A Method for Diagnosing a Multi-Copter Based on an Artificial Intelligence Algorism

Abstract

The present invention relates to a method for diagnosing the condition of a multicopter based on an artificial intelligence algorithm. The method for diagnosing the condition of a multicopter based on an artificial intelligence algorithm comprises: a step of detecting the operating condition of a plurality of parts of a multicopter by an inspection device; a step of acquiring and processing detection data obtained by the inspection device; a step of learning data related to the detection of the inspection device by an artificial intelligence learning algorithm; and a step of diagnosing the condition of the multicopter by analyzing the processed data by the artificial intelligence learning algorithm based on the learning result, wherein the operating condition of each of the plurality of parts is detected by a sensor of at least one inspection device.

Description

Technology Field

[0001] The present invention relates to a method for diagnosing the condition of a multicopter based on an artificial intelligence algorithm, and specifically, to a method for diagnosing the condition of a multicopter based on an artificial intelligence algorithm that can diagnose the health status of a multicopter by learning and analyzing measurement information obtained by an inspection device using an artificial intelligence algorithm. Background Technology

[0002] A multicopter is an aircraft that uses two or more rotors or propellers to take off or land, or to propel or rotate, and can be used for various purposes in diverse industrial or military fields. Among such multicopters, quadcopters with four wings are widely used due to their stable flight conditions. A multicopter includes components such as a frame, propellers, an ESC (Electric Speed ​​Controller), motors, and batteries, and for the multicopter to operate stably, each component must be maintained in an appropriate operating state. Furthermore, to ensure the operational stability of the multicopter, the condition of each component must be checked periodically, and components must be repaired or replaced as necessary. To this end, the condition of each component needs to be detected periodically. Patent Publication No. 10-2012-0006807 discloses a safety diagnostic system for a helicopter rotor. Patent Registration No. 10-2303118 discloses a method for diagnosing thrust abnormalities in a multicopter. In addition, Patent Publication No. 10-2024-0093267 discloses an actuator diagnostic device for an unmanned multi-copter, a control system for an unmanned multi-copter, and a method for diagnosing actuator failures for an unmanned multi-copter. For the safe operation of a multi-copter, it is necessary to diagnose the condition of all components of the multi-copter. To this end, it is necessary to classify the components forming the multi-copter, determine parameters capable of measuring the operating state of each component, and select measurement sensors capable of measuring the determined parameters to collect data for condition diagnosis. Furthermore, it is necessary to accurately analyze the collected data using an artificial intelligence algorithm to determine the soundness of the multi-copter. However, the prior art does not disclose a technology for determining the soundness of the multi-copter by using an appropriate measurement device, determining parameters using an artificial intelligence algorithm, and analyzing detection data using an artificial intelligence algorithm.

[0003] The present invention aims to solve the problems of the prior art and has the following objectives. Prior art literature

[0004] Prior Art 1: Patent Publication No. 10-2012-0006807 (Korea Aerospace Industries, Inc., published Jan. 19, 2012) Safety diagnostic system for a helicopter rotor Prior Art 2: Patent Registration No. 10-2303118 (J-Maple Co., Ltd., published Sep. 16, 2021) Method for diagnosing thrust abnormalities in a multicopter Prior Art 3: Patent Registration No. 10-2024-0093267 (Industry-Academic Cooperation Foundation of Chungnam National University, published June 24, 2024) Actuator fault diagnosis device for an unmanned multicopter, unmanned multicopter control system, and method for diagnosing actuator faults in an unmanned multicopter The problem to be solved

[0005] The objective of the present invention is to provide an artificial intelligence algorithm-based multicopter condition diagnosis method that prepares a measurement device by selecting parameters for condition diagnosis and diagnoses the health status of a multicopter by analyzing detection data obtained from the measurement device using an artificial intelligence algorithm. means of solving the problem

[0006] According to a suitable embodiment of the present invention, a multicopter condition diagnosis method based on an artificial intelligence algorithm comprises: a step of detecting the operating status of a plurality of parts of a multicopter by an inspection device; a step of acquiring and processing detection data obtained by the inspection device; a step of learning data related to the detection of the inspection device by an artificial intelligence learning algorithm; and a step of diagnosing the state of the multicopter by analyzing the processed data by the artificial intelligence learning algorithm based on the learning result, wherein the operating status of each of the plurality of parts is detected by a sensor of at least one inspection device.

