Flight performance testing methods and systems based on unmanned aerial vehicles (UAVs)

Abstract

The embodiment of the application provides a flight performance test method and system based on a UAV, template flight performance monitoring data sequences are acquired; for each template flight performance monitoring data, the current template flight performance monitoring data is taken as a loading parameter of an initialization abnormality analysis network of an initialization weight parameter, one or more candidate abnormality types corresponding to the current template flight performance monitoring data are obtained; based on one or more prior abnormality types in the current template flight performance monitoring data and the corresponding one or more candidate abnormality types, example abnormality weight parameter information in the initialization abnormality analysis network of the initialization weight parameter is optimized; evaluation index convergence in the initialization abnormality analysis network of the initialization weight parameter is taken as a knowledge learning direction, a target abnormality analysis network is obtained, and therefore the accuracy of subsequent flight performance abnormality monitoring is improved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and more specifically, to a method and system for testing the flight performance of unmanned aerial vehicles (UAVs). Background Technology

[0002] With the continuous development of drone technology, drones are being used more and more widely in various fields, making drone testing increasingly important. Drone testing involves conducting various tests on drones, including flight testing, payload testing, battery testing, remote control testing, sensor testing, safety testing, and environmental testing, to ensure that their performance and safety meet requirements. In an environment surrounded by a great deal of complex information, it is necessary to accurately and quickly identify anomalies in flight performance to improve the efficiency of subsequent drone testing. For example, artificial intelligence technology can be used to improve the accuracy of subsequent monitoring of flight performance anomalies in drone flight performance testing methods. Summary of the Invention

[0003] In view of this, the purpose of this invention is to provide a method and system for flight performance testing based on unmanned aerial vehicles (UAVs), which acquires template flight performance monitoring data sequences; for each template flight performance monitoring data, the current template flight performance monitoring data is used as the loading parameter of the initialization anomaly analysis network for initializing weight parameters to obtain one or more candidate anomaly types corresponding to the current template flight performance monitoring data; based on one or more prior anomaly types and corresponding one or more candidate anomaly types in the current template flight performance monitoring data, the example anomaly weight parameter information in the initialization anomaly analysis network for initializing weight parameters is optimized; the convergence of evaluation indicators in the initialization anomaly analysis network for initializing weight parameters is used as the knowledge learning direction to obtain the target anomaly analysis network, thereby improving the accuracy of subsequent flight performance anomaly monitoring.

[0004] According to one aspect of the present invention, a method and system for testing the flight performance of an unmanned aerial vehicle (UAV) are provided, the method comprising:

[0005] Obtain a template flight performance monitoring data sequence; wherein, the template flight performance monitoring data sequence includes multiple flight performance monitoring tags, and under different flight performance monitoring tags, there are multiple template flight performance monitoring data, and the template flight performance monitoring data includes sample flight performance data and prior anomaly types corresponding to the target performance anomaly location nodes;

[0006] For each template flight performance monitoring data, the current template flight performance monitoring data is used as the loading parameter of the initialization anomaly analysis network to generate one or more candidate anomaly types corresponding to the current template flight performance monitoring data.

[0007] For each template flight performance monitoring data, based on one or more prior anomaly types and one or more corresponding candidate anomaly types in the current template flight performance monitoring data, the example anomaly weight parameter information in the initial anomaly analysis network of the initial weight parameter is optimized;

[0008] The predicted reliability index in the initial anomaly analysis network with the initial weight parameters is used as the knowledge learning direction to generate a target anomaly analysis network; wherein, the target anomaly analysis network is used to detect the loaded abnormal flight performance data and generate anomaly types corresponding to the abnormal flight performance data.

[0009] In an alternative implementation, the acquisition of the template flight performance monitoring data sequence includes:

[0010] Acquire abnormal flight performance data, including the target performance anomaly location node;

[0011] Identify one or more performance fields corresponding to each abnormal flight performance data, and generate multiple abnormal flight performance data corresponding to each performance field;

[0012] Determine the prior anomaly type corresponding to each abnormal flight performance data; based on the abnormal flight performance data corresponding to each performance field and the corresponding prior anomaly type, determine each flight performance monitoring label in the template flight performance monitoring data sequence.

