Unmanned aerial vehicle and unmanned aerial vehicle guidance system

Abstract

An unmanned aerial vehicle comprises: a communication unit; an imaging unit that captures images of a surrounding situation; and an aerial vehicle body that flies while autonomously maintaining an attitude. The communication unit receives, as a generated image, an image that is automatically generated by a generative AI server, the image being generated on the basis of information that indicates a feature of a target object. The aerial vehicle body sequentially acquires images that are captured by the imaging unit as surrounding images, flies while sequentially determining whether the surrounding images are similar to the generated image that is generated by the generative AI server, and determines that a point where the target object is installed has been reached when it is determined that the surrounding images are similar to the generated image.

Description

Unmanned aerial vehicle and unmanned aerial vehicle guidance system

[0001] This disclosure relates to an unmanned aerial vehicle and an unmanned aerial vehicle guidance system.

[0002] In recent years, the use of unmanned aerial vehicles represented by drones for transporting materials, inspecting and maintaining target objects, etc. has been put into practical use.

[0003] Unmanned aerial vehicles generally fly by manual operation by an operator. In this case, the operator directly visually observes the unmanned aerial vehicle or views the video from the camera mounted on the unmanned aerial vehicle, and manually operates the unmanned aerial vehicle using a controller.

[0004] On the other hand, it is also practiced to set the coordinate values of the target point in advance in the unmanned aerial vehicle and make the unmanned aerial vehicle fly automatically without human intervention.

[0005] Also, the following systems related to unmanned aerial vehicles are disclosed (for example, see Patent Document 1). The exterior lamp monitoring system according to Patent Document 1 is a system for sequentially inspecting exterior lamps installed at various locations outdoors, and includes an unmanned aerial vehicle and a monitoring server. The unmanned aerial vehicle is equipped with a camera and automatically flies along a predetermined flight route. The monitoring server determines whether there is an abnormality in the exterior lamp based on the image of the exterior lamp photographed by the unmanned aerial vehicle. And when the monitoring server determines that there is an abnormality in the exterior lamp, it reports to a previously stored reporting destination.

[0006] Japanese Unexamined Patent Application Publication No. 2023-#

[0007] When flying an unmanned aerial vehicle by manual operation, if the operator does not know the characteristics of the target object, the target point, etc., the operator needs to receive an explanation in advance or receive an explanation each time while checking the situation on site. Also, in the case of manual operation, corresponding flight skills are required to avoid contact with other objects.

[0008] When an unmanned aerial vehicle (UAV) is to fly automatically without human intervention, the coordinates of the destination point must be set in advance on the UAV. In the streetlamp monitoring system described in Patent Document 1, the route on which the UAV will fly must also be set in advance, so a similar setup process is required.

[0009] This disclosure was made to solve the above-mentioned problems and aims to provide an unmanned aerial vehicle (UAV) that can fly to the installation site of a target object by a simpler method, and an UAV guidance system having said UAV.

[0010] The unmanned aerial vehicle according to this disclosure comprises a communication unit, an imaging unit that images the surrounding environment, and a main body of the aircraft that flies autonomously while maintaining its attitude. The communication unit receives images as generated images, which are automatically generated by a server and are generated based on information indicating the characteristics of the target object. The main body of the aircraft sequentially acquires images captured by the imaging unit as surrounding images, and flies while sequentially determining whether the surrounding images are similar to the generated images generated by the server. If it is determined that the surrounding images are similar to the generated images, it determines that it has reached the location where the target object is installed.

[0011] According to this disclosure, an unmanned aerial vehicle (UAV) that flies to the installation site of a target object, and an UAV guidance system having said UAV, can be obtained by a simpler method.

[0012] This figure shows the operation overview of the unmanned aerial vehicle guidance system in Embodiment 1 of this disclosure in step order. This block diagram shows an example configuration of the unmanned aerial vehicle guidance system in Embodiment 1. This flowchart shows an example of the operation of the unmanned aerial vehicle in Embodiment 1. This figure shows the operation overview of the unmanned aerial vehicle guidance system in Embodiment 2 of this disclosure in step order. This flowchart shows an example of the operation of the unmanned aerial vehicle in Embodiment 2.

[0013] Embodiment 1. Figure 1 is a diagram showing the overview of the operation of the unmanned aerial vehicle guidance system 1 in Embodiment 1 of this disclosure, in step order.

[0014] The unmanned aerial vehicle guidance system 1 is a system that flies an unmanned aerial vehicle 100 to a target object designated by the user, the worker 50, and has the vehicle perform inspection or maintenance work.