[0007] According to another suitable embodiment of the present invention, parameters for detecting the operating state of a component are extracted and set, and the setting of the parameters is performed by an artificial intelligence learning algorithm.

[0008] According to another suitable embodiment of the present invention, the inspection device comprises a fixed substrate to which a multicopter is fixed; at least one temperature sensor movable along a horizontal direction; and at least one sensor disposed on a sensor block movable up and down.

[0009] According to another suitable embodiment of the present invention, a plurality of parts include a motor, an ESC (Electric Speed ​​Controller), a propeller, a frame, and a battery, and sensors for measuring the status of the plurality of parts are synchronized.

[0010] According to another suitable embodiment of the present invention, for a diagnostic process by applying an artificial intelligence learning algorithm, a component diagnostic indicator is set and learned by the artificial intelligence learning algorithm. Effects of the invention

[0011] The artificial intelligence algorithm-based multicopter condition diagnosis method according to the present invention enables accurate condition diagnosis of a multicopter by analyzing data acquired from an inspection device based on an artificial intelligence algorithm. The multicopter condition diagnosis method according to the present invention enables effective condition diagnosis by measuring the condition of major components forming the multicopter in an operating state. Furthermore, the condition diagnosis method according to the present invention enables effective condition diagnosis based on an artificial intelligence algorithm from various types of parameters detected by various types of measurement sensors. The condition diagnosis method according to the present invention enables effective processing of data detected by measurement sensors suitable for measuring the operating state of each component forming the multicopter. The condition diagnosis method according to the present invention improves data accuracy through the synchronization of measurement sensors. Additionally, the condition diagnosis method according to the present invention enables the detection of abnormal conditions in real time from measured data so that necessary measures can be taken. The condition diagnosis method according to the present invention can be applied to the analysis of detection data collected from various measurement sensors using an artificial intelligence algorithm regarding the operating state of various components forming the multicopter, and the present invention is not limited thereto. Brief explanation of the drawing

[0012] FIG. 1 illustrates an example of a multicopter condition diagnosis method based on an artificial intelligence algorithm according to the present invention. Figure 2 illustrates an example of the process of performing a state diagnosis in the state diagnosis method according to the present invention. FIG. 3 illustrates an example of an inspection device for a condition diagnosis method according to the present invention. FIG. 4 illustrates an example of a type of sensor and a processing process of data measured by a sensor for a condition diagnosis method according to the present invention. FIG. 5 illustrates an example of the process in which a condition diagnosis is performed for each component in the condition diagnosis method according to the present invention. Specific details for implementing the invention

[0013] The present invention is described in detail below with reference to embodiments shown in the attached drawings, but the embodiments are for a clear understanding of the invention and the invention is not limited thereto. In the description below, components having the same reference numeral in different drawings have similar functions and are not described repeatedly unless necessary for understanding the invention, and known components are described briefly or omitted, but should not be understood as being excluded from the embodiments of the present invention.

[0014] FIG. 1 illustrates an example of a multicopter condition diagnosis method based on an artificial intelligence algorithm according to the present invention.

[0015] Referring to FIG. 1, the artificial intelligence algorithm-based multicopter state diagnosis method comprises the steps of: detecting the operating status of a plurality of parts of a multicopter by an inspection device (P11); obtaining and processing the detection data obtained by the inspection device (P12); learning the data related to the detection by the inspection device by an artificial intelligence learning algorithm (P14); and analyzing the processed data by the artificial intelligence learning algorithm based on the learning result to diagnose the state of the multicopter (P14), wherein the operating status of each of the plurality of parts is detected by a sensor of at least one inspection device.