[0013] In an alternative implementation, determining one or more performance fields corresponding to each abnormal flight performance data point, and generating multiple abnormal flight performance data points corresponding to each performance field, includes:

[0014] Encode each abnormal flight performance data and determine the encoded abnormal flight performance data;

[0015] Based on preset feature cleaning rules, feature cleaning is performed on the encoded abnormal flight performance data to generate target abnormal flight performance data.

[0016] Identify one or more performance fields corresponding to the abnormal flight performance data of each target.

[0017] In an alternative implementation, the method further includes:

[0018] Obtain a verification flight performance monitoring data sequence; wherein, the verification flight performance monitoring data sequence includes multiple flight performance monitoring tags, and each flight performance monitoring tag includes multiple verification flight performance monitoring data, and the flight performance monitoring tags are the same as the flight performance monitoring tags;

[0019] Each set of flight performance monitoring data is recorded into the trained target anomaly analysis network to generate the actual anomaly type corresponding to each set of flight performance monitoring data.

[0020] Based on the actual anomaly types and corresponding prior anomaly types of each flight performance monitoring data, the training loss function value under the same flight performance monitoring label is determined.

[0021] If there is a target flight performance monitoring label whose training loss function value is greater than the set loss function value, then the template flight performance monitoring data corresponding to the target flight performance monitoring label is obtained, and the target anomaly analysis network is iteratively optimized until the training loss function value of each flight performance monitoring label is not greater than the set loss function value.

[0022] In an alternative implementation, the step of acquiring template flight performance monitoring data corresponding to the target flight performance monitoring label, and continuing to iteratively optimize the target anomaly analysis network until the training loss function value of each flight performance monitoring label is not greater than a set loss function value, includes:

[0023] Acquire candidate template flight performance monitoring data corresponding to the target flight performance monitoring label, and retrain the target anomaly analysis network based on the candidate template flight performance monitoring data and the template flight performance monitoring data in the template flight performance monitoring data sequence, until the training loss function value of each flight performance monitoring label is determined to be no greater than the set loss function value based on the verified flight performance monitoring data sequence.

[0024] In an alternative implementation, the method further includes:

[0025] Acquire target flight performance monitoring data; wherein, the target flight performance monitoring data includes target performance anomaly location nodes;

[0026] The target flight performance monitoring data is recorded into the target anomaly analysis network to generate one or more anomaly types corresponding to the target flight performance monitoring data;

[0027] Based on the one or more anomaly types, determine the abnormal test path of the target performance anomaly location node in the target flight performance monitoring data.

[0028] According to another aspect of the present invention, a method and system for testing the flight performance of an unmanned aerial vehicle (UAV) are provided, the system comprising:

[0029] An acquisition unit is used to acquire a template flight performance monitoring data sequence; wherein, the template flight performance monitoring data sequence includes multiple flight performance monitoring tags, and under different flight performance monitoring tags, there are multiple template flight performance monitoring data, and the template flight performance monitoring data includes sample flight performance data and prior anomaly types corresponding to the target performance anomaly location nodes;

[0030] The generation unit is used to generate one or more candidate anomaly types corresponding to the current template flight performance monitoring data by using the current template flight performance monitoring data as the loading parameter of the initialization anomaly analysis network for initialization weight parameters.

[0031] The optimization unit is used to optimize the example anomaly weight parameter information in the initial anomaly analysis network of the initial weight parameters based on one or more prior anomaly types and one or more corresponding candidate anomaly types in the current template flight performance monitoring data.

[0032] The training unit is used to converge the evaluation indicators in the initial anomaly analysis network with the initialized weight parameters as the knowledge learning direction to generate a target anomaly analysis network; wherein, the target anomaly analysis network is used to detect the loaded abnormal flight performance data and generate anomaly types corresponding to the abnormal flight performance data.

[0033] According to another aspect of the present invention, an electronic device is provided, comprising: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; the memory is used to store a computer program; and the processor is used to execute the computer program to implement the steps of the flight performance testing method based on unmanned aerial vehicles described above.

[0034] According to another aspect of the present invention, a readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, can perform the steps of the above-described method for testing the flight performance of a drone.

[0035] To make the above-mentioned objects, features and advantages of the embodiments of the present invention more apparent and understandable, a detailed description will be given below in conjunction with the embodiments and the accompanying drawings. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 A schematic diagram of the components of a server provided in an embodiment of the present invention is shown;

[0038] Figure 2 A flowchart illustrating the flight performance testing method and system based on unmanned aerial vehicles (UAVs) provided in an embodiment of the present invention is shown.