[0015] In step S101 shown in Figure 1, the worker 50 verbally explains to the unmanned aerial vehicle 100 the characteristics of the target object to be inspected or maintained, and the details of the work to be done. The unmanned aerial vehicle 100, having received this explanation, extracts the characteristics of the target object from the information verbally explained by the worker 50.

[0016] Characteristics of the target object can include, for example, its name, shape, pattern, color, the name and characteristics of any structures surrounding it, and the relative positional relationship between the structures and the target object. Numerical characteristics of the target object can also be included; for example, if the target object is an escalator, the number of steps and its height can also be considered characteristics of the target object.

[0017] In step S102, the unmanned aerial vehicle 100 requests the generating AI server 200 to generate an image illustrating the characteristics of the target object, based on the information verbally explained by the operator 50.

[0018] For example, as shown in Figure 1, if the unmanned aerial vehicle 100 receives instructions from the worker 50 to "inspect the light fixture next to the vending machine on the bridge," it requests the generation AI server 200 to generate an image of "the light fixture located next to the vending machine installed on the bridge."

[0019] The unmanned aerial vehicle 100 then retrieves the image generated in response to this request from the generation AI server 200.

[0020] Subsequently, in step S103, the unmanned aerial vehicle 100 begins automatic flight and flies while searching for target objects similar to the images generated by the generating AI server 200.

[0021] If the unmanned aerial vehicle 100 finds a target object, in step S104, it begins inspection or maintenance work on the target object.

[0022] In this way, the unmanned aerial vehicle guidance system 1 in Embodiment 1 causes the generating AI server 200 to generate an image that matches the content of the explanation given by the worker 50. The unmanned aerial vehicle guidance system 1 then stores the generated image from the generating AI server 200 in the unmanned aerial vehicle 100 and automatically flies the unmanned aerial vehicle 100 to a location where a target object similar to the generated image is installed.

[0023] This allows the unmanned aerial vehicle 100 to be flown to the installation site of the target object using a simple method, and the target object can be inspected or maintained.

[0024] Figure 2 is a block diagram showing an example configuration of the unmanned aerial vehicle guidance system 1 in Embodiment 1.

[0025] As shown in Figure 2, the unmanned aerial vehicle guidance system 1 includes an unmanned aerial vehicle 100 and a generating AI server 200.

[0026] The unmanned aerial vehicle 100 is an aircraft that flies while maintaining its attitude autonomously. In Embodiment 1, the unmanned aerial vehicle 100 does not require human operation and performs autonomous flight while avoiding contact with obstacles by detecting obstacles on its own.

[0027] The unmanned aerial vehicle 100 comprises a communication unit 11, an imaging unit 16, a sound collection unit 17, an audio output unit 18, an inspection and maintenance unit 19, and the aircraft body 20.

[0028] The communication unit 11 communicates with the generation AI server 200 via other relay devices (not shown). As the communication means in this embodiment 1, for example, communication using a fourth-generation mobile communication system (4G), a fifth-generation mobile communication system (5G), or communication using a long-range wireless LAN (Local Area Network) may be employed.

[0029] The imaging unit 16 captures images of the surroundings of the unmanned aerial vehicle 100 in real time. The imaging unit 16 has cameras positioned to capture images of the four directions in front of, behind, to the right of, and to the left of the unmanned aerial vehicle 100.

[0030] Alternatively, the imaging unit 16 may have 360-degree cameras mounted on the top and bottom surfaces of the unmanned aerial vehicle 100. By mounting such 360-degree cameras on the top and bottom surfaces of the vehicle, it is possible to acquire images of the entire sphere in both the upper and lower directions.

[0031] The sound collection unit 17 has a microphone and collects ambient sounds. The sound collection unit 17 primarily collects the voices of the workers 50.

[0032] The audio output unit 18 has a speaker and outputs sound to the surrounding area. The audio output unit 18 mainly provides voice notifications to the worker 50.

[0033] The inspection and maintenance unit 19 is a module that performs inspection or maintenance work on the target object. The inspection and maintenance unit 19 includes, for example, LED lights that illuminate the surroundings or one direction, a micro camera that takes close-up shots from a nearby position, a laser device for distance measurement, and a robotic arm with a tool attached to its tip. The inspection and maintenance unit 19 incorporates various devices depending on the type of target object and the content of the inspection and maintenance.

[0034] The aircraft body 20 is the main body of the unmanned aircraft 100, including the frame and control board, and flies autonomously while maintaining its attitude. The aircraft body 20 has a control unit 12, a memory unit 13, an autonomous flight control unit 14, and a position information acquisition unit 15.