[0016] A multicopter can be various types of unmanned aerial vehicles that fly using two or more rotors or propellers, and includes propellers, an ESC (Electric Speed ​​Controller), a motor, a battery, or similar components for operation. The operating status of each component can be detected, and an inspection device equipped with multiple sensors for detecting the operating status may be prepared. The multicopter can be fixed to the inspection device, and the status of each component of the multicopter can be detected both when the multicopter is stationary and when the multicopter is in operation. The status of each component can be detected by at least one sensor, and the information detected by each sensor can be transmitted to a data acquisition / processing module for processing. The data can be processed into a form that can be processed by the data acquisition / processing module, and at the same time, the data can be processed into a form that can be learned by an artificial intelligence learning algorithm. The information detected by each sensor can be learned by the artificial intelligence learning algorithm, and at the same time, the artificial intelligence learning algorithm can learn various detection information regarding the operating status of various types of components that may appear through each sensor's detection. When information acquired by at least one sensor for each component is processed, a state diagnosis can be performed based on the processed data (P14). An artificial intelligence algorithm may be applied during the state diagnosis process, and the AI ​​algorithm may analyze the processed data based on the results of learning. Then, the state of each component of the multicopter can be diagnosed according to the analysis results by the AI ​​algorithm (P14). The diagnosis of a component may include determining whether the component has defects, the remaining lifespan of the component, whether the component needs to be replaced, and the condition of the component, such as a deteriorated state or a similar condition. Appropriate measures may be taken based on the diagnosis results of the component.Component diagnosis may include various items for determining the condition of the component, and the present invention is not limited thereto.

[0017] Figure 2 illustrates an example of the process of performing a state diagnosis in the state diagnosis method according to the present invention.

[0018] Referring to FIG. 2, a sensor module (21) may be positioned to detect the operating status of a component of a multicopter, and the sensor module (21) may include, for example, a vibration sensor, a temperature sensor, an ultrasonic sensor, a noise sensor, a laser scanner, a thermal imaging camera, an ammeter, or a voltmeter, but is not limited thereto. Detection data regarding a component of a multicopter detected by the sensor module (21) may be acquired by a data acquisition module (22). The sensor module (21) may transmit the detection data to the data acquisition module (22) via telemetry communication (T), but is not limited thereto. The data acquired by the data acquisition module (22) may be learned by a data learning module (23), and the data learning module (23) may include various types of artificial intelligence learning algorithms. The data learning module (23) may learn various forms of detection data that can be detected from the sensor module (21), and the learning result may be transmitted to an artificial intelligence learning algorithm (25). The acquired detection data can be transmitted to the parameter extraction / setting module (24), and parameter values ​​can be extracted by the parameter extraction / setting module (24) and parameter values ​​according to each state can be set. For example, values ​​for vibration, noise, temperature, shape, ultrasound, voltage, current, or similar parameters acquired by at least one sensor for each component can be extracted. Additionally, parameter values ​​according to the state of each component can be set based on the values ​​extracted in this way or data transmitted from the data learning module (23). For example, a range of temperature values ​​that can be detected according to the usage period of the component under a certain operating state can be set. Additionally, parameters to be detected for the diagnosis of each component can be selected. Furthermore, an artificial intelligence learning algorithm (25) can be applied for setting such parameter values ​​or selecting parameters.Extracted parameter values ​​or selected parameters can be transmitted to the parameter analysis module (26) for analysis. The parameter analysis module (26) can analyze the appropriateness of the selected parameters for each component and can analyze the parameters for the component extracted from the current detection data. Parameter analysis means analyzing whether the component is in an appropriate operating state based on the measured or detected parameter values. The parameter analysis module (26) can analyze changes in parameters over time based on the operating state and determine whether the component is normal based on the analysis results. The analysis results by the parameter analysis module (26) can be transmitted to the component diagnosis module (28). The component diagnosis module (28) can diagnose the condition of each component based on the analysis results by the parameter analysis module (26). For example, the component diagnosis module (28) can diagnose whether the component is normal or defective, diagnose whether the component is operating according to its normal service life, and diagnose the replacement time of the component based on the current operating state. An artificial intelligence learning algorithm (25) may be applied during the process of analyzing parameters by the parameter analysis module (26) or during the process of diagnosing parts by the part diagnosis module (28), and a database for the parts of the multicopter may be created by the parameter database creation module (27) according to the result of applying the artificial intelligence learning algorithm (25). The parameter database may include the unique number of the multicopter and the unique number of the parts, and may store the diagnosis results regarding the state of each part over time. The parameter database created by the parameter database creation module (27) may become historical data of the multicopter parts and may serve as training data for the artificial intelligence learning algorithm. The result of the part diagnosis by the part diagnosis module (28) may be transmitted to the diagnosis database creation module (29).The diagnostic database generation module (29) can generate and store a diagnostic database, and the diagnostic database can be transmitted to the manager of the multicopter, and the manager can take necessary measures for each component based on the diagnostic database. The diagnostic database can be utilized in various ways for the operation of the multicopter, and the present invention is not limited thereto.

[0019] FIG. 3 illustrates an example of an inspection device for a condition diagnosis method according to the present invention.

