Task reliability evaluation method and device, computer device and storage medium

Abstract

The application relates to a task reliability evaluation method and device, computer equipment and a storage medium. The method comprises the following steps: adjusting the sensor output of a UAV according to a fault test signal; performing image transformation on a test image to obtain a target image according to the flight state of the UAV after the sensor output is adjusted; performing image recognition on the test image and the target image respectively to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; wherein the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV; and the reliability of the UAV in performing a target recognition task is evaluated according to the first recognition result and the second recognition result. The method can more accurately evaluate the reliability of the UAV in performing a target recognition task.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a method, apparatus, computer equipment, storage medium, and computer program product for mission reliability assessment. Background Technology

[0002] With the rapid development of artificial intelligence and unmanned technologies, drones have been widely used in aerial photography, agricultural management, geographic surveying, search and rescue, environmental monitoring, traffic management, communication relay, and military applications. Ensuring that drones can safely and reliably complete their mission objectives has become a key research direction for the further development of drones.

[0003] In related technologies, the reliability assessment of UAVs performing target recognition tasks usually takes the target recognition algorithm used by the UAV as the research target. However, the influencing factors considered in the assessment process are relatively limited, resulting in low assessment accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide a task reliability assessment method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can improve the accuracy of assessment in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a method for assessing mission reliability, including:

[0006] Adjust the drone's sensor outputs based on the fault test signals;

[0007] Based on the flight status of the UAV after completing sensor output adjustments, image transformation is performed on the test image to obtain the target image;

[0008] Image recognition is performed on the test image and the target image respectively to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV.

[0009] The reliability of the UAV in performing target identification tasks is evaluated based on the first and second identification results.

[0010] In one embodiment, the fault test signal includes a sensor identifier, a parameter identifier, and a parameter setting value; adjusting the UAV's sensor output based on the fault test signal includes:

[0011] Based on the sensor identifier, identify the target sensor and obtain the corresponding fault function for the target sensor;

[0012] Adjust the output of the target sensor based on the parameter identifier, parameter setting value, and fault function.

[0013] In one embodiment, the flight state includes flight position and attitude angle; based on the flight state after the UAV completes sensor output adjustments, the test image is transformed to obtain the target image, including:

[0014] Based on the flight position, obtain the displacement matrix, and based on the attitude angle, obtain the rotation matrix;

[0015] Based on the rotation and translation matrices, the test image is transformed to obtain the target image.

[0016] In one embodiment, the first identification result includes a first number of identified targets and a first identification accuracy rate, and the second identification result includes a second number of identified targets and a second identification accuracy rate; based on the first and second identification results, the reliability of the UAV performing the target identification task is evaluated, including:

[0017] The target loss rate is obtained based on the number of first target identified and the number of second target identified.

[0018] The difference in recognition accuracy is obtained based on the first recognition accuracy and the second recognition accuracy.

[0019] The reliability of UAVs in performing target recognition tasks is evaluated based on the difference between the target loss rate and the recognition accuracy.

[0020] In one embodiment, the reliability of the UAV performing target recognition tasks is evaluated based on the difference between the target loss rate and the recognition accuracy, including:

[0021] The target loss rate is normalized to obtain the first evaluation parameter, and the difference in recognition accuracy is normalized to obtain the second evaluation parameter.

[0022] The target evaluation parameter is obtained by weighted summation of the first and second evaluation parameters.

[0023] The reliability of UAVs in performing target recognition tasks is evaluated based on target evaluation parameters.

[0024] In one embodiment, the reliability of the UAV performing target recognition tasks is evaluated based on target evaluation parameters, including:

[0025] Based on the mapping relationship between flight status and fault level, obtain the target fault level corresponding to the flight status of the UAV after completing sensor output;

[0026] The reliability of UAVs in performing target identification tasks is evaluated based on the target fault level and target evaluation parameters.