[0039] Figure 3 The diagram shows a functional block diagram of a flight performance testing system based on unmanned aerial vehicles (UAVs) provided in an embodiment of the present invention. Detailed Implementation

[0040] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] The terms “first,” “second,” “third,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0042] Figure 1An exemplary component diagram of server 100 is shown. Server 100 may include one or more processors 104, such as one or more central processing units (CPUs), each of which may implement one or more hardware threads. Server 100 may also include any storage medium 106 for storing any kind of information such as code, settings, data, etc. Without limitation, for example, storage medium 106 may include any type of RAM, any type of ROM, flash memory device, hard disk, optical disk, etc. More generally, any storage medium can use any technology to store information. Further, any storage medium may provide volatile or non-volatile retention of information. Further, any storage medium may represent a fixed or removable component of server 100. In one case, server 100 may perform any operation of the associated instructions when processor 104 executes associated instructions stored in any storage medium or combination of storage media. Server 100 also includes one or more drive units 108 for interacting with any storage medium, such as hard disk drive units, optical disk drive units, etc.

[0043] Server 100 also includes input / output 110 (I / O) for receiving various inputs (via input unit 112) and providing various outputs (via output unit 114). A specific output mechanism may include a presentation device 116 and an associated graphical user interface (GUI) 118. Server 100 may also include one or more network interfaces 120 for exchanging data with other devices via one or more communication units 122. One or more communication buses 124 couple the components described above together.

[0044] The communication unit 122 can be implemented in any manner, such as via a local area network, a wide area network (e.g., the Internet), a point-to-point connection, or any combination thereof. The communication unit 122 may include any combination of hardwired links, wireless links, routers, gateway functions, name server 100, etc., governed by any protocol or combination of protocols.

[0045] Figure 2 The diagram illustrates a flowchart of a UAV-based flight performance testing method and system provided in an embodiment of the present invention. This UAV-based flight performance testing method and system can be derived from... Figure 1 The server 100 shown in the figure executes the following detailed steps of the UAV-based flight performance testing method and system.

[0046] Step S110: Obtain template flight performance monitoring data sequence; wherein, the template flight performance monitoring data sequence includes multiple flight performance monitoring tags, and each flight performance monitoring tag includes multiple template flight performance monitoring data, and the template flight performance monitoring data includes sample flight performance data and prior anomaly types corresponding to the target performance anomaly location node;

[0047] Step S120: For each template flight performance monitoring data, the current template flight performance monitoring data is used as the loading parameter of the initialization anomaly analysis network to initialize the weight parameters, and one or more candidate anomaly types corresponding to the current template flight performance monitoring data are generated.

[0048] Step S130: For each template flight performance monitoring data, based on one or more prior anomaly types and one or more corresponding candidate anomaly types in the current template flight performance monitoring data, optimize the example anomaly weight parameter information in the initial anomaly analysis network of the initial weight parameters.

[0049] Step S140: The predicted reliability index in the initial anomaly analysis network of the initialized weight parameters is used as the knowledge learning direction to generate a target anomaly analysis network; wherein, the target anomaly analysis network is used to detect the loaded abnormal flight performance data and generate anomaly types corresponding to the abnormal flight performance data.

[0050] Based on the above steps, this embodiment obtains a sequence of template flight performance monitoring data; for each template flight performance monitoring data, the current template flight performance monitoring data is used as the loading parameter of the initialization anomaly analysis network for initializing weight parameters, resulting in one or more candidate anomaly types corresponding to the current template flight performance monitoring data; based on one or more prior anomaly types and corresponding one or more candidate anomaly types in the current template flight performance monitoring data, the example anomaly weight parameter information in the initialization anomaly analysis network for initializing weight parameters is optimized; the convergence of evaluation indicators in the initialization anomaly analysis network for initializing weight parameters is used as the knowledge learning direction to obtain the target anomaly analysis network, thereby improving the accuracy of subsequent flight performance anomaly monitoring.

[0051] In an alternative implementation, the acquisition of the template flight performance monitoring data sequence includes:

[0052] Acquire abnormal flight performance data, including the target performance anomaly location node;

[0053] Identify one or more performance fields corresponding to each abnormal flight performance data, and generate multiple abnormal flight performance data corresponding to each performance field;

[0054] Determine the prior anomaly type corresponding to each abnormal flight performance data; based on the abnormal flight performance data corresponding to each performance field and the corresponding prior anomaly type, determine each flight performance monitoring label in the template flight performance monitoring data sequence.