[0020] Referring to FIG. 3, the multicopter inspection device comprises: a base (B) having an upper surface with a flat shape; a fixing plate (FP) that is placed on the base (B) and to which the multicopter (M) is fixed; a vertical pillar (VP) that extends vertically with its lower end fixed to the base (B); a horizontal member (HP) that extends along a horizontal direction from the upper end of the vertical pillar (VP); an operating module (CP) that controls the operation of the inspection device; and a data acquisition / processing module (DQ / DP) that acquires and processes data detected by each sensor (33, 34, 35, 37) placed on the inspection device. The base (B) may have an overall rectangular plate shape, and a caster capable of being fixed at a fixed position may be attached to the bottom so that the base (B) can be movable and stably fixed at a fixed position. The upper surface of the base (B) may have a flat shape, and various configurations for inspecting the multicopter (M) may be placed on the upper surface of the base (B). A fixed block (FB) having a polyhedral shape such as a cuboid, a cylinder shape, or a similar shape may be placed on the base (B), and a fixed plate (FP) may be attached to the upper part of the fixed block (FB). The fixed plate (FP) may have a circular shape overall, and a multicopter (M) may be fixed to the upper surface of the fixed plate (FP). For example, the lower part of the multicopter (M) may include at least one pair of horizontal members, and the pair of horizontal members may be fixed to the upper surface of the fixed plate (FP) by at least one locking block (31a, 31b). The multicopter (M) may be fixed to the upper surface of the fixed plate (FP) in various ways. The fixed plate (FP) may have a rotatable structure, and the multicopter (M) may be rotated by the rotation of the fixed plate (FP). A vertical filler (VP) may be extended from the upper surface of the base (B), and a horizontal member (HM) may be extended horizontally from the upper end of the vertical filler (VP).Additionally, a scan guide module (36) may be coupled to the horizontal member (HM). The scan guide module (36) may have a structure that extends in a direction perpendicular to the extension direction of the horizontal extension part, and the scan guide module (36) may be coupled so as to be movable along the horizontal member (HM). A movable bracket (361) may be coupled to the scan guide module (36), and a laser scanner (37) may be coupled to the movable bracket (361). The movable bracket (361) can move along the scan guide module (36), and accordingly, the laser scanner (37) can move along the scan guide module (36). The multicopter (M) can be scanned by the laser scanner (37), and the laser scanner (37) can move along the first axis direction while moving along the scan guide module (36). Furthermore, as the scan guide module (36) moves along the horizontal member (HM), the laser scanner (37) can move along the second axis direction perpendicular to the first axis. Accordingly, the entire plane can be scanned by the laser scanner (37), and the entire multicopter (M) can be scanned by the laser scanner (37) to obtain, for example, a 3D model of the multicopter (M). And the external condition of the multicopter (M) can be inspected from the 3D model obtained by the laser scanner (37). A vibration sensor module (35) can be installed on the upper surface of the base (B), and the vibration sensor module (35) includes a sensor housing in the shape of a cuboid and a vibration sensor placed inside the sensor housing. The multicopter (M) fixed to the fixed plate (FP) can be operated, and vibration can be generated while the multicopter (M) is operated. The generated vibration can be transmitted along the fixed plate (FP) and the fixed block (FP) to the upper surface of the base (B), and then transmitted to the vibration sensor of the vibration sensor module (35) to be measured.The fixed plate (FP), fixed block (FB), and base (B) can be made of various materials or structures capable of transmitting vibrations of the multicopter (M) to the vibration sensor module (35). A temperature sensor (34) can be installed on the upper surface of the base (B), and the temperature sensor (34) can be installed so as to be movable along the upper surface of the base (B). A linear guide (341) can be installed on the upper surface of the base, and the linear guide (341) can extend linearly from one side of the fixed block (FB) toward one edge of the base (B). One end of a movable coupler (342) can be movably coupled to the linear guide (341), and the movable coupler (342) can be in the shape of a square plate that extends vertically overall. An adjustable bracket that can be directionally adjusted can be coupled to the upper end of the movable coupler (342), and the temperature sensor (34) can be coupled to the adjustable bracket. The temperature sensor (34) can be an infrared sensor, and the temperature of the frame or propeller can be measured by the temperature sensor (34). An operating motor can be installed at one end of the linear guide (341), and the moving coupler (342) can be moved along the linear guide (341) by the operation of the operating motor. Accordingly, as the temperature sensor (34) moves linearly along the linear guide (341), the temperature of the multicopter (M) in operation can be measured in a non-contact manner. A sensor block (33) can be installed on the vertical pillar (VP), and the sensor block (33) can be installed so as to be able to move up and down along the vertical extension. Specifically, a square plate-shaped vertical guide (32) can be installed on the vertical pillar (VP), and an induction coupler (321) can be coupled to the vertical guide (32). The induction coupler (321) can be moved up and down along the vertical guide (32), and a sensor block can be coupled to the induction coupler (321). At least one sensor (33a, 33b) can be coupled to the sensor block.At least one sensor (33a, 33b) may include, for example, a thermal imaging camera, an ultrasonic sensor, a noise sensor such as a microphone, or an RPM sensor. Data acquired or measured by the laser scanner (37) or the sensors (34, 35, 33a, 33b) may be collected and processed by a data acquisition / processing module (D / P). The data acquisition / processing module (DQ / DP) may be installed on one side of the upper surface of the base (B), and the data acquisition / processing module (DQ / DP) may include various software or hardware for processing scan data or detection data. For example, the data acquisition / processing module (DQ / DP) may include, for example, a central processing unit and storage means, and may include a processor for processing data. An operation module (CP) for operating the inspection device may be installed on the upper surface of the base (B), and the operation module (CP) may have a PLC (Programming Logic Controller) structure. The inspection process of the multicopter (M) in the inspection device can be carried out through a pre-programmed automated process.