[0035] The control unit 12 performs various calculation processes and outputs command signals to various hardware components within the unmanned aerial vehicle 100 to control them. The control unit 12 has a configuration that includes a calculation processing unit, a main memory, and interfaces for controlling various hardware components.

[0036] The memory unit 13 stores various information such as programs and control parameters for flying the unmanned aerial vehicle 100, as well as various images.

[0037] The autonomous flight control unit 14 is a module that controls the unmanned aircraft 100 to fly while maintaining its attitude and avoiding contact with obstacles. The autonomous flight control unit 14 includes an inertial measurement unit, multiple object detection sensors, a control circuit, and multiple motors for rotating each rotor.

[0038] Here, object detection sensors are installed at least above and below the aircraft, and in the front, back, left, and right positions, so as to be able to detect obstacles. The control circuit receives acceleration, rotational angular velocity, and axial direction measurement signals from the inertial measurement unit, as well as detection signals from the object detection sensors, and controls the rotational speed of each motor based on these signals.

[0039] The position information acquisition unit 15 has a receiver for the Global Navigation Satellite System (GNSS) and acquires the current position of the unmanned aerial vehicle 100 as latitude and longitude coordinate values. These coordinate values ​​are referred to as "position information".

[0040] The generation AI server 200 is a server that receives various commands from users and automatically generates and returns text, images, videos, audio data, etc., as requested by the user. These commands from users are called "prompts."

[0041] In Embodiment 1, the generating AI server 200 receives a prompt for image generation from the unmanned aerial vehicle 100. Based on the information contained in the prompt, the generating AI server 200 generates an image illustrating the target object and its surroundings.

[0042] Furthermore, a third-party AI generation cloud may be used as the AI ​​generation server 200.

[0043] Figure 3 is a flowchart showing an example of the operation of the unmanned aerial vehicle 100 in Embodiment 1.

[0044] In step S201, the sound collection unit 17 collects the audio of the worker 50 verbally explaining the target object. The audio collected here is referred to as the "user explanation audio." The user explanation audio includes audio describing the characteristics of the target object, such as its name and shape.

[0045] In step S202, the airframe 20 creates a prompt for image generation as a command to cause the generation AI server 200 to generate an image.

[0046] The airframe 20 has a voice processing function that performs speech-to-text conversion processing for generating text from voice and speech synthesis processing for generating voice from text. By using this function, the airframe 20 converts the user explanation voice into text. Then, the airframe 20 creates a prompt for image generation so as to include the converted text.

[0047] In step S203, the communication unit 11 transmits the created prompt for image generation to the generation AI server 200. Then, in step S204, the communication unit 11 receives the generated image generated based on the prompt for image generation from the generation AI server 200.

[0048] The airframe 20 stores the generated image received by the communication unit 11 in the storage unit 13. Then, in step S205, the airframe 20 uses the autonomous flight control unit 14 to start a search flight without human operation. Note that the mode of performing the search flight is not particularly limited.

[0049] In step S206, the airframe 20 acquires an image obtained by imaging the surrounding situation in real time from the imaging unit 16. This image is referred to as the "surrounding image".

[0050] In step S207, the airframe 20 calculates the similarity between the surrounding image and the generated image while performing image processing such as geometric transformation processing on the obtained surrounding image. Here, the similarity is an index value indicating how similar the two images of the surrounding image and the generated image are. In the first embodiment, the higher the value, the more similar the two are.

[0051] In step S208, the airframe 20 determines whether or not the calculated similarity is equal to or greater than a preset threshold value. If the similarity is equal to or greater than the threshold value, the airframe 20 proceeds to step S209.

[0052] In step S209, the aircraft body 20 determines that it has found the target object described by the operator 50 because the surrounding image and the generated image are similar to each other.

[0053] Then, the aircraft body 20 makes fine adjustments, such as moving to a suitable position. After that, the inspection and maintenance unit 19 performs inspection or maintenance work on the target object in step S210. These processes may also be performed by manual operation.

[0054] On the other hand, if the similarity is below the threshold in step S208, the aircraft body 20 determines that it has not yet found the target object and returns to step S205 to continue the search flight.

[0055] Furthermore, some of the processing performed by the unmanned aerial vehicle 100 may be performed by a terminal device such as a personal computer or smartphone held by the operator 50. For example, in Embodiment 1, the processing up to acquiring the generated image from the generation AI server 200, that is, the processing in steps S101 and S102 in Figure 1, or the processing in steps S201 to S204 in Figure 2, may be performed by a terminal device such as a personal computer.