[0055] In an alternative implementation, determining one or more performance fields corresponding to each abnormal flight performance data point, and generating multiple abnormal flight performance data points corresponding to each performance field, includes:

[0056] Encode each abnormal flight performance data and determine the encoded abnormal flight performance data;

[0057] Based on preset feature cleaning rules, feature cleaning is performed on the encoded abnormal flight performance data to generate target abnormal flight performance data.

[0058] Identify one or more performance fields corresponding to the abnormal flight performance data of each target.

[0059] In an alternative implementation, the method further includes:

[0060] Obtain a verification flight performance monitoring data sequence; wherein, the verification flight performance monitoring data sequence includes multiple flight performance monitoring tags, and each flight performance monitoring tag includes multiple verification flight performance monitoring data, and the flight performance monitoring tags are the same as the flight performance monitoring tags;

[0061] Each set of flight performance monitoring data is recorded into the trained target anomaly analysis network to generate the actual anomaly type corresponding to each set of flight performance monitoring data.

[0062] Based on the actual anomaly types and corresponding prior anomaly types of each flight performance monitoring data, the training loss function value under the same flight performance monitoring label is determined.

[0063] If there is a target flight performance monitoring label whose training loss function value is greater than the set loss function value, then the template flight performance monitoring data corresponding to the target flight performance monitoring label is obtained, and the target anomaly analysis network is iteratively optimized until the training loss function value of each flight performance monitoring label is not greater than the set loss function value.

[0064] In an alternative implementation, the step of acquiring template flight performance monitoring data corresponding to the target flight performance monitoring label, and continuing to iteratively optimize the target anomaly analysis network until the training loss function value of each flight performance monitoring label is not greater than a set loss function value, includes:

[0065] Acquire candidate template flight performance monitoring data corresponding to the target flight performance monitoring label, and retrain the target anomaly analysis network based on the candidate template flight performance monitoring data and the template flight performance monitoring data in the template flight performance monitoring data sequence, until the training loss function value of each flight performance monitoring label is determined to be no greater than the set loss function value based on the verified flight performance monitoring data sequence.

[0066] In an alternative implementation, the method further includes:

[0067] Acquire target flight performance monitoring data; wherein, the target flight performance monitoring data includes target performance anomaly location nodes;

[0068] The target flight performance monitoring data is recorded into the target anomaly analysis network to generate one or more anomaly types corresponding to the target flight performance monitoring data;

[0069] Based on the one or more anomaly types, determine the abnormal test path of the target performance anomaly location node in the target flight performance monitoring data.

[0070] Figure 3 A functional block diagram of a UAV-based flight performance testing system 200 according to an embodiment of the present invention is shown. The functions implemented by the UAV-based flight performance testing system 200 correspond to the steps performed by the above-described method. The UAV-based flight performance testing system 200 can be understood as the aforementioned server 100, or the processor of server 100, or it can be understood as a component independent of server 100 or the processor, but implementing the functions of the present invention under the control of server 100, such as... Figure 3 As shown below, the functions of each functional module of the UAV-based flight performance testing system 200 will be described in detail.

[0071] The acquisition unit 210 is used to acquire a template flight performance monitoring data sequence; wherein, the template flight performance monitoring data sequence includes multiple flight performance monitoring tags, and each flight performance monitoring tag includes multiple template flight performance monitoring data, and the template flight performance monitoring data includes sample flight performance data and prior anomaly types corresponding to the target performance anomaly location node;

[0072] The generation unit 220 is used to generate one or more candidate anomaly types corresponding to the current template flight performance monitoring data by using the current template flight performance monitoring data as the loading parameter of the initialization anomaly analysis network for initialization weight parameters.

[0073] Optimization unit 230 is used to optimize the example anomaly weight parameter information in the initial anomaly analysis network of the initial weight parameters based on one or more prior anomaly types and one or more corresponding candidate anomaly types in the current template flight performance monitoring data.

[0074] Training unit 240 is used to converge the evaluation index in the initial anomaly analysis network with the initial weight parameters as the knowledge learning direction to generate a target anomaly analysis network; wherein, the target anomaly analysis network is used to detect the loaded abnormal flight performance data and generate anomaly types corresponding to the abnormal flight performance data.