[0027] Secondly, this application also provides a mission reliability assessment device, comprising:

[0028] The adjustment module is used to adjust the sensor output of the UAV based on the fault test signal;

[0029] The transformation module is used to transform the test image based on the flight state of the UAV after the sensor output adjustment, so as to obtain the target image.

[0030] The recognition module is used to perform image recognition on the test image and the target image respectively, and obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; wherein, the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV;

[0031] The evaluation module is used to evaluate the reliability of the UAV in performing target recognition tasks based on the first and second recognition results.

[0032] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0033] Adjust the drone's sensor outputs based on the fault test signals;

[0034] Based on the flight status of the UAV after completing sensor output adjustments, image transformation is performed on the test image to obtain the target image;

[0035] Image recognition is performed on the test image and the target image respectively to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV.

[0036] The reliability of the UAV in performing target identification tasks is evaluated based on the first and second identification results.

[0037] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0038] Adjust the drone's sensor outputs based on the fault test signals;

[0039] Based on the flight status of the UAV after completing sensor output adjustments, image transformation is performed on the test image to obtain the target image;

[0040] Image recognition is performed on the test image and the target image respectively to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV.

[0041] The reliability of the UAV in performing target identification tasks is evaluated based on the first and second identification results.

[0042] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0043] Adjust the drone's sensor outputs based on the fault test signals;

[0044] Based on the flight status of the UAV after completing sensor output adjustments, image transformation is performed on the test image to obtain the target image;

[0045] Image recognition is performed on the test image and the target image respectively to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV.

[0046] The reliability of the UAV in performing target identification tasks is evaluated based on the first and second identification results.

[0047] The aforementioned reliability assessment method, apparatus, computer equipment, storage medium, and computer program product first adjust the UAV's sensor output based on the fault test signal, and then transform the test image to obtain the target image based on the UAV's flight state after the sensor output adjustment. Next, image recognition is performed on both the test image and the target image to obtain a first recognition result for the test image and a second recognition result for the target image. Finally, the reliability of the UAV performing the target recognition task is assessed based on the first and second recognition results. This approach comprehensively considers the impact of hardware-level faults propagating to the software level, incorporating both hardware and software systems into the reliability assessment scope, thus enabling a more comprehensive and accurate evaluation of the UAV's reliability in target recognition tasks. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a diagram illustrating the application environment of a task reliability assessment method in one embodiment;

[0050] Figure 2 This is a flowchart illustrating a task reliability assessment method in one embodiment;

[0051] Figure 3 This is a comparative diagram of target recognition results in one embodiment;

[0052] Figure 4 This is a flowchart illustrating the task reliability assessment method in another embodiment;

[0053] Figure 5 This is a structural block diagram of a task reliability assessment device in one embodiment;

[0054] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0056] In one embodiment, such as Figure 1 As shown, a method for assessing task reliability is provided. This embodiment uses the application of this method to a computer device as an example for illustration. It can be understood that, in situations such as... Figure 1 In the illustrated application environment, the computer device can specifically be terminal 102 or server 104. Terminal 102 communicates with server 104 via a network. The data storage system can store the data that server 104 needs to process. The data storage system can be integrated on server 104 or located in the cloud or on other network servers. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, and tablets. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0057] In one exemplary embodiment, such as Figure 2 As shown, a task reliability assessment method is provided, which can be applied to... Figure 1 Taking server 104 as an example, the following steps are included:

[0058] S202: Adjust the drone's sensor output based on the fault test signal.

[0059] The drone used for mission reliability assessment can be a multi-rotor drone, a fixed-wing drone, or a hybrid drone, etc., without specific limitations.

[0060] Optionally, during the reliability assessment of the UAV's target recognition task, the server can first obtain fault test signals from the terminal, determine the fault injection method, and complete the fault injection based on the fault test signals. For example, fault injection can be performed on key components of the UAV such as the gyroscope, accelerometer, magnetometer, positioning system, power system, and communication system. Taking the gyroscope as an example, the specific fault injection method can involve assigning fault values ​​to parameters such as the constant deviation of the three axes, scale factor deviation, random walk noise variance, and data impulse coefficient, thereby affecting the gyroscope's output.

