Coaxial eight-blade manned multi-rotor aircraft detection method and device and storage medium

Abstract

This application discloses a detection method, device, and storage medium for a coaxial eight-rotor manned multirotor aircraft, relating to the field of aircraft control and fault diagnosis technology. Within a preset time period after the aircraft is powered on, the integrity of each component of the four power arms is checked in parallel to obtain self-test results. If the self-test result is passed, a preset amplitude excitation signal is applied to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification, obtaining pre-flight baseline data for each power arm. A five-dimensional residual vector is constructed based on the difference between the currently collected operational characteristic data and the pre-flight baseline data, and a comprehensive health score is determined for each power arm. When the updated comprehensive health score is less than a preset health score threshold, a target power arm is identified, and fault detection is performed based on the five-dimensional residual vector of the target power arm. This method can improve the self-test efficiency, health assessment accuracy, and fault location precision of manned multirotor aircraft.

Description

Technical Field

[0001] This application relates to the field of aircraft control and fault diagnosis technology, and in particular to the detection method, device and storage medium of coaxial eight-rotor manned multi-rotor aircraft. Background Technology

[0002] With the rise of Urban Air Mobility (UAM), manned multirotor aircraft, as the core carrier of electric vertical takeoff and landing (eVTOL) aircraft, are rapidly moving towards commercial operation. Among them, the coaxial eight-rotor configuration has become the mainstream design solution to ensure the safety of manned flight due to its high power redundancy, strong anti-disturbance capability, and structural symmetry. This configuration typically consists of four powered arms, each integrating a coaxial counter-rotating propeller, and achieves lift distribution and attitude stability through independent motors and ESCs in coordinated control. However, manned application scenarios place stringent requirements on system reliability, requiring a full-aircraft health status assessment to be completed within a very short time after power-on, and millisecond-level response and precise isolation of potential faults during flight to support fault-tolerant control or emergency landing strategies.

[0003] However, existing general self-testing methods are mostly geared towards small unmanned multi-rotor platforms and lack adaptability to manned, highly complex power systems, resulting in low efficiency and accuracy in fault detection. Summary of the Invention

[0004] The main purpose of this application is to provide a detection method, device and storage medium for a coaxial eight-rotor manned multi-rotor aircraft, which aims to solve the technical problems of low detection efficiency and insufficient fault location accuracy in traditional methods.

[0005] To achieve the above objectives, this application proposes a detection method for a coaxial eight-rotor manned multi-rotor aircraft, the detection method comprising: Within a preset time period after the aircraft is powered on, the integrity checks of the motor drivers, three-phase current sensors, speed encoders, inertial measurement units, and flight control communication links of the four power arms are initiated in parallel to obtain the self-test results. After the self-test result is that the self-test is passed, a preset amplitude excitation signal is applied to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification and obtain the pre-flight reference data corresponding to each power arm. A five-dimensional residual vector is constructed based on the difference between the currently collected operational characteristic data and the pre-flight baseline data, and the comprehensive health score corresponding to each power arm is determined based on the five-dimensional residual vector. During the real-time flight of the aircraft, the comprehensive health score is updated to obtain an updated comprehensive health score; When the updated comprehensive health score is less than a preset health score threshold, the five-dimensional residual vector of the target power arm corresponding to the updated comprehensive health score being less than the preset health score threshold is determined; Fault detection is performed based on the five-dimensional residual vector of the target power arm.

[0006] In one embodiment, the step of simultaneously initiating integrity checks on the motor drivers, three-phase current sensors, speed encoders, inertial measurement units, and flight control communication links of the four power arms within a preset time period after the aircraft is powered on, and obtaining self-test results, includes: Within a preset time period after the aircraft is powered on, power-on enable signals are sent in parallel to the power management units of the four power arms, so that the power management units sequentially activate the power supply rails of the corresponding motor drivers, three-phase current sensors, speed encoders, inertial measurement units, and flight control communication links based on the power-on enable signals. The hardware handshake protocol is initiated, and read commands are sent to each motor driver, heartbeat handshake commands are sent to the three-phase current sensor, speed encoder and flight control communication link, and self-test commands are sent to the inertial measurement unit to perform integrity detection. The system also receives power supply status data fed back by the motor driver, communication link response signals fed back by the three-phase current sensor, speed encoder and flight control communication link, and basic function response flags fed back by the inertial measurement unit. The self-test result is obtained based on the power supply status data, the communication link response signal, and the basic function response flag.

[0007] In one embodiment, the step of obtaining the self-test result based on the power supply status data, the communication link response signal, and the basic function response flag includes: The voltage monitoring register value of the motor driver is obtained based on the power supply status data; If the voltage monitoring register value is within a preset safe range, the first detection result is determined to be that the power supply is normal. If the communication link response signal is received within the preset time window, the second detection result is determined to be that the communication is normal; If the basic function response flag is set to a preset value, the third detection result is determined to be that the inertial measurement unit has no calibration deviation or hardware failure. The self-test result is obtained based on the first test result, the second test result, and the third test result.

[0008] In one embodiment, the method further includes: Acceleration and angular velocity data are collected from multiple consecutive sampling points of the inertial measurement units corresponding to the four power arms under static conditions. The mean zero-bias ratio of each inertial measurement unit in the three-axis directions is calculated based on the acceleration data and the angular velocity data. Calculate the root mean square value between any two inertial measurement units based on the mean zero bias values ​​in the three-axis directions; The root mean square value is compared with a preset zero bias difference threshold. When the root mean square value is greater than the preset zero bias difference threshold, a verification result is generated indicating that the inertial measurement unit has installation stress or abnormal device drift. The verification results are sent to the user for re-inspection until the inertial measurement unit is free from installation stress or device drift abnormalities. Then, a pre-flight power link verification is performed by applying a preset amplitude excitation signal to the upper and lower coaxial motor assemblies of each power arm to obtain the pre-flight reference data corresponding to each power arm.

[0009] In one embodiment, the step of applying a preset amplitude excitation signal to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification after the self-test result is a pass, and obtaining the pre-flight reference data corresponding to each power arm, includes: When the ground is stationary, pulse width modulation excitation signals with preset amplitude and preset frequency and preset duty cycle are sent sequentially to the upper and lower motors of each power arm. The instantaneous values ​​of the three-phase current, motor speed feedback, fuselage vibration acceleration, and flight control communication link delay data of each power arm under the action of the pulse width modulation excitation signal are collected simultaneously. The instantaneous values ​​of the three-phase current are filtered to calculate the mean current and total harmonic distortion rate; the motor speed feedback value is smoothed to extract the steady-state speed deviation; the fuselage vibration acceleration value is subjected to fast Fourier transform to calculate the proportion of high-frequency vibration energy; and the standard deviation of flight control communication link delay jitter is calculated based on the flight control communication link delay data. The mean current, total harmonic distortion rate, steady-state speed deviation, high-frequency vibration energy ratio, and standard deviation of flight control communication link delay jitter are used as five-dimensional initial feature data, and the five-dimensional initial feature data are saved as pre-flight reference data for each power arm.