[0075] In an alternative implementation, the acquisition unit 210 is further configured to:

[0076] Acquire abnormal flight performance data, including the target performance anomaly location node;

[0077] Identify one or more performance fields corresponding to each abnormal flight performance data, and generate multiple abnormal flight performance data corresponding to each performance field;

[0078] Determine the prior anomaly type corresponding to each abnormal flight performance data; based on the abnormal flight performance data corresponding to each performance field and the corresponding prior anomaly type, determine each flight performance monitoring label in the template flight performance monitoring data sequence.

[0079] In an alternative implementation, the acquisition unit 210 is further configured to:

[0080] Encode each abnormal flight performance data and determine the encoded abnormal flight performance data;

[0081] Based on preset feature cleaning rules, feature cleaning is performed on the encoded abnormal flight performance data to generate target abnormal flight performance data.

[0082] Identify one or more performance fields corresponding to the abnormal flight performance data of each target.

[0083] In an alternative implementation, the acquisition unit 210 is further configured to:

[0084] Obtain a verification flight performance monitoring data sequence; wherein, the verification flight performance monitoring data sequence includes multiple flight performance monitoring tags, and each flight performance monitoring tag includes multiple verification flight performance monitoring data, and the flight performance monitoring tags are the same as the flight performance monitoring tags;

[0085] Each set of flight performance monitoring data is recorded into the trained target anomaly analysis network to generate the actual anomaly type corresponding to each set of flight performance monitoring data.

[0086] Based on the actual anomaly types and corresponding prior anomaly types of each flight performance monitoring data, the training loss function value under the same flight performance monitoring label is determined.

[0087] If there is a target flight performance monitoring label whose training loss function value is greater than the set loss function value, then the template flight performance monitoring data corresponding to the target flight performance monitoring label is obtained, and the target anomaly analysis network is iteratively optimized until the training loss function value of each flight performance monitoring label is not greater than the set loss function value.

[0088] In an alternative implementation, the acquisition unit 210 is further configured to:

[0089] Acquire candidate template flight performance monitoring data corresponding to the target flight performance monitoring label, and retrain the target anomaly analysis network based on the candidate template flight performance monitoring data and the template flight performance monitoring data in the template flight performance monitoring data sequence, until the training loss function value of each flight performance monitoring label is determined to be no greater than the set loss function value based on the verified flight performance monitoring data sequence.

[0090] In an alternative implementation, the acquisition unit 210 is further configured to:

[0091] Acquire target flight performance monitoring data; wherein, the target flight performance monitoring data includes target performance anomaly location nodes;

[0092] The target flight performance monitoring data is recorded into the target anomaly analysis network to generate one or more anomaly types corresponding to the target flight performance monitoring data;

[0093] Based on the one or more anomaly types, determine the abnormal test path of the target performance anomaly location node in the target flight performance monitoring data.

[0094] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0095] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of equivalents of the claims be included within the present invention.