[0061] In one optional implementation, to facilitate experimental analysis and cost control, a drone simulation platform can be deployed on the server. The drone's fault injection and the monitoring of the drone's flight status after fault injection can be accomplished through the simulation platform.

[0062] S204: Based on the flight status of the UAV after completing the sensor output adjustment, perform image transformation on the test image to obtain the target image.

[0063] Flight status includes, but is not limited to, flight position, flight speed, and flight attitude. Flight attitude can be characterized by attitude angles such as roll angle, pitch angle, and yaw angle, or Euler angles. Test images refer to images captured by the UAV in normal flight conditions without fault injection, and there may be one or more test images.

[0064] Optionally, after fault injection, the drone's perception of its own flight status may be affected, and the drone's flight status may change, thereby affecting the drone's ability to capture the target scene. In order to assess the impact of the change in flight status on the execution of the target recognition task, the server further obtains the drone's flight status after fault injection.

[0065] Furthermore, based on the camera's imaging principle and the UAV's flight state before and after fault injection, image transformation is performed on the test image to obtain the target image. Obtaining the target image through image transformation eliminates the need to control the UAV to photograph the target scene based on its flight attitude after fault injection, thereby improving the efficiency of the evaluation process.

[0066] S206: Perform image recognition on the test image and the target image respectively to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; wherein, the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV.

[0067] The identification results may include accuracy, precision, recall, etc., without specific limitations.

[0068] Optionally, after obtaining the target image, the server performs image recognition on the test image according to the target recognition algorithm used by the UAV to obtain a first recognition result, and performs image recognition on the target graphic according to the same target recognition algorithm to obtain a second recognition result.

[0069] S208: Based on the first and second identification results, assess the reliability of the UAV in performing the target identification task.

[0070] Optionally, the target recognition algorithm used by the UAV is typically based on deep learning neural network algorithms, such as the YOLO series and Faster CNN convolutional neural network algorithms. Since deep learning models are data-driven, changes in the input image can lead to variations in the target recognition performance of the deep learning algorithm, thus affecting the completion of the target recognition task. Therefore, the server further evaluates the reliability of the UAV's target recognition task by comparing and analyzing the first and second recognition results. For example, Figure 3 This is a comparative illustration of target recognition results in one embodiment, such as... Figure 3 As shown, one target is missing in the target recognition result after fault injection. The reliability of the UAV in performing target recognition tasks can be evaluated based on the target loss rate.

[0071] In the aforementioned reliability assessment method, the UAV's sensor output is first adjusted based on the fault test signal. Then, based on the UAV's flight state after the sensor output adjustment, the test image is transformed to obtain the target image. Next, image recognition is performed on both the test image and the target image to obtain a first recognition result for the test image and a second recognition result for the target image. Finally, the reliability of the UAV in performing the target recognition task is assessed based on the first and second recognition results. This approach comprehensively considers the impact of hardware-level faults on the software level, incorporating both hardware and software systems into the reliability assessment scope. As a result, the reliability of the UAV in performing the target recognition task can be assessed more comprehensively and accurately.

[0072] In one embodiment, the fault test signal includes a sensor identifier, a parameter identifier, and a parameter setting value; adjusting the UAV's sensor output based on the fault test signal includes: determining the target sensor based on the sensor identifier and obtaining the fault function corresponding to the target sensor; adjusting the output of the target sensor based on the parameter identifier, the parameter setting value, and the fault function.

[0073] The sensor identifier indicates the sensor used for output adjustment, such as a gyroscope, accelerometer, or magnetometer. The parameter identifier indicates the type of parameter used for output adjustment; for example, for a gyroscope, this could be constant deviation, scale factor deviation, random walk noise variance, or data impulse coefficient.