[0010] In one embodiment, the step of constructing a five-dimensional residual vector based on the difference between the currently collected operational characteristic data and the pre-flight baseline data, and determining the comprehensive health score corresponding to each power arm based on the five-dimensional residual vector, includes: Real-time acquisition of the mean current, harmonic distortion rate, speed deviation, high-frequency vibration energy ratio, and standard deviation of delay jitter of each power arm at the current moment. The differences between the real-time average current, the real-time harmonic distortion rate, the real-time speed deviation, the real-time high-frequency vibration energy ratio, and the standard deviation of the real-time flight control communication link delay jitter and the corresponding items in the pre-flight reference data are calculated respectively to obtain the first residual component, the second residual component, the third residual component, the fourth residual component, and the fifth residual component. The first residual component to the fifth residual component are combined in sequence to construct the five-dimensional residual vector at the current time. After standardizing the five-dimensional residual vector at the current moment, calculate the probability density value of belonging to the healthy state; Based on the probability density value, a comprehensive health score corresponding to each power arm is obtained by mapping it through a preset scoring mapping function.

[0011] In one embodiment, the step of fault detection based on the five-dimensional residual vector of the target power arm includes: The five-dimensional residual vector of the target power arm is reduced in dimension by principal component analysis to obtain the dimension-reduced residual vector. The Mahalanobis distance between the dimension-reduced residual vector and the various fault prototype vectors in the pre-stored fault prototype library is calculated to obtain the matching score set. The pre-stored fault prototype library stores at least one or more fault types, such as propeller damage, motor bearing wear, ESC response lag, sensor drift, motor winding abnormality, and propeller loosening, and the fault prototype vector is associated with the power arm number, upper / lower coaxial position identifier, and component type identifier. The fault type corresponding to the maximum matching score is selected from the matching score set as the fault detection result of the target boom. When the maximum matching score is greater than the preset matching threshold and the difference between the maximum matching score and the second largest matching score is greater than the preset confidence threshold, the fault detection result is locked as the final fault type.

[0012] In one embodiment, the method further includes: Based on the final fault type, the target power arm, and the upper / lower coaxial position identifier and component type identifier associated with the fault prototype vector, the specific component identifier where the fault occurred is parsed out; Based on the specific component identifier, query the preset control allocation matrix adjustment table to obtain the weight suppression parameters for the target power arm and the torque compensation parameters for the other normal power arms; The output commands of the faulty motor in the target power arm are reduced or cut off based on the weight suppression parameter, and the motor speed commands of the remaining normal power arms are redistributed using the torque compensation parameter to maintain the attitude stability of the aircraft.

[0013] Furthermore, to achieve the above objectives, this application also proposes a detection device for a coaxial eight-rotor manned multi-rotor aircraft, the detection device comprising: The detection module is used to perform parallel integrity checks on the motor drivers, three-phase current sensors, speed encoders, inertial measurement units, and flight control communication links of the four power arms within a preset time period after the aircraft is powered on, and obtain self-test results. The verification module is used to apply a preset amplitude excitation signal to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification after the self-test result is that the self-test is passed, so as to obtain the pre-flight reference data corresponding to each power arm. The construction module is used to construct a five-dimensional residual vector based on the difference between the currently collected operational feature data and the pre-flight baseline data, and to determine the comprehensive health score corresponding to each power arm based on the five-dimensional residual vector. The update module is used to update the comprehensive health score during the real-time flight of the aircraft to obtain an updated comprehensive health score. The determination module is used to determine the five-dimensional residual vector of the target power arm corresponding to the updated comprehensive health score being less than the preset health score threshold when the updated comprehensive health score is less than the preset health score threshold. The detection module is also used for fault detection based on the five-dimensional residual vector of the target power arm.

[0014] In addition, to achieve the above objectives, this application also proposes a testing device for a coaxial eight-rotor manned multi-rotor aircraft. The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The computer program is configured to implement the steps of the testing method for the coaxial eight-rotor manned multi-rotor aircraft as described above.

[0015] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the detection method for a coaxial eight-rotor manned multi-rotor aircraft as described above.

[0016] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the detection method for a coaxial eight-rotor manned multi-rotor aircraft as described above.

[0017] This application establishes pre-flight baseline data by performing integrity checks on four powered arms in parallel during the power-on phase and applying preset amplitude excitation signals to the upper and lower coaxial motor assemblies of each powered arm after the self-check passes. During flight, a five-dimensional residual vector is constructed based on the difference between the current operating characteristic data and the pre-flight baseline data, and this five-dimensional residual vector is mapped to a comprehensive health score. When the comprehensive health score of any powered arm is lower than a preset health score threshold, fault detection is performed based on the five-dimensional residual vector of that powered arm. This enables rapid pre-flight self-checks, dynamic in-flight health assessments, and fault location, providing a diagnostic basis for the fault-tolerant control of manned multi-rotor aircraft. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating the first embodiment of the detection method for the coaxial eight-rotor manned multi-rotor aircraft of this application; Figure 2 A flowchart illustrating the second embodiment of the detection method for the coaxial eight-rotor manned multi-rotor aircraft of this application; Figure 3 A flowchart illustrating Embodiment 3 of the detection method for the coaxial eight-rotor manned multi-rotor aircraft of this application; Figure 4 This is a schematic diagram of the module structure of the detection device for the coaxial eight-rotor manned multi-rotor aircraft according to an embodiment of this application; Figure 5 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the detection method of the coaxial eight-rotor manned multi-rotor aircraft in the embodiments of this application.

[0021] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0022] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0023] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0024] The main solution of this application embodiment is as follows: within a preset time period after the aircraft is powered on, the integrity detection of the corresponding devices of each of the four power arms is started in parallel to obtain the self-test result; after the self-test result is that the self-test is passed, a preset amplitude excitation signal is applied to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification to obtain the pre-flight reference data corresponding to each power arm; during the operation of the aircraft, a five-dimensional residual vector is constructed based on the difference between the currently collected operation feature data and the pre-flight reference data, and the comprehensive health score corresponding to each power arm is determined accordingly; when the updated comprehensive health score of any power arm is less than the preset health score threshold, the corresponding target power arm is determined, and fault detection is performed based on the five-dimensional residual vector of the target power arm.

[0025] Because existing general self-testing methods are mostly geared towards small unmanned multi-rotor platforms, they lack adaptability to highly complex manned power systems. Firstly, traditional self-testing processes rely on sequential, item-by-item checks, making it difficult to complete the integrity verification of eight motors, eight propellers, ESC communication links, power distribution networks, and flight control buses within seconds. Secondly, fault diagnosis is often based on single sensor threshold judgments (such as current over-limit), failing to effectively distinguish between combined fault modes such as coupling interference between coaxial rotors, ESC response delays, or mechanical jamming. Thirdly, existing solutions lack a joint residual model for multi-source heterogeneous data (including three-phase current waveforms, speed feedback, IMU vibration spectrum, and CAN bus communication status), resulting in coarse-grained fault location, often only able to identify "one side of the arm" rather than pinpointing a specific motor or propeller, severely restricting the accuracy and safety of subsequent control reconfiguration.

[0026] This application provides a solution that enables parallel and hierarchical verification of the entire power chain during the power-on self-start phase, and achieves precise isolation of faults to specific power units based on multi-dimensional sensor residual characteristics during flight. This constructs a highly reliable health status assessment system, realizing second-level full system self-check, high-precision fault isolation, and closed-loop health management, which can improve the safety and maintenance efficiency of manned aircraft.