[0056] Furthermore, in Embodiment 1, the unmanned aerial vehicle 100 was flown to a specific location primarily for the purpose of inspection and maintenance. In contrast, the unmanned aerial vehicle 100 may be flown to a specific location for various purposes, such as transporting supplies, surveying disaster areas, and searching for resources.

[0057] The features of the unmanned aerial vehicle 100 and the unmanned aerial vehicle guidance system 1 in this embodiment 1 can be summarized as follows, and the effects can be achieved.

[0058] The unmanned aerial vehicle 100 in Embodiment 1 comprises a communication unit 11, an imaging unit 16 for capturing images of the surrounding environment, and an aerial vehicle body 20 that flies while autonomously maintaining its attitude.

[0059] The communication unit 11 receives the image automatically generated by the generation AI server 200 as a generated image. This generated image is an image generated based on information that indicates the characteristics of the target object.

[0060] The aircraft body 20 sequentially acquires images captured by the imaging unit 16 as surrounding images, and flies while sequentially determining whether the surrounding images are similar to the generated images generated by the generation AI server 200.

[0061] Then, if the main body of the aircraft determines that the surrounding image is similar to the generated image, it determines that it has reached the location where the target object is installed.

[0062] Therefore, it is possible to fly to the installation location of the target object using a simpler method.

[0063] Furthermore, the unmanned aerial vehicle 100 is equipped with a sound collection unit 17 that collects sounds from the surrounding area. The sound collection unit 17 collects the voice of the worker 50 when he explains the target object, and uses this voice as user explanation audio.

[0064] The aircraft body unit 20 creates a command to the generation AI server 200 to generate a generated image, based on the user's explanatory voice. The communication unit 11 transmits the created command to the generation AI server 200 and receives the generated image generated based on the command from the generation AI server 200.

[0065] Therefore, by giving verbal instructions to the unmanned aerial vehicle 100, it is possible to make the unmanned aerial vehicle 100 fly to the installation site of the target object.

[0066] The unmanned aerial vehicle 100 further includes an inspection and maintenance unit 19 that performs inspection or maintenance on the target object. The inspection and maintenance unit 19 performs inspection or maintenance on the target object after the aircraft body 20 has determined that it has reached the location where the target object is installed.

[0067] Therefore, the unmanned aerial vehicle 100 can be used for the purpose of inspecting or maintaining target objects.

[0068] Furthermore, a configuration including the generation AI server 200 and the unmanned aerial vehicle 100 having the above-described features can be provided as the unmanned aerial vehicle guidance system 1.

[0069] Embodiment 2. Figure 4 is a diagram showing the operation overview of the unmanned aerial vehicle guidance system 1 in Embodiment 2 of this disclosure in step order. The unmanned aerial vehicle guidance system 1 in Embodiment 2 flies the unmanned aerial vehicle 100 to the installation site of the target object by a method different from that of Embodiment 1.

[0070] Here, with reference to Figure 4, the operation of the unmanned aerial vehicle guidance system 1 will be explained in step order. Note that the configuration of the unmanned aerial vehicle 100 in Embodiment 2 is the same as the configuration shown in Figure 2.

[0071] Furthermore, the generation AI server 200 in Embodiment 2 performs a different process from the process in Embodiment 1. Specifically, the generation AI server 200 in Embodiment 2 receives an image, recognizes objects contained in the image, and transmits the names of the objects. This process is referred to as "object recognition processing."

[0072] First, the unmanned aerial vehicle 100 automatically performs a patrol flight while taking images of its surroundings using the imaging unit 16, as shown in steps S301 to S303.

[0073] Specifically, as shown in step S302, the unmanned aerial vehicle 100 transmits the images it has captured during its patrol flight to the generating AI server 200. The unmanned aerial vehicle 100 then receives the processing results from the object recognition process, that is, the names of the objects captured in the images, from the generating AI server 200. The unmanned aerial vehicle 100 then records the received object names in association with the location information of the object at the time it was captured.

[0074] The unmanned aerial vehicle 100 repeats the process shown in step S302 each time it moves. This allows the unmanned aerial vehicle 100 to create a list of objects it has discovered during its patrol flight through steps S301 to S303.

[0075] Subsequently, upon arriving at the designated patrol completion point, the unmanned aircraft 100 reports the objects obtained during the patrol to the operator 50 in step S304.

[0076] The worker 50 selects which of the objects found during the patrol should be inspected or maintained, and gives verbal instructions to the unmanned aerial vehicle 100.