[0021] FIG. 4 illustrates an example of a type of sensor and a processing process of data measured by a sensor for a condition diagnosis method according to the present invention.

[0022] Referring to FIG. 4, a plurality of components include a motor, an ESC (Electric Speed ​​Controller), a propeller, a frame, and a battery, and sensors for measuring the status of the plurality of components are synchronized. A sensor module (21) placed in an inspection device may include, but is not limited to, a vibration sensor (41a), an ultrasonic sensor (41b), a noise sensor (41c), a voltage / amplifier (41d), and a laser / thermal imaging sensor (41e). Additionally, it may include an infrared temperature sensor, an RPM sensor, or similar sensors. Various vibrations generated during the operation of the multicopter can be measured by the vibration sensor (41a), and, for example, motor status data (42a) or propeller status data (42c) can be obtained. Motor status data (42a) or frame status data (42d) can be obtained by the ultrasonic sensor (41b). Propeller status data (42c) or frame status data (42d) can be obtained by a noise sensor (41c), and ESC status data (42b) or battery status data (42e) can be obtained by a voltmeter or ammeter (41d). ESC status data (42b), propeller status data (42c), frame status data (42d), or battery status data (42e) can be obtained by a laser scanner or thermal imaging scanner (41e). Additionally, propeller temperature data or battery temperature data in an operating state can be obtained by an infrared temperature sensor, and the propeller RPM can be measured by an RPM sensor. Data regarding various operating states of multicopter components can be obtained by various sensors, and the present invention is not limited by this. When operating state data for the multicopter is obtained by the sensor module (21), the data can be processed by the data processing module (43).The data processing module (43) may have the function of verifying the validity of data obtained from the sensor module (21), aligning the data based on the measurement time, and combining the operating conditions and the detection time. Additionally, the data processing module (43) may convert the data transmitted from the sensor module (21) into a signal form that is electrically processable or computer-readable, and create a data form that can be learned by the artificial intelligence learning module. The data processed by the data processing module (43) may be transmitted to the data synchronization module (44) to be synchronized. The data synchronization module (44) may synchronize the data detected by each sensor (41a to 41e) based on the time, and the synchronized detection data may be aligned according to the passage of time. The synchronized detection data may be transmitted to the state diagnosis indicator extraction module (45), and indicator data for state diagnosis may be extracted by the state diagnosis indicator extraction module (45). The indicator data may be data corresponding to the normal operating conditions in each operating condition, and the indicator data may be generated in advance. For example, an artificial intelligence learning algorithm (47) may be applied to predetermine indicator data or indicator values ​​for each component under each operating condition. Alternatively, indicator values ​​or indicator data may be determined based on the average value of the detection data. In this way, when indicator data or indicator values ​​are extracted by the state diagnosis indicator extraction module (45), the detection data is analyzed by the data analysis module (46) based on this, and each component can be diagnosed. The artificial intelligence learning algorithm (47) may be applied during the data processing, data synchronization, indicator extraction, or data analysis process, thereby improving the diagnostic efficiency of the component. The artificial intelligence learning algorithm (47) may be various algorithms, and the present invention is not limited by this.