[0074] Optionally, during the fault injection process, the server first identifies the target sensor requiring output adjustment based on the sensor identifier and obtains the corresponding fault function for the target sensor. Then, based on the parameter identifier, parameter settings, and fault function, the server adjusts the output of the target sensor.

[0075] For example, the fault function of a gyroscope is as follows:

[0076]

[0077] In the formula, This indicates the adjusted output value. This represents the output value under normal flight conditions. Indicates the data impact factor. Indicates scale factor bias. Indicates constant deviation, This represents the variance of the random walk noise.

[0078] Under normal flight conditions , , , If the fault detection signal indicates that the gyroscope's scale factor deviation has been modified to 1.2, then... .

[0079] In this embodiment, by first identifying the target sensor based on the sensor identifier and obtaining the corresponding fault function of the target sensor, and then adjusting the output of the target sensor based on the parameter identifier, parameter setting value, and fault function, the sensor output can be easily adjusted according to the fault injection requirements.

[0080] In one embodiment, the flight state includes flight position and attitude angle; based on the flight state after the UAV completes sensor output adjustment, the test image is transformed to obtain the target image, including: obtaining a displacement matrix based on the flight position and a rotation matrix based on the attitude angle; and transforming the test image based on the rotation matrix and displacement matrix to obtain the target image.

[0081] Optionally, during the image transformation of the test image, the server first determines the displacement matrix based on the UAV's flight position before and after the fault injection, and determines the rotation matrix based on the UAV's attitude angles before and after the fault injection.

[0082] The flight position refers to the three-dimensional spatial position of the UAV, including the horizontal coordinates and the flight altitude (Z-axis direction).

[0083] Then, based on the displacement and rotation matrices, image transformation is performed on the test image to obtain the target image. The image transformation formulas involved include:

[0084]

[0085] In the formula, This represents the coordinates of the drone's camera in the geodetic coordinate system. Represents the rotation matrix. Represents the displacement matrix. This represents the intrinsic parameters of the drone's camera. Represents pixel coordinates.

[0086] In this embodiment, by first obtaining the displacement matrix based on the flight position and the rotation matrix based on the attitude angle, and then performing image transformation on the test image based on the rotation matrix and the displacement matrix to obtain the target image, the target image can be obtained quickly without controlling the UAV to take pictures of the target scene based on the flight attitude after fault injection, thereby improving the efficiency of the evaluation process.

[0087] In one embodiment, the first identification result includes a first number of identified targets and a first identification accuracy rate, and the second identification result includes a second number of identified targets and a second identification accuracy rate. Based on the first and second identification results, the reliability of the UAV performing the target identification task is evaluated, including: obtaining a target loss rate based on the first and second number of identified targets; obtaining a difference in identification accuracy based on the first and second identification accuracy rates; normalizing the target loss rate to obtain a first evaluation parameter, and normalizing the difference in identification accuracy to obtain a second evaluation parameter; weighted summing of the first and second evaluation parameters to obtain a target evaluation parameter; and evaluating the reliability of the UAV performing the target identification task based on the target evaluation parameter.

[0088] Among them, the number of targets identified refers to the number of targets that can be identified in an image by the target recognition algorithm, and the recognition accuracy rate refers to the proportion of the number of correctly identified targets to the total number of targets identified.

[0089] Optionally, after obtaining the first and second identification results, the server can calculate the target loss rate based on the number of first and second identified targets, and use the difference between the second and first identification accuracy rates as the identification accuracy difference. The specific calculation method for the target loss rate is as follows:

[0090]

[0091] In the formula, Indicates the target loss rate. Indicates the number of targets identified in the first phase. This indicates the number of second targets identified.

[0092] Furthermore, the server can normalize the target loss rate to obtain a first evaluation parameter, and normalize the difference in recognition accuracy to obtain a second evaluation parameter. The first and second evaluation parameters are then weighted and summed to obtain the target evaluation parameter. The weighting coefficients can be pre-set or manually assigned by the terminal; no specific limitations are made here.

