary shape Bs is rectangular is given (see FIG. 9 and the like).
By generating the inclined path Pd using the rectangular boundary shape Bs, the inclined path Pd can be easily generated regardless of the shape of the detection target area 410.

In the embodiment, in the second step (step S108) of the path generation process, a partial path is generated so that the adjacent partial paths (inclined paths Pd, Pd) form a predetermined path angle (inclined path angle α). An example is given (see FIG. 7).
By using the predetermined inclined path angle α, the inclined path Pd can be easily generated in the boundary shape Bs.

In the embodiment, an example of setting a predetermined path angle so as to satisfy a predetermined minimum coverage rate has been described (see FIG. 7).
As described with reference to FIG. 7, the desired minimum coverage rate Rmin is realized by calculating the inclined path angle α based on the minimum coverage rate Rmin and generating the inclined path Pd forming the inclined path angle α. Sampling can be done.
It is also conceivable to set the tilt path angle α so as to satisfy a predetermined coverage rate Ra. In this case, by generating the inclined path Pd having the inclined path angle α, sampling that realizes a desired coverage rate can be performed.

In the fourth embodiment, an example is given in which the intersection points Ip in the adjacent partial paths (inclined paths Pd) are connected to generate the connection path Pc (see FIG. 13).
As a result, the adjacent inclined paths Pd and Pd are connected by the connection path Pc, and the entire path P3 without interruption can be generated. For example, when the entire path P3 is used as the movement path of a moving body such as the flying object 200, it is not necessary to connect the inclined path Pd again to generate one path in which the moving body can move without interruption.

In the embodiment, an example in which the partial path (inclined path Pd) is formed by a line segment having a straight line is given (see FIGS. 6 and 13). As a result, the shortest line segment from one point to the other can be generated as the inclined path Pd. For example, when the inclined path Pd is used as a moving path of a moving body such as a flying object 200, the moving distance of the moving body can be shortened.
Further, the partial path (inclined path Pd) may be formed by a line segment which is a curved line. For example, FIG. 15A shows a state in which the entire path P4 including the inclined path Pd formed by the curved line segment is set in the detection target area 310. When the curved inclined path Pd is used, the probability of being sampled at each point of the detection target area 310 is further increased. Therefore, the sampling accuracy can be further improved.

In the embodiment, as one of the paths included in the entire path (overall path P3), an example in which the connection path Pc along the outer edge of the detection target area 410 is included is given (see FIG. 13).
By providing the connection path Pc along the outer edge, for example, when the entire path P3 is used as the movement path of the moving body such as the flying object 200, the moving body unnecessarily moves in the area outside the detection target area 410. There is no.

In the embodiment, as one of the paths included in the entire path (overall path P2), an example in which the connection path Pc along the outer edge of the detection target area (detection target area 310) is not included is given (see FIG. 6 and the like). ).
Since the connection path Pc along the outer edge of the detection target area 310 is not included, the distance of the entire path P2 can be shortened.
Further, it is preferable that the sensing data acquired by the connection path Pc along the outer edge of the detection target area 310 is not used as a sample for statistical analysis. Therefore, by not providing the connection path Pc, it is possible to save the trouble of acquiring and storing the sensing data that is not used for the analysis, and to reduce the processing load of the information processing apparatus that executes sampling and analysis.

In the embodiment, an example of generating an entire path P2 in which adjacent partial paths (inclined paths Pd, Pd) are connected at an end is given (see FIG. 6 and the like).
For example, as shown in FIGS. 6 and 15A, by connecting adjacent inclined paths Pd and Pd that cross the detection target area 310 at the ends, sensing can be performed on all the paths included in the entire path. can.

In the embodiment, an example of generating an entire path (overall paths P2, P3) in which adjacent partial paths (inclined paths Pd, Pd) are non-parallel is given (see FIGS. 6 and 13).
As a result, for example, the adjacent inclined paths Pd and Pd are inclined in different directions. Therefore, for example, it is possible to generate the entire paths P2 and P3 having a zigzag shape at least in part.
It should be noted that the entire path in which the adjacent inclined paths Pd and Pd are parallel may be generated. For example, FIG. 15B shows a state in which the entire path P5 is generated so that a plurality of adjacent inclined paths Pd from the start point to the end point on the outer edge of the detection target area 310 are parallel to the detection target area 310. In the example of FIG. 15B, the ends of the adjacent inclined paths Pd and Pd are connected by the connection path Pc along the outer edge of the detection target area 310.