[0023] FIG. 5 illustrates an example of the process in which a condition diagnosis is performed for each component in the condition diagnosis method according to the present invention.

[0024] Referring to FIG. 5, a method for diagnosing the state of a multicopter component comprises the following steps: a step of distinguishing multicopter components to be diagnosed (P51); a step of extracting component parameters for diagnosing each distinguished component (P52); a step of learning the extracted parameters and the operating state of each component (P53); a step of creating a parameter database based on the learning results (P54); a step of detecting determined parameters for each component by a sensor module (P55); a step of processing the detected parameters or detection data and synchronizing multiple parameters (P56); a step of setting diagnostic indicators for diagnosing each component based on the operating state (P57); a step of learning the set diagnostic indicators by an artificial intelligence learning algorithm (P58); a step of analyzing the detection data based on the diagnostic indicators to diagnose the component state (P59); and a step of creating a component diagnostic database (P591). The parameter database may be created based on various detection data for the component (P54), and the parameter database may be learned by an artificial intelligence learning algorithm. Furthermore, based on the learning results of the parameter database, diagnostic indicators for each component can be set by an artificial intelligence learning algorithm. Additionally, the artificial intelligence learning algorithm can be applied to the condition diagnosis process of each component to improve diagnostic accuracy and efficiency (P59). Once the component diagnosis database is created (P591), necessary measures for each component can be taken based thereon. Measures based on the diagnosis can be taken in various ways, and the present invention is not limited thereto.

[0025] Although the present invention has been described in detail above with reference to the presented embodiments, those skilled in the art may make various modifications and variations without departing from the technical spirit of the invention by referring to the presented embodiments. The present invention is not limited by such modifications and variations, but is limited only by the claims appended below. Explanation of the symbols

[0026] 21: Sensor module 22: Data acquisition module 23: Data Training Module 24: Parameter Extraction / Setting Module 25: Artificial Intelligence Learning Algorithm 26: Parameter Analysis Module 27: Parameter DB Creation Module 28: Part Diagnosis Module 29: Diagnostic DB Generation Module

Claims

Claim 1 The method comprises: a step of detecting the operating status of multiple parts of a multicopter by an inspection device; a step of acquiring and processing detection data obtained by the inspection device; a step of learning data related to the detection by the inspection device by an artificial intelligence learning algorithm; and a step of analyzing the processed data by the artificial intelligence learning algorithm based on the learning results to diagnose the state of the multicopter, wherein the operating status of each of the multiple parts is detected by multiple sensors placed in the inspection device, and the multiple parts include a motor, an ESC (Electric Speed ​​Controller), a propeller, a frame, and a battery, and the multiple sensors for measuring the state of the multiple parts are synchronized, and the inspection device comprises a base (B) having a flat upper surface; a fixing plate (FP) placed on the base (B) to which the multicopter (M) is fixed; a vertical pillar (VP) extending vertically with its lower end fixed to the base (B); a horizontal member (HP) extending along a horizontal direction from the upper end of the vertical pillar (VP); and an operating module (CP) that controls the operation of the inspection device. A method for diagnosing the condition of a multicopter based on an artificial intelligence algorithm, comprising: a data acquisition / processing module (DQ / DP) that acquires and processes data detected by a plurality of sensors, wherein the plurality of sensors include a vibration sensor module (35) installed on the upper surface of the base (B) to measure vibrations transmitted to the upper surface of the base (B) that are generated while the multicopter (M) fixed to the fixed plate (FP) is operating; and a temperature sensor (34) installed to be movable along a linear guide (341) installed on the upper surface of the base (B) to measure the temperature of the frame or propeller of the multicopter. Claim 2 A multicopter state diagnosis method based on an artificial intelligence algorithm according to claim 1, characterized in that parameters for detecting the operating state of a component are extracted and set, and the setting of the parameters is performed by an artificial intelligence learning algorithm. Claim 3 delete Claim 4 delete Claim 5 A multicopter state diagnosis method based on an artificial intelligence algorithm according to claim 1, characterized in that a component diagnosis indicator is set and learned by the artificial intelligence learning algorithm for a diagnosis process by applying an artificial intelligence learning algorithm.

Images