[0093] In one optional implementation, the server directly outputs target evaluation parameters to the terminal as a reliability assessment index. The smaller the reliability assessment index, the higher the reliability of the UAV in performing the target recognition task.

[0094] In this embodiment, the target evaluation parameter is obtained by weighted summing the difference between the target loss rate and the recognition accuracy. Then, based on the target evaluation parameter, the reliability of the UAV in performing the target recognition task is evaluated, which can better measure the impact of UAV failure on target recognition, thereby improving the accuracy of the evaluation process.

[0095] In one embodiment, the reliability of the UAV performing the target recognition task is evaluated based on the target evaluation parameters, including: obtaining the target fault level corresponding to the flight state after the UAV completes sensor output adjustment based on the mapping relationship between flight state and fault level; and evaluating the reliability of the UAV performing the target recognition task based on the target fault level and the target evaluation parameters.

[0096] In one optional implementation, the impact of UAV malfunctions on flight status is also considered as a reliability assessment dimension. After obtaining the target evaluation parameters, the server can first acquire the changes in roll, pitch, and yaw angles after fault injection, and use the maximum value as the target angle. Then, based on the target angle and the mapping relationship between the target angle and the fault level, the target fault level after UAV fault injection is determined. For example, a target angle less than 5 degrees is defined as fault level 1, not less than 5 degrees and less than 10 degrees as fault level 2, not less than 10 degrees and less than 15 degrees as fault level 3, and not less than 15 degrees as level 4.

[0097] Furthermore, the reliability of the UAV in performing target identification tasks is evaluated based on the target fault level and target evaluation parameters.

[0098] In one optional implementation, the product of the target fault level and the target evaluation parameter can be used as a reliability assessment index. The smaller the reliability assessment index, the higher the reliability of the UAV in performing the target identification task. The specific calculation method is as follows:

[0099]

[0100] In the formula, Indicates reliability assessment metrics, Indicates the fault level. Indicates the first weighting coefficient. This represents the second weighting coefficient. Indicates normalization, Indicates the target loss rate. This represents the difference in recognition accuracy.

[0101] In this embodiment, by obtaining the target fault level corresponding to the flight state of the UAV after completing the sensor output adjustment according to the mapping relationship between flight state and fault level, and evaluating the reliability of the UAV in performing the target recognition task according to the target fault level and target evaluation parameters, the impact of UAV fault on UAV flight state and target recognition can be comprehensively considered, thereby improving the accuracy of the evaluation results.

[0102] In one embodiment, such as Figure 4 As shown, a method for assessing mission reliability is provided, which includes the following steps:

[0103] Acquire fault test signals; fault test signals include sensor identifiers, parameter identifiers, and parameter settings.

[0104] Based on the sensor identifier, identify the target sensor and obtain its corresponding fault function. Adjust the target sensor's output based on the parameter identifier, parameter settings, and fault function. Obtain the UAV's flight attitude after sensor output adjustment; flight status includes flight position and attitude angles.

[0105] Based on the flight position, obtain the displacement matrix, and based on the attitude angle, obtain the rotation matrix; based on the rotation matrix and displacement matrix, perform image transformation on the test image to obtain the target image.

[0106] Image recognition is performed on the test image and the target image respectively to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; wherein the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV.

[0107] The target loss rate is obtained based on the number of first targets identified and the number of second targets identified; the difference in recognition accuracy is obtained based on the first recognition accuracy and the second recognition accuracy.

[0108] The target loss rate is normalized to obtain the first evaluation parameter, and the difference in recognition accuracy is normalized to obtain the second evaluation parameter. The first evaluation parameter and the second evaluation parameter are weighted and summed to obtain the target evaluation parameter.

[0109] Based on the mapping relationship between flight status and fault level, the target fault level corresponding to the flight status of the UAV after completing sensor output adjustment is obtained.

[0110] The reliability of UAVs in performing target identification tasks is evaluated based on the target fault level and target evaluation parameters.