[0027] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device capable of performing the above functions, such as a testing device for a coaxial octagonal manned multirotor aircraft. The following description uses a testing device for a coaxial octagonal manned multirotor aircraft as an example to illustrate this embodiment and the following embodiments. All actions involving the acquisition of signals, information, or data in this application are performed in accordance with the relevant data protection regulations of the country where the application is located and with authorization from the owner of the corresponding device.

[0028] Based on this, the embodiments of this application provide a detection method for a coaxial eight-rotor manned multi-rotor aircraft, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the detection method for the coaxial eight-rotor manned multi-rotor aircraft of this application.

[0029] In this embodiment, the detection method for the coaxial eight-rotor manned multi-rotor aircraft includes steps S10 to S60: Step S10: Within a preset time period after the aircraft is powered on, the integrity of the motor drivers, three-phase current sensors, speed encoders, inertial measurement units and flight control communication links of the four power arms is started in parallel to obtain the self-test results. Each of the four power arms is equipped with upper and lower coaxial motor assemblies to form a coaxial eight-propeller configuration.

[0030] It should be noted that the preset time period can be set according to requirements, such as 2.5s, 2s, etc. That is, no more than 2 seconds after the main control system of the coaxial eight-rotor manned multi-rotor aircraft is powered on, the integrity detection of the motor drivers, three-phase current sensors, speed encoders, inertial measurement units and flight control communication links of the four power arms of the aircraft is started in parallel. Specifically, this includes: power supply status confirmation, communication link connectivity verification and basic function response test.

[0031] Understandably, each of the four power arms is equipped with upper and lower coaxial motor assemblies, and each motor assembly drives a propeller. The four power arms are equipped with a total of eight propellers, thus forming a stable lift configuration of eight coaxial propellers. This can improve power redundancy while reducing the overall rotor footprint, adapting to the spatial layout requirements of manned aircraft.

[0032] In practice, if any of the three stages of the test fails, it means that the self-test result is a failure. At this time, the fault code can be recorded and output to the flight control log.

[0033] Step S20: After the self-test result is that the self-test is passed, apply a preset amplitude excitation signal to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification and obtain the pre-flight reference data corresponding to each power arm.

[0034] It should be noted that if the self-test result is that the self-test passes, the next stage of testing can be carried out, namely the pre-flight power link verification. This stage of verification is carried out when the aircraft is on the ground in standby phase, so as to obtain the pre-flight reference data corresponding to each power arm.

[0035] It should be noted that the parameters of the preset amplitude excitation signal can be set according to actual needs, for example, set to 20% to 30% of the rated operating current, to ensure that the motor assembly can operate stably and generate measurable dynamic response under no-load conditions. By applying this excitation signal to the upper and lower coaxial motor assemblies, multi-source data of each power arm under steady-state operating conditions are collected, including but not limited to three-phase current waveform characteristics, speed fluctuation range, high-frequency vibration spectrum distribution, and communication bus delay characteristics. After filtering and normalization, these data form the pre-flight baseline dataset for each power arm and are stored in the non-volatile storage area of ​​the flight control unit for reference in subsequent health status assessment and fault diagnosis.

[0036] In one feasible implementation, step S20 may include steps A11 to A14: Step A11: When the ground is stationary, send pulse width modulation excitation signals with preset amplitude, preset frequency and preset duty cycle to the upper and lower motors of each power arm in sequence. In practice, while the aircraft is stationary on the ground, pulse width modulation (PWM) excitation signals with preset frequency, preset duty cycle, and preset amplitude are sequentially sent to the upper and lower motors of each power arm to ensure signal stability and consistency. This method effectively simulates the power output of the aircraft during actual operation while avoiding data deviations caused by external environmental interference.

[0037] It should be noted that the preset amplitude is a low amplitude that does not produce significant lift. The preset frequency is 50–200 Hz, the preset duty cycle is 10%–20%, and the duration is 300–800 ms. In this embodiment, the frequency is set to 100 Hz, the duty cycle is set to 15%, and the duration is set to 500 ms. This combination of parameters can effectively stimulate the dynamic response characteristics of the motor and transmission system without causing significant lift.

[0038] Step A12: Synchronously collect the instantaneous values ​​of the three-phase current, motor speed feedback values, fuselage vibration acceleration values, and flight control communication link delay data of each power arm under the action of the pulse width modulation excitation signal; In practice, the three-phase current waveform, speed feedback data and vibration spectrum of each power arm under the action of this signal can be collected simultaneously to obtain the instantaneous value of the three-phase current, the motor speed feedback value and the vibration acceleration value of the machine body.

[0039] In one feasible implementation, the pulse width modulation excitation signal can also be a swept-frequency sine wave instead of a fixed-frequency PWM, covering a frequency range of 10Hz to 200Hz, to excite more comprehensive mechanical resonance characteristics. Vibration spectrum analysis uses wavelet packet decomposition to extract multi-scale energy features, combined with a support vector machine (SVM) classifier for early microcrack detection.

[0040] Flight control communication link delay data is the transmission time difference between the flight control sending a command and the motor driver responding. It is obtained by synchronous counting through timestamps and reflects the stability and reliability of the communication link.

[0041] Step A13: Filter the instantaneous values ​​of the three-phase currents and calculate the mean current and total harmonic distortion rate; smooth the motor speed feedback values ​​and extract the steady-state speed deviation; perform a fast Fourier transform on the fuselage vibration acceleration values ​​and calculate the proportion of high-frequency vibration energy; and calculate the standard deviation of flight control communication link delay jitter based on the flight control communication link delay data. It should be noted that the instantaneous values ​​of the three-phase current can be filtered to calculate the average current and total harmonic distortion rate, and the motor speed feedback value can be smoothed to extract the steady-state speed deviation. Then, the fuselage vibration acceleration value can be processed by fast Fourier transform to obtain the proportion of high-frequency vibration energy, and the standard deviation of flight control communication link delay jitter can be calculated based on the flight control communication link delay data.

[0042] Step A14: Use the mean current, total harmonic distortion rate, steady-state speed deviation, high-frequency vibration energy ratio, and standard deviation of flight control communication link delay jitter as five-dimensional initial feature data, and save the five-dimensional initial feature data as pre-flight reference data for each power arm.

[0043] Understandably, the calculated average current, total harmonic distortion rate, steady-state speed deviation, and high-frequency vibration energy ratio can be compared with the benchmark values ​​to determine whether there are any anomalies. If there are no anomalies, the above data can be used as initial characteristic data and then saved as the pre-flight benchmark data for each power arm.