In the embodiment, the detection target area (detection target area 310, 410) is a field (field 300, 400), and the wax is a line in which the crop is planted in the field.
As a result, sensing is performed on the fields 300 and 400 in which the lines where the crops are planted are parallel, and sampling can be performed on the fields 300 and 400.
Sampling in the field can be carried out for various purposes such as estimation of the number of crops (stands) planted and germinated in the field, weed detection, detection of plant diseases and water stress, and the like.
The field may be not only cultivated land of outdoor farmland but also land for hydroponics or house cultivation, and the technique of the embodiment can be used for sampling of crops grown in various places. can.
In addition to crops, the technique of the embodiment can also be used for sampling sensing data related to the growth of, for example, fruit trees, trees used as wood, and weeds. Therefore, it can also be applied to remote sensing for forests, vacant lots, and the like.
As an example of the sensing data, it is assumed that the image image data is captured by the imaging device 220, but of course, the detection data of various sensors such as the detection data by the thermo sensor and the detection data by the ultrasonic sensor is assumed.
Further, the technique of the embodiment can be applied not only to plants but also to sensing the quality of soil in fields and the like.
The detection target area of the embodiment is not limited to the field, but various places and areas can be considered. For example, the technology of the embodiment can be used for an area that can be assumed on the water including forests, areas and plots affected by a disaster, the ocean, and the like.

In the embodiment, the entire path (overall paths P2, P3) is the flight path of the flying object 200, and an example in which sensing is performed by the flying object 200 has been described.
The aircraft body 200 includes a so-called drone, a small radio-controlled fixed-wing airplane, and a small radio-controlled helicopter.

The path generation device (information processing device 1) of the embodiment is a path generation device that generates a path for sensing for a detection target area (detection target areas 310, 410) in which rows, which are an array of detection target objects, are parallel to each other. , The path generation process for generating the entire path (overall path P2, P3) is performed so that the partial path (inclined path Pd) passing through the detection target area crosses at least two rows.

The program of the embodiment is executed by a path generator (information processing device 1) that generates a path for sensing for the detection target areas (detection target areas 310 and 410) in which rows, which are an array of detection target objects, are parallel to each other. The program causes the path generator to execute a path generation process for generating an entire path so that a partial path passing through the detection target area crosses at least two of the rows.
That is, it is a program that causes the information processing apparatus 1 to execute the process described with reference to FIG.

Such a program facilitates the realization of the information processing device 1 of the present embodiment.
Such a program can be stored in advance in a recording medium built in a device such as a computer device, a ROM in a microcomputer having a CPU, or the like. Alternatively, it can be temporarily or permanently stored (stored) in a removable recording medium such as a semiconductor memory, a memory card, an optical disk, a magneto-optical disk, or a magnetic disk. Further, such a removable recording medium can be provided as so-called package software.
In addition to installing such a program from a removable recording medium on a personal computer or the like, it can also be downloaded from a download site via a network such as a LAN or the Internet.

Note that the effects described in this specification are merely examples and are not limited, and other effects may be obtained.