[0111] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0112] Based on the same inventive concept, this application also provides a task reliability assessment apparatus for implementing the task reliability assessment method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more task reliability assessment apparatus embodiments provided below can be found in the limitations of the task reliability assessment method described above, and will not be repeated here.

[0113] In one exemplary embodiment, such as Figure 5 As shown, a task reliability assessment device is provided, including: an adjustment module 510, a transformation module 520, an identification module 530, and an assessment module 540, wherein:

[0114] Adjustment module 510 is used to adjust the sensor output of the UAV according to the fault test signal;

[0115] The transformation module 520 is used to transform the test image according to the flight state of the UAV after the sensor output adjustment, so as to obtain the target image.

[0116] The recognition module 530 is used to perform image recognition on the test image and the target image respectively, and obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; wherein, the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV;

[0117] Evaluation module 540 is used to evaluate the reliability of the UAV in performing target recognition tasks based on the first recognition result and the second recognition result.

[0118] In one embodiment, the adjustment module 510 is further configured to determine the target sensor based on the sensor identifier and obtain the fault function corresponding to the target sensor; and adjust the output of the target sensor based on the parameter identifier, parameter setting value, and fault function.

[0119] In one embodiment, the transformation module 520 is further configured to obtain a displacement matrix based on the flight position and a rotation matrix based on the attitude angle; and to perform image transformation on the test image based on the rotation matrix and the displacement matrix to obtain the target image.

[0120] In one embodiment, the evaluation module 540 is further configured to obtain a target loss rate based on the first target recognition count and the second target recognition count; obtain a difference in recognition accuracy based on the first recognition accuracy and the second recognition accuracy; and evaluate the reliability of the UAV in performing the target recognition task based on the target loss rate and the difference in recognition accuracy.

[0121] In one embodiment, the evaluation module 540 is further configured to normalize the target loss rate to obtain a first evaluation parameter, and normalize the difference in recognition accuracy to obtain a second evaluation parameter; to perform a weighted summation of the first evaluation parameter and the second evaluation parameter to obtain a target evaluation parameter; and to evaluate the reliability of the UAV in performing the target recognition task based on the target evaluation parameter.

[0122] In one embodiment, the evaluation module 540 is further configured to obtain the target fault level corresponding to the flight state after the UAV completes the sensor output based on the mapping relationship between flight state and fault level; and to evaluate the reliability of the UAV in performing the target recognition task based on the target fault level and target evaluation parameters.

[0123] Each module in the aforementioned task reliability assessment device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0124] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 6As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores evaluation process data and other business data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When executed by the processor, the computer program implements a task reliability evaluation method.

[0125] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0126] In one exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps: adjusting the sensor output of a UAV based on a fault test signal; performing image transformation on a test image based on the flight state of the UAV after completing the sensor output adjustment to obtain a target image; performing image recognition on the test image and the target image respectively to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; wherein the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV; and evaluating the reliability of the UAV in performing the target recognition task based on the first recognition result and the second recognition result.

[0127] In one embodiment, when the processor executes the computer program, it further performs the following steps: determining the target sensor based on the sensor identifier and obtaining the fault function corresponding to the target sensor; adjusting the output of the target sensor based on the parameter identifier, parameter setting value and fault function.

[0128] In one embodiment, when the processor executes the computer program, it further performs the following steps: obtaining a displacement matrix based on the flight position and a rotation matrix based on the attitude angle; and performing image transformation on the test image based on the rotation matrix and the displacement matrix to obtain the target image.

[0129] In one embodiment, when the processor executes the computer program, it further performs the following steps: obtaining a target loss rate based on a first target recognition count and a second target recognition count; obtaining a difference in recognition accuracy based on a first recognition accuracy and a second recognition accuracy; and evaluating the reliability of the UAV performing the target recognition task based on the target loss rate and the difference in recognition accuracy.