[0044] It should be noted that this embodiment pre-sets a health baseline digital model, which allows for the calculation of the dynamic response residuals of each channel based on this data model, thereby determining whether there is mechanical jamming, abnormal ESC response, or propeller installation deviation. First, the currently valid health baseline digital model is loaded from non-volatile memory. This health baseline digital model uses the feature vectors of each power unit... For the object (default d=5), a statistical baseline is established under standard conditions; the standard conditions are defined as: the aircraft is stationary on the ground, the propeller is free from external load interference, the ambient temperature is 15–35℃, the battery SOC is 30%–90%, and the pre-takeoff rest time is ≥60s; the baseline data storage format is the mean vector of each power unit. With covariance matrix (or mean + variance), and save metadata such as sampling rate, filter parameters, and feature extraction frequency band to ensure reproducibility. During this process, the three-phase current sensor operates at a rate no lower than... The sampling frequency acquires the instantaneous current value of each phase, the speed encoder outputs motor speed feedback with a resolution better than 0.1%, and the triaxial accelerometer... The sampling rate records the vibration signal at the base of the boom. All acquired data is transmitted to the main control system's buffer via a high-speed DMA channel to avoid CPU interrupt overhead affecting synchronization accuracy. The steady-state rotational speed deviation... The vibration signal is processed by Fast Fourier Transform (FFT) to calculate... Energy percentage of the above frequency bands These features are compared with their corresponding values ​​in the healthy baseline model to calculate the dynamic response residuals:

[0045] in For the first Measured values ​​of each characteristic quantity Its baseline value, The baseline standard deviation. If any residual... If the value exceeds a preset threshold (e.g., 3.0), the power unit is deemed to be malfunctioning, requiring further analysis of the specific failure mode. The execution of this entire process ensures that potential mechanical or electrical hazards are eliminated before actual takeoff, thereby improving flight safety.

[0046] Understandably, the health baseline digital model is updated online after each maintenance. By comparing the KL divergence between the current and historical baselines, a baseline recalibration process is triggered when the difference exceeds a preset difference threshold (e.g., 0.15). The calibration process is automatically executed while the device is stationary on the ground, requiring no manual intervention. The CAN bus communication status monitoring includes baud rate consistency verification, node response timeout statistics, and error frame counting. If a single boom node accumulates more than 10 error frames within one second or experiences three consecutive response timeouts, its communication link is determined to have a physical layer or protocol layer fault. Furthermore, in other embodiments, the health baseline model can also be updated based on the calibration data after maintenance or historical flight data.

[0047] Step S30: Construct a five-dimensional residual vector based on the difference between the currently collected operational feature data and the pre-flight baseline data, and determine the comprehensive health score corresponding to each power arm based on the five-dimensional residual vector.

[0048] In practice, operational characteristic data of each power arm can be collected in real time, and the operational characteristic data can be compared with the corresponding items in the pre-flight baseline data to obtain a five-dimensional residual vector. The five-dimensional residual vector is further standardized and mapped to obtain the comprehensive health score corresponding to each power arm.

[0049] In one feasible implementation, step S30 may include steps A21 to A25: Step A21: Real-time acquisition of the mean current, harmonic distortion rate, speed deviation, high-frequency vibration energy ratio, and standard deviation of delay jitter of each power arm at the current moment. In practice, multiple key parameters of each boom are collected in real time, including the real-time average current, real-time current harmonic distortion rate, real-time rotational speed deviation, real-time high-frequency vibration energy ratio, and real-time flight control communication link delay jitter standard deviation. These parameters are acquired synchronously through a sensor network distributed across each boom, ensuring the temporal consistency and spatial correlation of the data.

[0050] Step A22: Calculate the difference between the real-time average current, the real-time harmonic distortion rate, the real-time speed deviation, the real-time high-frequency vibration energy ratio, and the standard deviation of the real-time flight control communication link delay jitter and the corresponding items in the pre-flight reference data, respectively, to obtain the first residual component, the second residual component, the third residual component, the fourth residual component, and the fifth residual component. In practice, the differences between the real-time collected parameters and the corresponding items in the pre-flight baseline data are calculated to obtain five-dimensional residual components. These residual components reflect the degree of deviation between the current actual operating state of the boom and the baseline state, providing a quantitative basis for subsequent health assessments.

[0051] Step A23: Combine the first residual component to the fifth residual component in sequence to construct the five-dimensional residual vector at the current time. In one feasible implementation, the preset health state model is a health state model based on a Gaussian mixture distribution. The preset health state model takes a five-dimensional residual vector as input, which includes the mean current residual, total harmonic distortion rate residual, rotational speed deviation residual, high-frequency vibration energy proportion residual, and flight control communication link delay jitter standard deviation residual. The preset health state model is trained based on healthy samples and typical fault samples collected during the offline calibration phase, and is used to calculate the probability density value of the current five-dimensional residual vector belonging to a healthy state.

[0052] During online detection, the main control system inputs the current five-dimensional residual vector into the preset health status model to obtain the corresponding health status probability density value, and maps the health status probability density value into a comprehensive health score through a preset scoring mapping function.

[0053] In practical implementation, a comprehensive health score for each power unit can be generated through a pre-trained multidimensional joint residual analysis model. The multidimensional joint residual analysis model is a Gaussian mixture model (GMM) with an input dimension of d=5 (or expanded to d=8 / 10), corresponding to the mean deviation of current, the content of the 5th harmonic, the steady-state error of rotational speed, etc. Vibration energy percentage and communication delay jitter standard deviation; the mixture component K is selected within the range of K=3–8 using the BIC criterion (default K=5); the model is trained using the EM algorithm, with a maximum number of iterations ≤200, and the log-likelihood increment is less than 10 consecutive iterations. The training samples are healthy samples and typical fault samples collected during the offline calibration phase. The samples are z-score standardized before input and the standardized parameters are saved for consistent use online.

[0054] The multidimensional joint residual analysis model is a probabilistic graphical model based on Gaussian mixture distribution. Its input dimensions are preset numbers, corresponding to the mean deviation of current, the content of specific harmonics, the steady-state error of rotational speed, the proportion of vibration energy above a specific frequency, and the standard deviation of communication delay jitter, respectively. The model parameters are obtained by training multiple sets of healthy samples and multiple sets of typical fault samples collected during the offline calibration stage. The convergence threshold is set to the point where the change in the log-likelihood function is less than the preset convergence threshold.

[0055] The construction process of this multidimensional joint residual analysis model is as follows: First, during the offline calibration phase, at least 500 sets of aircraft operation data under healthy conditions and 200 sets of sample data covering typical fault modes such as motor winding short circuits, ESC MOSFET aging, propeller micro-cracks, and bearing wear were collected. For each sample, a 5-dimensional feature vector was extracted. ,in The mean deviation of current (unit: A). The content of the 5th harmonic (unit: %). Steady-state speed error (unit: ), for The above percentages of vibration energy (unit: %) Standard deviation of CAN bus communication delay jitter (unit: Subsequently, the Gaussian Mixture Model (GMM) is fitted using the Expectation-Maximization (EM) algorithm, and its probability density function is expressed as:

[0056] in The number of components in the mixture. For the first The weights of each Gaussian component, and These are their mean vector and covariance matrix, respectively. During training, when the change in the log-likelihood function over 10 consecutive iterations is less than... When the model converges, it is determined that the model has converged.

[0057] Step A24: After standardizing the five-dimensional residual vector at the current moment, calculate the probability density value of belonging to the healthy state; It should be noted that the above formula can be used to calculate the probability density value of a healthy state, thereby obtaining the probability density value of each power arm.

[0058] Step A25: Based on the probability density values, a comprehensive health score corresponding to each power arm is obtained by mapping.