Claims

1. A method for testing flight performance based on a UAV, characterized in that, The method comprises: obtaining a template flight performance monitoring data sequence; wherein the template flight performance monitoring data sequence comprises a plurality of flight performance monitoring tags, each flight performance monitoring tag comprises a plurality of template flight performance monitoring data, and each template flight performance monitoring data comprises sample flight performance data corresponding to a target performance anomaly positioning node and a prior abnormal type; for each template flight performance monitoring data, taking the current template flight performance monitoring data as a loading parameter of an initialized abnormality analysis network with an initialized weight parameter to generate one or more candidate abnormal types corresponding to the current template flight performance monitoring data; for each template flight performance monitoring data, optimizing the example abnormal weight parameter information in the initialized abnormality analysis network with the initialized weight parameter based on one or more prior abnormal types in the current template flight performance monitoring data and the corresponding one or more candidate abnormal types; taking a prediction credibility index in the initialized abnormality analysis network with the initialized weight parameter as a knowledge learning direction to generate a target abnormality analysis network; wherein the target abnormality analysis network is used to detect loaded abnormal flight performance data to generate an abnormal type corresponding to the abnormal flight performance data. 2.The UAV-based flight performance testing method of claim 1, wherein, The method comprises: obtaining abnormal flight performance data comprising a target performance anomaly positioning node; determining one or more performance fields corresponding to each abnormal flight performance data to generate a plurality of abnormal flight performance data corresponding to each performance field; determining a prior abnormal type corresponding to each abnormal flight performance data; based on the abnormal flight performance data corresponding to each performance field and the corresponding prior abnormal type, determining each flight performance monitoring tag in the template flight performance monitoring data sequence. 3.The UAV-based flight performance testing method of claim 2, wherein, The method comprises: encoding each abnormal flight performance data to determine encoded abnormal flight performance data; based on a preset feature cleaning rule, performing feature cleaning on the encoded abnormal flight performance data to generate target abnormal flight performance data; determining one or more performance fields corresponding to each target abnormal flight performance data. 4.The UAV-based flight performance testing method of claim 1, wherein, The method further comprises: obtaining a verification flight performance monitoring data sequence; wherein the verification flight performance monitoring data sequence comprises a plurality of flight performance monitoring tags, each flight performance monitoring tag comprises a plurality of verification flight performance monitoring data, and the flight performance monitoring tags are the same as the flight performance monitoring tags; recording each verification flight performance monitoring data to the trained target abnormality analysis network to generate an actual abnormal type corresponding to each verification flight performance monitoring data; based on the actual abnormal type of each verification flight performance monitoring data and the corresponding prior abnormal type, determining a training loss function value under the same flight performance monitoring tag; If there is a target flight performance monitoring label with a training loss function value greater than a set loss function value, template flight performance monitoring data corresponding to the target flight performance monitoring label is obtained, and iterative optimization of the target anomaly analysis network is continued until the training loss function value of each flight performance monitoring label is not greater than the set loss function value. 5.The UAV-based flight performance testing method of claim 4, wherein, The obtaining of the template flight performance monitoring data corresponding to the target flight performance monitoring label and the continuing of the iterative optimization of the target anomaly analysis network until the training loss function value of each flight performance monitoring label is not greater than the set loss function value includes: The candidate template flight performance monitoring data corresponding to the target flight performance monitoring label is obtained, and the target anomaly analysis network is retrained based on the candidate template flight performance monitoring data and the template flight performance monitoring data in the sequence of template flight performance monitoring data until the training loss function value of each flight performance monitoring label is determined based on the sequence of verification flight performance monitoring data to be not greater than the set loss function value. 6.The UAV-based flight performance testing method of claim 1, wherein, Further comprising: Obtaining target flight performance monitoring data; wherein the target flight performance monitoring data includes a target performance anomaly positioning node; Recording the target flight performance monitoring data into the target anomaly analysis network to generate one or more abnormal types corresponding to the target flight performance monitoring data; Based on the one or more abnormal types, determining an abnormal test path of the target performance anomaly positioning node in the target flight performance monitoring data. 7.A system of a UAV-based flight performance test method, the system implementing the UAV-based flight performance test method of any one of claims 1-6, wherein, Comprising: An obtaining unit is configured to obtain a sequence of template flight performance monitoring data; wherein the sequence of template flight performance monitoring data includes a plurality of flight performance monitoring labels, and each flight performance monitoring label includes a plurality of template flight performance monitoring data, and each template flight performance monitoring data includes sample flight performance data corresponding to a target performance anomaly positioning node and prior abnormal types; A generating unit is configured to, for each template flight performance monitoring data, generate one or more candidate abnormal types corresponding to the current template flight performance monitoring data by taking the current template flight performance monitoring data as the loading parameter of the initialized anomaly analysis network with the initialized weight parameter; An optimizing unit is configured to, for each template flight performance monitoring data, optimize the example abnormal weight parameter information in the initialized anomaly analysis network with the initialized weight parameter based on one or more prior abnormal types in the current template flight performance monitoring data and the corresponding one or more candidate abnormal types; A training unit is configured to generate a target anomaly analysis network by taking the convergence of the evaluation index in the initialized anomaly analysis network with the initialized weight parameter as a knowledge learning direction; wherein the target anomaly analysis network is configured to detect loaded abnormal flight performance data to generate abnormal types corresponding to the abnormal flight performance data.

8. An electronic device, comprising: The apparatus includes one or more processors; and a storage device storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method for testing flight performance of a UAV according to any one of claims 1-6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the method for testing flight performance of a UAV according to any one of claims 1-6.

Images