[0077] The unmanned aerial vehicle 100 identifies the object to be inspected or maintained based on the instructions, and obtains location information corresponding to the name of the identified object from a list.

[0078] Subsequently, in step S305, the unmanned aircraft 100 automatically flies to the point indicated by the position information.

[0079] Thus, the unmanned aerial vehicle guidance system 1 in Embodiment 2 is configured to allow the operator 50 to select a target object from among the objects discovered by the unmanned aerial vehicle 100 during its patrol flight. The unmanned aerial vehicle guidance system 1 then flies the unmanned aerial vehicle 100 to the installation location of the target object selected by the operator 50.

[0080] This allows the unmanned aerial vehicle 100 to be flown to the installation site of the target object using a simple method, and the target object can be inspected or maintained.

[0081] Figure 5 is a flowchart showing an example of the operation of the unmanned aerial vehicle 100 in Embodiment 2.

[0082] In step S401, the aircraft body 20 starts an automatic patrol flight. The unmanned aircraft 100 may fly along a predetermined route as its patrol route, or it may conduct a patrol flight only within a predetermined range, following certain flight rules.

[0083] In step S402, the aircraft body 20 uses the imaging unit 16 to capture images of its surroundings in real time. The aircraft body 20 also uses the position information acquisition unit 15 to acquire position information indicating the position at the time of imaging by the imaging unit 16.

[0084] In step S403, the aircraft body 20 creates a predetermined prompt for object recognition. Then, in step S403, the communication unit 11 transmits the created object recognition prompt along with the captured image to the generation AI server 200.

[0085] In step S404, the communication unit 11 receives from the generation AI server 200 the names of the objects captured in the image as a result of object recognition processing by the generation AI server 200.

[0086] In step S405, the aircraft body 20 acquires the name of the object received by the communication unit 11 and registers it in association with the position information acquired when the object was imaged.

[0087] In step S406, the aircraft body 20 determines whether it has reached a predetermined patrol end point. The aircraft body 20 makes the determination in step S406 by comparing the current location information obtained from the location information acquisition unit 15 with the predetermined location information of the patrol end point.

[0088] If the aircraft body 20 determines that it has reached a predetermined patrol end point, it proceeds to step S407. On the other hand, if the aircraft body 20 does not determine that it has reached a predetermined patrol end point, it returns to step S401 and continues the patrol flight.

[0089] In this way, steps S401 to S405 are repeated until the patrol end point is reached. As a result, the aircraft body 20 can sequentially record the objects it discovers during the patrol flight and add the names of the objects to the list.

[0090] In step S406, it is stated that the aircraft determines whether or not it has reached the patrol end point, but it is not limited to this. The aircraft body 20 may, for example, determine whether or not a predetermined patrol end time has elapsed, and if it determines that the patrol end time has elapsed, it may terminate the patrol flight and head towards the patrol end point.

[0091] After reaching the patrol completion point, the aircraft body 20 notifies the operator 50 of the name of the object registered in the list in step S407. At this time, the voice output unit 18 reads out the name of the object registered in the list by voice.

[0092] The sound collection unit 17 collects the voice emitted by the worker 50. The voice collected here is the voice that the worker 50 selects from among the object names read out by the voice output unit 18, and is referred to as the "user-selected voice".

[0093] User-selectable voice prompts include those that simply state the name of an object, such as "Please inspect the escalator," and those that state the order in which they are read, such as "Please inspect the third item on the list." As you can see, there are various types of user-selectable voice prompts.

[0094] Then, in step S408, the main body of the aircraft 20 identifies the name of the object selected by the operator 50 based on the user-selected voice collected by the sound collection unit 17. In step S409, the main body of the aircraft 20 attempts to obtain location information corresponding to the name of the identified object from a list.

[0095] For example, if the name of the object selected by the worker 50 is "escalator," the aircraft body 20 attempts in step S409 to see if it can obtain location information corresponding to "escalator" from the list.

[0096] If the corresponding location information cannot be obtained, such as when "escalator" does not exist in the list, the process returns to step S407. This causes the names of the objects registered in the list to be re-recorded.

[0097] The aircraft body 20 may repeat this regeneration process a specified number of times, and if it still cannot acquire position information, it may notify the system of this fact and terminate the process shown in the flowchart of Figure 5.

[0098] If the aircraft body 20 acquires positional information corresponding to the name of the object in step S409, in step S410 it flies to the point indicated by the acquired positional information. As a result, the aircraft body 20 can fly to the installation location of the target object selected by the operator 50.