[0059] When used online, the main control system inputs the 5-dimensional feature vectors collected in real time into the GMM and calculates its log-likelihood value. And map it to a comprehensive health score of 0-100. ,in For the Sigmoid function, and To calibrate the parameters, ensure that a healthy state corresponds to a high score (>90) and a serious fault corresponds to a low score (<30).

[0060] Therefore, the comprehensive health score corresponding to each power arm can be obtained based on the current five-dimensional residual vector.

[0061] By constructing a joint residual model based on four-dimensional heterogeneous data of current, speed, vibration and communication, it can effectively distinguish complex fault modes such as ESC response delay, motor winding short circuit and propeller micro-cracks. The fault location granularity is accurate to the specific motor or propeller, and the location accuracy is greater than or equal to the preset accuracy threshold.

[0062] Step S40: During the real-time flight of the aircraft, the comprehensive health score is updated to obtain an updated comprehensive health score.

[0063] It should be noted that after the aircraft begins real-time flight, the health scores of each power arm can be continuously updated at a preset cycle during the flight. For example, the construction of the five-dimensional residual vector and the calculation of the comprehensive health score can be performed cyclically every 20ms control cycle. The latest comprehensive health score calculated each time will overwrite the score of the previous cycle and be used as the updated comprehensive health score at the current moment. At the same time, the comprehensive health score at the start of the flight will be recorded as the health score benchmark for subsequent trend comparison.

[0064] Step S50: When the updated comprehensive health score is less than the preset health score threshold, determine the five-dimensional residual vector of the target power arm corresponding to the updated comprehensive health score being less than the preset health score threshold.

[0065] It should be noted that when the updated comprehensive health score of any powered arm is less than a preset health score threshold, a fault isolation mechanism can be triggered. Specifically, the five-dimensional residual vector of the target powered arm with the fault is first determined. The preset health score threshold can be determined based on health and fault samples collected during the offline calibration phase, or it can be preset according to the aircraft model, power system parameters, and safety level requirements. When the updated comprehensive health score corresponding to any powered arm is less than the preset health score threshold, the fault isolation process is triggered.

[0066] Step S60: Perform fault detection based on the five-dimensional residual vector of the target power arm.

[0067] In practice, the five-dimensional residual vector of the target power arm can be used to perform similarity matching between the currently observed residual vector and the pre-stored fault prototype vectors by using the residual space projection algorithm and the Mahalanobis distance metric after dimensionality reduction by principal component analysis. This allows for the determination of the specific fault type and fault location.

[0068] This embodiment uses parallelized hardware handshake and multi-channel synchronous excitation to compress the integrity verification of the entire power link, including 8 motors, 8 propellers, power distribution network and flight control bus, into a predetermined time period. This significantly speeds up the process compared to traditional item-by-item testing and meets the strict pre-flight preparation time requirements for manned aircraft.

[0069] This embodiment provides a detection method for a coaxial eight-rotor manned multi-rotor aircraft. The method involves performing integrity checks on four powered arms in parallel within a preset time period after the aircraft is powered on, and establishing pre-flight baseline data for each arm after a successful self-test. During flight, a five-dimensional residual vector is constructed based on the difference between the current operational characteristic data and the pre-flight baseline data, and this vector is mapped to a comprehensive health score. When the comprehensive health score of any powered arm falls below a preset health score threshold, the system performs fault detection based on the five-dimensional residual vector of that arm. This improves the efficiency of pre-flight self-tests, the accuracy of in-flight health assessments, and the precision of fault location.

[0070] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 Step S10 includes steps S101 to S103: Step S101: Within a preset time period after the aircraft is powered on, power-on enable signals are sent in parallel to the power management units of the four power arms, so that the power management units sequentially activate the power rails of the corresponding motor drivers, three-phase current sensors, speed encoders, inertial measurement units, and flight control communication links based on the power-on enable signals.

[0071] It should be noted that the parallel self-test process in this embodiment adopts a scheduling mechanism that combines time-slice polling and interrupt response to ensure that the testing tasks of the four power arms are completed within a predetermined time period. Specifically, power-on enable signals can be sent to the power management units (PMUs) of the four power arms in parallel. After receiving the enable signal, each PMU sequentially activates the power supply rails of its downstream motor driver, current sensor, encoder, and IMU module.

[0072] Step S102: Initiate the hardware handshake protocol, send read commands to each motor driver, send heartbeat handshake commands to the three-phase current sensor, speed encoder and flight control communication link, and send a self-test command to the inertial measurement unit to perform integrity detection, and receive power supply status data fed back by the motor driver, communication link response signals fed back by the three-phase current sensor, speed encoder and flight control communication link, and basic function response flags fed back by the inertial measurement unit.

[0073] Specifically, the motor driver status is obtained via SPI bus polling, communication link connectivity is verified via CAN bus loopback testing, and the inertial measurement unit initialization status is read from its internal self-test flag via the I2C interface. The parallel self-test process is uniformly scheduled by the real-time scheduling module in the aircraft's main control system, ensuring that the detection tasks corresponding to the four power arms are executed in parallel within a preset time period. The real-time scheduling module can schedule power supply status detection, communication link detection, and basic function response detection according to preset task priorities to guarantee that the self-test process is completed within the preset time period.

[0074] It should be noted that a hardware handshake protocol can be initiated, which defines a three-stage interaction process: the first stage is power supply status confirmation, i.e., sending read commands to each motor driver to confirm the power supply status; the second stage is communication link connectivity verification. In one feasible implementation, the hardware handshake protocol includes two stages: handshake request and handshake response. The main control system sends handshake request information to the communication nodes corresponding to each power arm. The handshake request information includes node identifier, sequence number, and verification information. After receiving the handshake request information, each communication node returns handshake response information within a preset time window. The handshake response information includes the corresponding node identifier, sequence number, and node status code. The main control system determines whether the corresponding communication link is normal based on the handshake response information.

[0075] The flight control communication link detection includes one or more of the following: node response timeout statistics, error frame count, communication link delay statistics, and communication link delay jitter statistics. The main control system sends handshake commands to the communication nodes corresponding to each power arm within a preset time window and receives the corresponding communication link response signals. When any communication node fails to return a communication link response signal within the preset time window, or when the error frame count exceeds a preset error frame threshold, the flight control communication link of the corresponding power arm is determined to be abnormal.

[0076] In specific implementation, the system can receive the internal voltage monitoring register value returned by the motor driver, the communication link response signals fed back by the three-phase current sensor, speed encoder and each node of the flight control communication link, and the basic function response flag bit returned by the inertial measurement unit.

[0077] Step S103: Obtain the self-test result based on the power supply status data, the communication link response signal, and the basic function response flag.

[0078] In practice, power supply status data, communication link response signals, and basic function response flags can be tested according to the set rules to obtain self-test results.

[0079] In one feasible implementation, step S103 may include steps B11 to B15: Step B11: Obtain the voltage monitoring register value of the motor driver based on the power supply status data; It should be noted that the power supply status data includes the voltage monitoring register value of the motor driver, which reflects the stability and health status of the internal power supply circuit of the motor driver. By analyzing the voltage monitoring register value, it is possible to determine whether there are problems such as overvoltage, undervoltage, or abnormal power fluctuations.