[0130] In one embodiment, when the processor executes the computer program, it further performs the following steps: normalizing the target loss rate to obtain a first evaluation parameter, and normalizing the difference in recognition accuracy to obtain a second evaluation parameter; weighted summing of the first evaluation parameter and the second evaluation parameter to obtain a target evaluation parameter; and evaluating the reliability of the UAV in performing the target recognition task based on the target evaluation parameter.

[0131] In one embodiment, when the processor executes the computer program, it also performs the following steps: based on the mapping relationship between flight state and fault level, obtain the target fault level corresponding to the flight state after the UAV completes sensor output adjustment; and evaluate the reliability of the UAV in performing the target recognition task based on the target fault level and target evaluation parameters.

[0132] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it performs the following steps: adjusting the sensor output of a UAV based on a fault test signal; performing image transformation on a test image based on the flight state of the UAV after the sensor output adjustment to obtain a target image; performing image recognition on the test image and the target image respectively to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; wherein the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV; and evaluating the reliability of the UAV in performing the target recognition task based on the first recognition result and the second recognition result.

[0133] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: determining the target sensor based on the sensor identifier and obtaining the fault function corresponding to the target sensor; adjusting the output of the target sensor based on the parameter identifier, parameter setting value and fault function.

[0134] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: obtaining a displacement matrix based on the flight position and a rotation matrix based on the attitude angle; and performing image transformation on the test image based on the rotation matrix and the displacement matrix to obtain the target image.

[0135] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: obtaining a target loss rate based on a first target recognition count and a second target recognition count; obtaining a difference in recognition accuracy based on a first recognition accuracy rate and a second recognition accuracy rate; and evaluating the reliability of the UAV performing the target recognition task based on the target loss rate and the difference in recognition accuracy rate.

[0136] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: normalizing the target loss rate to obtain a first evaluation parameter, and normalizing the difference in recognition accuracy to obtain a second evaluation parameter; weighted summing of the first evaluation parameter and the second evaluation parameter to obtain a target evaluation parameter; and evaluating the reliability of the UAV in performing the target recognition task based on the target evaluation parameter.

[0137] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: obtaining the target fault level corresponding to the flight state of the UAV after completing the sensor output adjustment, based on the mapping relationship between flight state and fault level; and evaluating the reliability of the UAV in performing the target recognition task based on the target fault level and target evaluation parameters.

[0138] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps: adjusting the sensor output of a UAV based on a fault test signal; performing image transformation on a test image based on the flight state of the UAV after completing the sensor output adjustment to obtain a target image; performing image recognition on the test image and the target image respectively to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; wherein the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV; and evaluating the reliability of the UAV in performing the target recognition task based on the first recognition result and the second recognition result.

[0139] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: determining the target sensor based on the sensor identifier and obtaining the fault function corresponding to the target sensor; adjusting the output of the target sensor based on the parameter identifier, parameter setting value and fault function.

[0140] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: obtaining a displacement matrix based on the flight position and a rotation matrix based on the attitude angle; and performing image transformation on the test image based on the rotation matrix and the displacement matrix to obtain the target image.

[0141] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: obtaining a target loss rate based on a first target recognition count and a second target recognition count; obtaining a difference in recognition accuracy based on a first recognition accuracy rate and a second recognition accuracy rate; and evaluating the reliability of the UAV performing the target recognition task based on the target loss rate and the difference in recognition accuracy rate.

[0142] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: normalizing the target loss rate to obtain a first evaluation parameter, and normalizing the difference in recognition accuracy to obtain a second evaluation parameter; weighted summing of the first evaluation parameter and the second evaluation parameter to obtain a target evaluation parameter; and evaluating the reliability of the UAV in performing the target recognition task based on the target evaluation parameter.

[0143] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: obtaining the target fault level corresponding to the flight state of the UAV after completing the sensor output adjustment, based on the mapping relationship between flight state and fault level; and evaluating the reliability of the UAV in performing the target recognition task based on the target fault level and target evaluation parameters.