[0099] Then, after the unmanned aerial vehicle 100 reaches the installation site of the target object, it makes fine adjustments such as moving to a suitable position. Then, in step S411, the inspection and maintenance unit 19 performs inspection or maintenance work on the target object. These processes may also be performed by manual operation.

[0100] In the second embodiment, some of the processing performed by the unmanned aerial vehicle 100 may also be performed by a terminal device such as a personal computer or smartphone held by the worker 50.

[0101] For example, in Embodiment 2, among the processes described above, the process of reporting to the worker 50 and obtaining the instructions, i.e., steps S304, S407, and S408, may be performed via a terminal device such as a personal computer. In other words, the means of having the worker 50 select an object from the list may be performed using the display device and input devices such as a touch panel and keyboard of the terminal device.

[0102] The features described in Embodiment 1 and Embodiment 2 above may be combined with each other.

[0103] The features of the unmanned aerial vehicle 100 and the unmanned aerial vehicle guidance system 1 in this embodiment 2 can be summarized as follows, and the effects can be achieved.

[0104] In Embodiment 2, the generating AI server 200 performs object recognition processing to recognize objects captured in the image and transmit the name of the object. In Embodiment 2, the aircraft body 20 performs a patrol flight while capturing images of its surroundings using the imaging unit 16.

[0105] Furthermore, the communication unit 11 transmits the captured images taken by the imaging unit 16 during the patrol flight, and receives the names of the objects captured in the captured images as a result of object recognition processing by the generation AI server 200.

[0106] The aircraft body 20 creates a list of objects discovered during its patrol flight by sequentially registering the names of objects received by the communication unit 11 in association with location information indicating the position of the object when it was captured by the imaging unit 16.

[0107] This makes it possible to understand what is present within the route that the unmanned aerial vehicle 100 has patrolled.

[0108] Furthermore, the aircraft body 20 notifies the operator 50 of the names of the objects registered in the list. The aircraft body 20 then identifies the name of the object selected by the operator 50 and obtains the location information corresponding to the name of the identified object from the list.

[0109] The aircraft body 20 then flies to the location indicated by the acquired position information.

[0110] Therefore, it is possible to fly to the installation location of the target object using a simpler method.

[0111] Furthermore, the unmanned aerial vehicle 100 in the second embodiment includes an audio output unit 18 that outputs sound to the surroundings and an audio collection unit 17 that collects surrounding sounds.

[0112] The audio output unit 18 outputs the names of the objects registered in the list by voice, and the sound collection unit 17 collects the user-selected voice, which is the voice emitted when the operator 50 selects one of the names of the objects output by the audio output unit 18.

[0113] The main body of the aircraft 20 identifies the name of the object selected by the operator 50 based on the user-selected voice collected by the sound collection unit 17, and obtains location information corresponding to the name of the identified object from a list. The main body of the aircraft 20 then flies to the location indicated by the obtained location information.

[0114] Therefore, by interacting with the unmanned aerial vehicle 100, it is possible to select a target object and fly the unmanned aerial vehicle 100 to the location where the target object is to be installed.

[0115] Furthermore, in the second embodiment as well, a configuration including a generating AI server 200 and an unmanned aerial vehicle 100 having the above-described features can be provided as an unmanned aerial vehicle guidance system 1.

[0116] Furthermore, as an application of the above, the following implementation is also possible. That is, one or more pieces of information indicating the characteristics of the target object are stored in the memory unit 13 of the unmanned aerial vehicle 100 in advance. An example of the characteristics of the target object is, if the target object is, for example, an escalator, the color of the escalator, the number of steps, and its height.

[0117] The aircraft body 20 flies while taking images of its surroundings using the imaging unit 16, and the communication unit 11 transmits the images captured by the imaging unit 16 during flight. In response to this transmission, the communication unit 11 receives the characteristics of objects captured in the images as a result of object recognition processing by the generation AI server 200.

[0118] Subsequently, if the main body of the aircraft 20 determines that the characteristics of the object are similar to the information stored in the memory unit 13, it determines that it has reached the location where the target object is installed.

[0119] This allows the system to fly to the location where the target object is located, based on the characteristics of the target object that have been previously stored in the memory unit 13.

[0120] Although preferred embodiments of unmanned aerial vehicles and unmanned aerial vehicle guidance systems have been described above, the system is not limited to the embodiments described above. Various modifications and transformations can be made to the unmanned aerial vehicles and unmanned aerial vehicle guidance systems described above without departing from the scope of the claims.