[0080] Step B12: If the voltage monitoring register value is within the preset safe range, determine the first detection result as normal power supply; The preset safety zone range can be set according to needs, for example, set to If the voltage health register value is within this range, the power supply is considered normal; if it does not fall within this range, the power supply is considered abnormal.

[0081] Step B13: If the communication link response signal is received within the preset time window, the second detection result is determined to be that the communication is normal; The preset time window can be pre-set according to the communication cycle and node response delay requirements of the flight control communication link. After the main control system sends a handshake command to the communication node corresponding to each power arm, it waits for the corresponding communication node to return a communication link response signal within the preset time window. If the communication link response signal is received within the preset time window and the error frame count does not exceed the preset error frame threshold, the second detection result is determined to be normal communication; otherwise, the second detection result is determined to be abnormal communication.

[0082] Step B14: If the basic function response flag is set to a preset value, determine that the third detection result is that the inertial measurement unit has no calibration deviation or hardware failure. It should be noted that the preset value is set to 0x55. If the basic function response flag is 0x55, it indicates that the IMU has completed the bias calibration of the internal gyroscope and accelerometer and there is no hardware fault. That is, the third detection result is that the inertial measurement unit does not have calibration deviation or hardware fault.

[0083] Step B15: Obtain the self-test result based on the first test result, the second test result, and the third test result.

[0084] Understandably, the three test results can be considered together as the self-test result. If any one of the test results is abnormal, the self-test result is considered failed; if all three test results are normal, the self-test result is considered passed. For example, if the first test result is a power supply abnormality, regardless of the second and third test results, the self-test result is considered failed.

[0085] In one feasible implementation, in addition to integrity detection, inertial measurement unit consistency verification can also be performed simultaneously. Therefore, the method further includes steps S11 to S16: Step S11: Collect acceleration and angular velocity data from multiple consecutive sampling points of the inertial measurement units corresponding to the four power arms under static conditions; Understandably, the inertial measurement unit (IMU) consistency verification is performed synchronously in step S10. By comparing the zero-bias outputs of the four power arms' IMUs under static conditions, if the root mean square of the zero-bias values ​​of the X, Y, and Z axes between any two units is greater than a preset zero-bias difference threshold, then the IMU is marked as having installation stress or device drift anomalies. Specifically, after completing the initialization of all IMUs, the main control system collects static acceleration and angular velocity data at 1000 consecutive sampling points (sampling frequency of 1000Hz).

[0086] Step S12: Calculate the mean zero offset of each inertial measurement unit in the three-axis directions based on the acceleration data and the angular velocity data; In practice, the mean zero-offset value of each axis of each inertial measurement unit can be calculated based on acceleration and angular velocity data. and The mean zero bias values ​​of each inertial measurement unit in the three-axis directions are obtained.

[0087] Step S13: Calculate the root mean square value between any two inertial measurement units based on the mean zero bias values ​​in the three-axis directions; It should be noted that the root mean square value between any two inertial measurement units (IMUs) can be calculated based on the mean zero bias values ​​in the three axes. and The root mean square value of the zero-bias difference is calculated as follows:

[0088] In the above formula, This is the root mean square value.

[0089] Step S14: Compare the root mean square value with a preset zero bias difference threshold; It should be noted that the preset zero bias difference threshold can be set in advance, for example, different preset zero bias difference thresholds can be set according to acceleration and angular velocity.

[0090] Step S15: When the root mean square value is greater than the preset zero bias difference threshold, a verification result is generated indicating that the inertial measurement unit has installation stress or abnormal device drift. In practical implementation, if the root mean square value is greater than the preset zero bias difference threshold, for example... , (Regarding acceleration) or (For angular velocity), the verification result indicates that the inertial measurement unit has abnormal installation stress or device drift. If the root mean square value is less than or equal to the preset zero bias difference threshold, it means that the verification result indicates that the inertial measurement unit does not have abnormal installation stress or device drift.

[0091] Step S16: Send the verification results to the user for re-inspection until the inertial measurement unit is free from installation stress or device drift abnormalities, and perform the step of applying a preset amplitude excitation signal to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification, and obtain the pre-flight reference data corresponding to each power arm.

[0092] Understandably, if the verification result shows that the inertial measurement unit has a consistency verification anomaly, this result will be sent to the user for manual re-inspection before takeoff until the inertial measurement unit no longer has a consistency verification anomaly, and then the pre-flight power link verification will be performed.

[0093] In this embodiment, within a preset time period after the aircraft is powered on, power-on enable signals are sent in parallel to the power management units of the four power arms. These signals enable the power management units to sequentially activate the power rails of the corresponding motor drivers, three-phase current sensors, speed encoders, inertial measurement units, and flight control communication links. A hardware handshake protocol is initiated, sending read commands to each motor driver, heartbeat handshake commands to the three-phase current sensors, speed encoders, and flight control communication links, and a self-test command to the inertial measurement unit for integrity checks. The system also receives power supply status data from the motor drivers, communication link response signals from the three-phase current sensors, speed encoders, and flight control communication links, and basic function response flags from the inertial measurement unit. The self-test result is obtained based on the power supply status data, the communication link response signals, and the basic function response flags. Through these steps, this embodiment significantly improves the detection efficiency and fault location accuracy of rotorcraft, meeting the stringent reliability requirements of manned application scenarios. This technical solution can not only complete the full aircraft health status assessment in a very short time after the aircraft is powered on, but also achieve millisecond-level response and precise fault isolation during flight, providing strong support for fault-tolerant control or emergency landing strategies.

[0094] Based on the first embodiment of this application, in the third embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 3 Step S60 includes steps S601 to S603: Step S601: The five-dimensional residual vector of the target power arm is reduced in dimension by principal component analysis to obtain the dimension-reduced residual vector.

[0095] In one feasible implementation, the five-dimensional residual vector of the target power arm is reduced in dimensionality by principal component analysis, and the principal components whose cumulative contribution rate meets the preset contribution rate threshold are retained to obtain the dimensionality-reduced residual vector.

[0096] In one feasible implementation, the main control system updates the comprehensive health score of each power arm according to a preset update cycle. When the comprehensive health score of any power arm is lower than a preset health score threshold, the system identifies that power arm as the target power arm and extracts the current five-dimensional residual vector corresponding to the target power arm. Subsequently, the system performs dimensionality reduction processing on the current five-dimensional residual vector using principal component analysis, and retains the principal components whose cumulative contribution rate meets the preset contribution rate threshold, thus obtaining the dimensionality-reduced residual vector.

[0097] Step S602: Calculate the Mahalanobis distance between the dimension-reduced residual vector and the various fault prototype vectors in the pre-stored fault prototype library, and obtain the matching score set based on the Mahalanobis distance through a preset distance-score conversion function.

[0098] It is understood that the pre-stored fault prototype library includes fault prototype vectors corresponding to multiple types of typical faults. These typical faults include one or more of the following: propeller damage, propeller loosening, motor bearing wear, motor winding abnormality, ESC response lag, and sensor drift. Each type of fault prototype vector is associated with a power arm number, upper / lower coaxial position identifier, and component type identifier.

[0099] The Mahalanobis distance is calculated as follows:

[0100] in and Let be the mean and covariance of the i-th type of fault prototype in the PCA subspace.