The present technology can also adopt the following configurations.
(1)
As a path generation method for generating a path for sensing for a detection target area in which rows, which are an array of detection targets, are parallel,
A path generation method for performing a path generation process for generating an entire path so that a partial path passing through the detection target area crosses at least two of the rows.
(2)
The path generation method according to (1) above, wherein at least a part of the partial paths in the entire path is a path that crosses all the waxes in the detection target area.
(3)
The path generation process is
The first step of setting the boundary shape surrounding the detection target area and
A second step of generating the partial path within the boundary shape,
A third step of extracting the intersection of the outer edge of the detection target area and the partial path generated in the second step, and
The path generation method according to (1) or (2) above, further comprising a fourth step of generating the entire path by a process of connecting the intersections.
(4)
The path generation method according to (3) above, wherein the boundary shape is rectangular.
(5)
The path generation method according to (3) or (4) above, wherein in the second step, the partial paths are generated so that the adjacent partial paths form a predetermined path angle.
(6)
The path generation method according to (5) above, wherein the predetermined path angle is set so as to satisfy a predetermined minimum coverage rate.
(7)
In the third step, the intersection of the partial path and the detection target area is detected.
The path generation method according to any one of (3) to (6) above, wherein the intersections in the adjacent partial paths are connected to generate a connection path.
(8)
The path generation method according to any one of (1) to (7) above, wherein the partial path is formed by a line segment that is a straight line or a curved line.
(9)
The path generation method according to any one of (1) to (8) above, wherein a connection path along the outer edge of the detection target area is included as one of the paths included in the entire path.
(10)
The path generation method according to any one of (1) to (8) above, wherein the connection path along the outer edge of the detection target area is not included as one of the paths included in the entire path.
(11)
The path generation method according to any one of (1) to (9) above, which generates the entire path such that adjacent partial paths are connected to each other at an end.
(12)
The path generation method according to any one of (1) to (11) above, which generates the entire path such that adjacent partial paths are non-parallel.
(13)
The detection target area is a field,
The path generation method according to any one of (1) to (12) above, wherein the wax is a line on which crops are planted in the field.
(14)
The whole path is the flight path of the flying object.
The path generation method according to any one of (1) to (13) above, wherein the sensing is performed by the flying object.
(15)
As a path generator that generates a path for sensing to the detection target area where rows, which are an array of detection targets, are parallel.
A path generation device that performs a path generation process that generates an entire path so that a partial path passing through the detection target area crosses at least two of the rows.
(16)
It is a program to be executed by a path generator that generates a path for sensing for a detection target area in which rows that are a sequence of detection targets are parallel.
A program that causes the path generator to execute a path generation process that generates an entire path so that a partial path passing through the detection target area crosses at least two of the rows.

1 Information processing device 2 Path generator 310 Detection target area 410 Detection target area Bs Boundary shape Pd Inclined path Pc Connection path P2, P3, P4, P5 Overall path α Inclined path angle

Claims (16)

1. As a path generation method for generating a path for sensing for a detection target area in which rows, which are an array of detection targets, are parallel,
   A path generation method for performing a path generation process for generating an entire path so that a partial path passing through the detection target area crosses at least two of the rows.
2. The path generation method according to claim 1, wherein at least a part of the partial paths in the entire path is a path that crosses all the rows in the detection target area.
3. The path generation process is
   The first step of setting the boundary shape surrounding the detection target area and
   A second step of generating the partial path within the boundary shape,
   A third step of extracting the intersection of the outer edge of the detection target area and the partial path generated in the second step, and
   The path generation method according to claim 1, further comprising a fourth step of generating the entire path by a process of connecting the intersections.
4. The path generation method according to claim 3, wherein the boundary shape is rectangular.
5. The path generation method according to claim 3, wherein in the second step, the partial paths are generated so that the adjacent partial paths form a predetermined path angle.
6. The path generation method according to claim 5, wherein the predetermined path angle is set so as to satisfy a predetermined minimum coverage rate.
7. The path generation method according to claim 3, wherein in the fourth step, the intersections in the adjacent partial paths are connected to generate a connection path.
8. The path generation method according to claim 1, wherein the partial path is formed by a line segment that is a straight line or a curved line.
9. The path generation method according to claim 1, wherein a connection path along the outer edge of the detection target area is included as one of the paths included in the entire path.
10. The path generation method according to claim 1, wherein the connection path along the outer edge of the detection target area is not included as one of the paths included in the entire path.
11. The path generation method according to claim 1, wherein the entire path is generated so that adjacent partial paths are connected to each other at the ends.
12. The path generation method according to claim 1, wherein the entire path is generated such that the adjacent partial paths are non-parallel.
13. The detection target area is a field,
    The path generation method according to claim 1, wherein the wax is a line on which crops are planted in the field.
14. The whole path is the flight path of the flying object.
    The path generation method according to claim 1, wherein the sensing is performed by the flying object.
15. As a path generator that generates a path for sensing to the detection target area where rows, which are an array of detection targets, are parallel.
    A path generation device that performs a path generation process that generates an entire path so that a partial path passing through the detection target area crosses at least two of the rows.
16. It is a program to be executed by a path generator that generates a path for sensing for a detection target area in which rows that are a sequence of detection targets are parallel.
    A program that causes the path generator to execute a path generation process that generates an entire path so that a partial path passing through the detection target area crosses at least two of the rows.

Images