[0144] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0145] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0146] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0147] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method of mission reliability assessment, characterized by, The method includes: Adjust the drone's sensor outputs based on the fault test signals; Based on the flight state of the UAV after completing sensor output adjustments, the test image is transformed to obtain the target image; the flight state includes flight position and attitude angle. The step of performing image transformation on the test image to obtain the target image based on the flight state of the UAV after completing sensor output adjustment includes: Based on the flight position, obtain the displacement matrix, and based on the attitude angle, obtain the rotation matrix; Based on the rotation matrix and the displacement matrix, the test image is transformed to obtain the target image; Image recognition is performed on the test image and the target image respectively to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; wherein the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV; Based on the first identification result and the second identification result, the reliability of the UAV in performing the target identification task is evaluated; the first identification result includes a first number of identified targets and a first identification accuracy rate, and the second identification result includes a second number of identified targets and a second identification accuracy rate; The step of evaluating the reliability of the UAV in performing the target identification task based on the first identification result and the second identification result includes: The target loss rate is obtained based on the first target identification count and the second target identification count. The difference in recognition accuracy is obtained based on the first recognition accuracy and the second recognition accuracy. The reliability of the UAV in performing the target recognition task is evaluated based on the difference between the target loss rate and the recognition accuracy.

2. The method of claim 1, wherein, The fault test signal includes a sensor identifier, a parameter identifier, and a parameter setting value; adjusting the UAV's sensor output based on the fault test signal includes: Based on the sensor identifier, the target sensor is determined, and the fault function corresponding to the target sensor is obtained; The output of the target sensor is adjusted based on the parameter identifier, the parameter setting value, and the fault function.

3. The method of claim 1, wherein, The method of evaluating the reliability of the UAV in performing the target recognition task based on the difference between the target loss rate and the recognition accuracy includes: The target loss rate is normalized to obtain a first evaluation parameter, and the difference in recognition accuracy is normalized to obtain a second evaluation parameter. The first evaluation parameter and the second evaluation parameter are weighted and summed to obtain the target evaluation parameter; The reliability of the UAV in performing the target recognition task is evaluated based on the target evaluation parameters.

4. The method of claim 3, wherein, The step of evaluating the reliability of the UAV in performing the target recognition task based on the target evaluation parameters includes: Based on the mapping relationship between flight status and fault level, the target fault level corresponding to the flight status of the UAV after completing sensor output adjustment is obtained; The reliability of the UAV in performing the target identification task is evaluated based on the target fault level and the target evaluation parameters.

5. A mission reliability assessment apparatus, characterized by comprising: The device includes: The adjustment module is used to adjust the sensor output of the UAV based on the fault test signal; The transformation module is used to transform the test image to obtain a target image based on the flight state of the UAV after sensor output adjustment; the flight state includes flight position and attitude angle; the transformation of the test image to obtain the target image based on the flight state of the UAV after sensor output adjustment includes: obtaining a displacement matrix based on the flight position and a rotation matrix based on the attitude angle; and transforming the test image based on the rotation matrix and the displacement matrix to obtain the target image. The recognition module is used to perform image recognition on the test image and the target image respectively, to obtain a first recognition result corresponding to the test image and a second recognition result corresponding to the target image; wherein the recognition algorithm used in the recognition process is the same as the recognition algorithm used by the UAV; An evaluation module is used to evaluate the reliability of the UAV performing the target recognition task based on the first recognition result and the second recognition result; the first recognition result includes a first number of recognized targets and a first recognition accuracy rate, and the second recognition result includes a second number of recognized targets and a second recognition accuracy rate; the evaluation of the reliability of the UAV performing the target recognition task based on the first recognition result and the second recognition result includes: obtaining a target loss rate based on the first number of recognized targets and the second number of recognized targets; obtaining a difference in recognition accuracy based on the first recognition accuracy rate and the second recognition accuracy rate; and evaluating the reliability of the UAV performing the target recognition task based on the difference in recognition accuracy based on the target loss rate and the difference in recognition accuracy rate. 6.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-5 when the computer program is executed by the processor. When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

8. A computer program product comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images