[0101] After calculating the matching scores for each type of fault, the main control system selects the fault type with the highest matching score from the set of matching scores as a candidate fault type. When the highest matching score is greater than a preset matching threshold, and the difference between the highest matching score and the second-highest matching score is greater than a preset confidence threshold, the candidate fault type is locked as the final fault type. The fault detection process is completed within a preset detection time to meet the real-time fault isolation requirements of the aircraft.

[0102] Step S603: Select the fault type corresponding to the maximum matching score from the matching score set as the fault detection result of the target power arm, and lock the fault detection result as the final fault type when the maximum matching score is greater than the preset matching threshold and the difference between the maximum matching score and the second largest matching score is greater than the preset confidence threshold.

[0103] In this embodiment, the fault type corresponding to the highest matching score can be selected from the matching score set as the fault detection result of the target power arm. When the difference between the highest matching score and the second highest matching score is greater than a preset confidence threshold, the system will lock the fault detection result as the final fault type and trigger the corresponding fault handling mechanism. For example, if the matching score of the "propeller damage" category is detected as the highest and meets the confidence condition, it is determined that the power arm has a propeller damage fault, and an emergency plan is immediately activated, such as limiting the maximum output power of the power arm or adjusting the load distribution of other power arms to maintain the overall balance of the aircraft.

[0104] In one feasible implementation, if the final fault type is determined, fault location and handling can be performed. Therefore, after step S603, the method further includes: parsing the specific component identifier where the fault occurred based on the final fault type, the target power arm, and the upper / lower coaxial position identifier and component type identifier associated with the fault prototype vector; querying a preset control allocation matrix adjustment table based on the specific component identifier to obtain the weight suppression parameter for the target power arm and the torque compensation parameter for the other normal power arms; reducing or cutting off the output command of the faulty motor in the target power arm based on the weight suppression parameter, and redistributing the motor speed command of the other normal power arms using the torque compensation parameter to maintain the attitude stability of the aircraft.

[0105] It should be noted that after identifying the specific component, the system automatically triggers the corresponding fault handling procedure. First, based on the parsed component identification, the system queries a preset control allocation matrix adjustment table to extract the weight suppression parameters related to the target power arm and the torque compensation parameters for the remaining normal power arms. These parameters are designed to minimize the impact of the fault on the overall performance of the aircraft by dynamically adjusting the output commands of each power arm. Next, based on the extracted weight suppression parameters, the system will proactively reduce or completely cut off the output capability of the faulty motor in the target power arm to prevent a more serious chain reaction caused by a single point of failure. At the same time, the system recalculates and allocates the motor speed commands of the remaining normal power arms using the torque compensation parameters to ensure that the aircraft can maintain attitude stability and trajectory controllability even when some power support is lost. In addition, to further improve the system's fault tolerance, while performing the above operations, the main control module will also monitor the overall state changes of the aircraft in real time and perform closed-loop feedback adjustment by combining multi-source data such as the inertial measurement unit, three-phase current sensor, and speed encoder. This multi-layered redundancy design not only effectively mitigates the impact of sudden failures, but also enables millisecond-level response in complex environments, providing reliable assurance for possible emergency landing or return-to-base strategies.

[0106] It is worth noting that the entire fault handling process strictly follows a priority principle. That is, when multiple power arms malfunction simultaneously, the system will prioritize handling the fault type with the higher threat level according to preset rules. For example, if a power arm is detected to have a propeller damage fault, while another power arm has a motor bearing wear problem, the system will prioritize handling the propeller damage fault because it poses a greater immediate threat to the aircraft's safety.

[0107] In one feasible implementation, the main control system also writes the fault detection results into the flight control log. The flight control log includes one or more of the following: fault occurrence time, target power arm identifier, specific component identifier, final fault type, overall health score, and fault detection confidence level. After the flight, the flight control log can be uploaded to the ground maintenance platform for maintenance recording, fault review, and health status tracking.

[0108] To illustrate the implementation effect of this invention, the following application scenario is constructed: A coaxial eight-propeller manned eVTOL is performing a test flight mission at a test site. After power-on, the main control system completes a parallel self-check within 1800 milliseconds, discovering that the IMU zero-bias root mean square value of power arm 2 exceeds the preset zero-bias difference threshold, marking it as an abnormal installation stress and prompting a re-check. After passing the re-check, the system enters the pre-flight power link verification stage. The system applies low-amplitude excitation signals to the upper and lower coaxial motor assemblies of each power arm, collects current, speed, vibration, and flight control communication link delay data, and establishes pre-flight baseline data. During flight, the system monitors that the comprehensive health score of power arm 4 drops below the preset health score threshold, and determines that the lower propeller is loose based on the five-dimensional residual vector of the power arm, with the confidence level meeting the preset requirements. Accordingly, the system reduces the output command of the lower motor in power arm 4 and increases the compensation output of the diagonal power arm and adjacent power arms according to the control allocation matrix adjustment table to maintain the attitude stability of the aircraft. After the flight, a complete self-check and fault log are uploaded to the ground operation and maintenance platform for subsequent maintenance and health record updates.

[0109] This embodiment uses principal component analysis to reduce the dimensionality of the five-dimensional residual vector of the target power arm, and determines the fault type based on the matching relationship between the dimensionality-reduced residual vector and various fault prototype vectors in a pre-stored fault prototype library. Specifically, the system calculates the Mahalanobis distance between the dimensionality-reduced residual vector and various fault prototype vectors, and obtains a set of matching scores based on the Mahalanobis distance using a preset distance-score transformation function. When the maximum matching score meets a preset matching threshold requirement, and the difference between the maximum matching score and the second largest matching score meets a preset confidence threshold requirement, the final fault type is locked. Further, the system determines the specific component identifier based on the final fault type, the target power arm, and the upper / lower coaxial position identifier and component type identifier associated with the fault prototype vector, and adjusts the output commands of the faulty motor and the remaining normal power arms based on the control allocation matrix adjustment table to maintain the stability of the aircraft attitude.

[0110] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the detection method of the coaxial eight-rotor manned multi-rotor aircraft of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0111] This application also provides a testing device for a coaxial eight-rotor manned multi-rotor aircraft. Please refer to [reference needed]. Figure 4 The detection device for the coaxial eight-rotor manned multi-rotor aircraft includes: The detection module 10 is used to perform parallel integrity checks on the motor drivers, three-phase current sensors, speed encoders, inertial measurement units, and flight control communication links of the four power arms within a preset time period after the aircraft is powered on, and obtain self-test results. Each of the four power arms is equipped with upper and lower coaxial motor assemblies to form a coaxial eight-propeller configuration.

[0112] The verification module 20 is used to apply a preset amplitude excitation signal to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification after the self-test result is that the self-test is passed, so as to obtain the pre-flight reference data corresponding to each power arm.

[0113] The construction module 30 is used to construct a five-dimensional residual vector based on the difference between the currently collected operational feature data and the pre-flight baseline data, and to determine the comprehensive health score corresponding to each power arm based on the five-dimensional residual vector.

[0114] The update module 40 is used to update the comprehensive health score during the real-time flight of the aircraft to obtain an updated comprehensive health score.

[0115] The determination module 50 is used to determine the five-dimensional residual vector of the target power arm corresponding to the updated comprehensive health score being less than the preset health score threshold when the updated comprehensive health score is less than the preset health score threshold.