[0121] The various aspects of this disclosure are summarized below as appendices. (Appendix 1) An unmanned aerial vehicle comprising: a communication unit; an imaging unit for imaging the surrounding environment; and an aircraft body that flies autonomously while maintaining its attitude, wherein the communication unit receives as a generated image an image that is automatically generated by a server and is generated based on information indicating the characteristics of a target object, and the aircraft body sequentially acquires images captured by the imaging unit as surrounding images, and flies while sequentially determining whether the surrounding images are similar to the generated images generated by the server, and when it is determined that the surrounding images are similar to the generated images, it determines that it has reached the location where the target object is installed. (Note 2) The unmanned aerial vehicle according to Note 1, further comprising a sound collection unit for collecting ambient sounds, wherein the sound collection unit collects the sound of a user describing the target object as user description sound, the main body of the aircraft creates a command for the server to generate the generated image based on the user description sound, and the communication unit transmits the created command to the server and receives the generated image generated based on the command from the server. (Note 3) The unmanned aerial vehicle according to Note 1 or 2, further comprising an inspection and maintenance unit for performing inspection or maintenance on the target object, wherein the inspection and maintenance unit performs inspection or maintenance on the target object after the main body of the aircraft has determined to have reached the location where the target object is installed. (Note 4) The unmanned aerial vehicle according to any one of Notes 1 to 3, wherein the server further performs object recognition processing to recognize objects captured in the image and transmit the name of the object, the main body of the aircraft further performs a patrol flight while capturing images of the surroundings using the imaging unit, the communication unit transmits the captured images captured by the imaging unit during the patrol flight, the name of the object captured in the captured image is received by the server as a result of the object recognition processing, and the main body of the aircraft sequentially registers the name of the object received by the communication unit in association with position information indicating the position of the object when it was captured by the imaging unit, thereby creating a list of objects discovered during the patrol flight.(Note 5) The unmanned aerial vehicle according to Note 4, wherein the main body of the aircraft notifies the user of the names of the objects registered in the list, identifies the name of the object selected by the user, obtains location information corresponding to the identified name of the object from the list, and flies to the location indicated by the obtained location information. (Note 6) The unmanned aerial vehicle according to Note 5, further comprising: an audio output unit that outputs sound to the surroundings; and a sound collection unit that collects surrounding sounds, wherein the audio output unit outputs the names of the objects registered in the list by voice; the sound collection unit collects user-selected voice, which is the sound emitted when the user selects one of the names of the objects output by the audio output unit; the main body of the aircraft identifies the name of the object selected by the user based on the user-selected voice collected by the sound collection unit; obtains location information corresponding to the identified name of the object from the list, and flies to the location indicated by the obtained location information. (Note 7) An unmanned aerial vehicle guidance system comprising: a server that generates an image corresponding to the content desired by the user based on the content explained by the user; and an unmanned aerial vehicle described in any one of Notes 1 to 6.(Note 8) An unmanned aerial vehicle comprising: a communication unit; an imaging unit for imaging the surrounding environment; and an aircraft body that flies autonomously while maintaining its attitude, wherein the aircraft body performs a patrol flight while imaging the surrounding environment using the imaging unit; the communication unit transmits the images captured by the imaging unit during the patrol flight; receives the names of objects captured in the images as a result of object recognition processing by a server; the aircraft body creates a list of objects discovered during the patrol flight by sequentially registering the names of the objects received by the communication unit in association with location information indicating the position of the object when it was imaged by the imaging unit; after creating the list, notifies the user of the names of the objects registered in the list; identifies the name of the object selected by the user; obtains location information corresponding to the name of the identified object from the list; and flies to the point indicated by the obtained location information. (Note 9) An unmanned aerial vehicle guidance system comprising: a server that performs object recognition processing to recognize an object captured in an image and transmit the name of the object; and an unmanned aerial vehicle as described in Note 8. (Note 10) An unmanned aerial vehicle comprising: a communication unit; an imaging unit that captures images of the surrounding environment; an aerial vehicle body that flies while autonomously maintaining its attitude; and a storage unit that stores information indicating the characteristics of a target object, wherein the aerial vehicle body flies while capturing images of the surrounding environment using the imaging unit; the communication unit transmits the captured images captured by the imaging unit during the flight; the communication unit receives the characteristics of an object captured in the captured image as a result of object recognition processing by the server; and the aerial vehicle body determines that it has reached the location where the target object is installed if it determines that the characteristics of the object are similar to the information stored in the storage unit.

[0122] 1 Unmanned aircraft guidance system, 11 Communication unit, 12 Control unit, 13 Memory unit, 14 Autonomous flight control unit, 15 Position information acquisition unit, 16 Imaging unit, 17 Sound collection unit, 18 Audio output unit, 19 Inspection and maintenance unit, 20 Aircraft body unit, 25 Audio output unit, 50 Operator (user), 100 Unmanned aircraft, 200 Generation AI server.

Claims

1. An unmanned aerial vehicle comprising: a communication unit; an imaging unit for imaging the surrounding environment; and an aircraft body that flies autonomously while maintaining its attitude, wherein the communication unit receives as a generated image an image that is automatically generated by a server and is generated based on information indicating the characteristics of a target object; and the aircraft body sequentially acquires images captured by the imaging unit as surrounding images, and flies while sequentially determining whether the surrounding images are similar to the generated images generated by the server, and if it is determined that the surrounding images are similar to the generated images, it determines that it has reached the location where the target object is installed.

2. The unmanned aerial vehicle according to claim 1, further comprising a sound collection unit for collecting ambient sounds, wherein the sound collection unit collects the sounds of a user describing the target object as user description sounds, the main body of the aerial vehicle creates a command for the server to generate the generated image based on the user description sounds, and the communication unit transmits the created command to the server and receives the generated image generated based on the command from the server.

3. The unmanned aircraft according to claim 1 or 2, further comprising an inspection and maintenance unit for performing inspection or maintenance on the target object, wherein the inspection and maintenance unit performs inspection or maintenance on the target object after the aircraft body has determined that it has reached the location where the target object is installed.

4. The unmanned aerial vehicle according to any one of claims 1 to 3, wherein the server further performs object recognition processing to recognize an object captured in the image and transmit the name of the object; the main body of the aircraft further performs a patrol flight while capturing images of the surroundings using the imaging unit; the communication unit transmits the captured images captured by the imaging unit during the patrol flight; the communication unit receives the names of objects captured in the captured images as a result of the object recognition processing by the server; and the main body of the aircraft creates a list of objects discovered during the patrol flight by sequentially registering the names of the objects received by the communication unit in association with position information indicating the position of the object when it was captured by the imaging unit.

5. The unmanned aerial vehicle according to claim 4, wherein the main body of the aircraft notifies the user of the names of the objects registered in the list, identifies the name of the object selected by the user, obtains location information corresponding to the name of the identified object from the list, and flies to the location indicated by the obtained location information.

6. The unmanned aerial vehicle according to claim 5, further comprising: an audio output unit that outputs sound to the surroundings; and a sound collection unit that collects surrounding sounds, wherein the audio output unit outputs the names of objects registered in the list by voice; the sound collection unit collects user-selected voice, which is the sound emitted when the user selects one of the names of objects output by the audio output unit; and the aircraft body unit identifies the name of the object selected by the user based on the user-selected voice collected by the sound collection unit, obtains location information corresponding to the identified name of the object from the list, and flies to the location indicated by the obtained location information.

7. An unmanned aerial vehicle guidance system comprising: a server that generates an image corresponding to the content desired by the user based on the content described by the user; and an unmanned aerial vehicle according to any one of claims 1 to 6.

8. An unmanned aerial vehicle comprising: a communication unit; an imaging unit for imaging the surrounding environment; and an aircraft body that flies autonomously while maintaining its attitude, wherein the aircraft body performs a patrol flight while imaging the surrounding environment using the imaging unit; the communication unit transmits the images captured by the imaging unit during the patrol flight; receives the names of objects captured in the images as a result of object recognition processing by a server; the aircraft body creates a list of objects discovered during the patrol flight by sequentially registering the names of the objects received by the communication unit in association with location information indicating the position of the object when it was imaged by the imaging unit; after creating the list, notifies the user of the names of the objects registered in the list; identifies the name of the object selected by the user; obtains location information corresponding to the name of the identified object from the list; and flies to the point indicated by the obtained location information.

9. An unmanned aerial vehicle guidance system comprising: a server that performs object recognition processing to recognize an object captured in an image and transmit the name of the object; and an unmanned aerial vehicle as described in claim 8.

10. An unmanned aerial vehicle comprising: a communication unit; an imaging unit for imaging the surrounding environment; a main body of the aircraft that flies while autonomously maintaining its attitude; and a storage unit for storing information indicating the characteristics of a target object, wherein the main body of the aircraft flies while imaging the surrounding environment using the imaging unit; the communication unit transmits the images captured by the imaging unit during the flight; the main body of the aircraft receives the characteristics of objects captured in the images as a result of object recognition processing by a server; and the main body of the aircraft determines that it has reached the location where the target object is installed if it determines that the characteristics of the object are similar to the information stored in the storage unit.

Images