[0116] The detection module 10 is also used for fault detection based on the five-dimensional residual vector of the target power arm.

[0117] The detection device for a coaxial eight-rotor manned multirotor aircraft provided in this application adopts the detection method for coaxial eight-rotor manned multirotor aircraft in the above embodiments, which can solve the technical problems of low detection efficiency and insufficient fault location accuracy in traditional methods. Compared with the prior art, the beneficial effects of the detection device for a coaxial eight-rotor manned multirotor aircraft provided in this application are the same as the beneficial effects of the detection method for a coaxial eight-rotor manned multirotor aircraft provided in the above embodiments, and other technical features in the detection device for a coaxial eight-rotor manned multirotor aircraft are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0118] In one embodiment, the detection module 10 is further configured to send power-on enable signals to the power management units of the four power arms in parallel within a preset time period after the aircraft is powered on, so that the power management units sequentially activate the power supply rails of the corresponding motor drivers, three-phase current sensors, speed encoders, inertial measurement units and flight control communication links based on the power-on enable signals. The hardware handshake protocol is initiated, and read commands are sent to each motor driver, heartbeat handshake commands are sent to the three-phase current sensor, speed encoder and flight control communication link, and self-test commands are sent to the inertial measurement unit to perform integrity detection. The system also receives power supply status data fed back by the motor driver, communication link response signals fed back by the three-phase current sensor, speed encoder and flight control communication link, and basic function response flags fed back by the inertial measurement unit. The self-test result is obtained based on the power supply status data, the communication link response signal, and the basic function response flag.

[0119] In one embodiment, the detection module 10 is further configured to obtain the voltage monitoring register value of the motor driver based on the power supply status data; If the voltage monitoring register value is within a preset safe range, the first detection result is determined to be that the power supply is normal. If the communication link response signal is received within the preset time window, the second detection result is determined to be that the communication is normal; If the basic function response flag is set to a preset value, the third detection result is determined to be that the inertial measurement unit has no calibration deviation or hardware failure. The self-test result is obtained based on the first test result, the second test result, and the third test result.

[0120] In one embodiment, the detection module 10 is further configured to collect acceleration and angular velocity data from multiple consecutive sampling points of the inertial measurement units corresponding to the four power arms under static conditions; The mean zero-bias ratio of each inertial measurement unit in the three-axis directions is calculated based on the acceleration data and the angular velocity data. Calculate the root mean square value between any two inertial measurement units based on the mean zero bias values ​​in the three-axis directions; The root mean square value is compared with a preset zero bias difference threshold. When the root mean square value is greater than the preset zero bias difference threshold, a verification result is generated indicating that the inertial measurement unit has installation stress or abnormal device drift. The verification results are sent to the user for re-inspection until the inertial measurement unit is free from installation stress or device drift abnormalities. Then, a pre-flight power link verification is performed by applying a preset amplitude excitation signal to the upper and lower coaxial motor assemblies of each power arm to obtain the pre-flight reference data corresponding to each power arm.

[0121] In one embodiment, the verification module 20 is further configured to send pulse width modulation excitation signals with preset amplitude and preset frequency and preset duty cycle to the upper and lower motors of each power arm in sequence when the ground is stationary. The instantaneous values ​​of the three-phase current, motor speed feedback, fuselage vibration acceleration, and flight control communication link delay data of each power arm under the action of the pulse width modulation excitation signal are collected simultaneously. The instantaneous values ​​of the three-phase current are filtered to calculate the mean current and total harmonic distortion rate; the motor speed feedback value is smoothed to extract the steady-state speed deviation; the fuselage vibration acceleration value is subjected to fast Fourier transform to calculate the proportion of high-frequency vibration energy, and the standard deviation of flight control communication link delay jitter is calculated based on the flight control communication link delay data; The mean current, total harmonic distortion rate, steady-state speed deviation, high-frequency vibration energy ratio, and standard deviation of flight control communication link delay jitter are used as five-dimensional initial feature data, and the five-dimensional initial feature data are saved as pre-flight reference data for each power arm.

[0122] In one embodiment, the detection module 10 is also used to collect in real time the average current, harmonic distortion rate, rotational speed deviation, high-frequency vibration energy ratio, and standard deviation of delay jitter of each power arm at the current moment. The differences between the real-time average current, the real-time harmonic distortion rate, the real-time speed deviation, the real-time high-frequency vibration energy ratio, and the standard deviation of the real-time flight control communication link delay jitter and the corresponding items in the pre-flight reference data are calculated respectively to obtain the first residual component, the second residual component, the third residual component, the fourth residual component, and the fifth residual component. The first residual component to the fifth residual component are combined in sequence to construct the five-dimensional residual vector at the current time. After standardizing the five-dimensional residual vector at the current moment, calculate the probability density value of belonging to the healthy state; Based on the probability density value, a comprehensive health score corresponding to each power arm is obtained by mapping it through a preset scoring mapping function.

[0123] In one embodiment, the detection module 10 is further configured to perform dimensionality reduction processing on the five-dimensional residual vector of the target power arm using principal component analysis to obtain the dimensionality-reduced residual vector; The Mahalanobis distance between the dimension-reduced residual vector and various fault prototype vectors in the pre-stored fault prototype library is calculated, and a matching score set is obtained based on the Mahalanobis distance through a preset distance-score conversion function. The pre-stored fault prototype library stores at least one or more fault types, such as propeller damage, motor bearing wear, ESC response lag, sensor drift, motor winding abnormality, and propeller loosening, and the fault prototype vector is associated with the power arm number, upper / lower coaxial position identifier, and component type identifier. The fault type corresponding to the maximum matching score is selected from the matching score set as the fault detection result of the target boom. When the maximum matching score is greater than the preset matching threshold and the difference between the maximum matching score and the second largest matching score is greater than the preset confidence threshold, the fault detection result is locked as the final fault type.

[0124] In one embodiment, the detection module 10 is further configured to parse the specific component identifier where the fault occurred based on the final fault type, the target power arm, and the upper / lower coaxial position identifier and component type identifier associated with the fault prototype vector; Based on the specific component identifier, query the preset control allocation matrix adjustment table to obtain the weight suppression parameters for the target power arm and the torque compensation parameters for the other normal power arms; The output commands of the faulty motor in the target power arm are reduced or cut off based on the weight suppression parameter, and the motor speed commands of the remaining normal power arms are redistributed using the torque compensation parameter to maintain the attitude stability of the aircraft.

[0125] This application provides a testing device for a coaxial eight-rotor manned multi-rotor aircraft. The testing device for the coaxial eight-rotor manned multi-rotor aircraft includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the testing method for the coaxial eight-rotor manned multi-rotor aircraft in the first embodiment described above.

[0126] The following is for reference. Figure 5 The diagram illustrates a structural schematic of a testing device suitable for implementing the embodiments of this application for a coaxial eight-rotor manned multi-rotor aircraft. The testing device in this application embodiment can be a flight controller, health management controller, onboard computing unit mounted on the aircraft, or a ground-based testing device communicatively connected to the flight controller. Figure 5 The testing equipment for the coaxial eight-rotor manned multi-rotor aircraft shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0127] like Figure 5 As shown, the detection device for a coaxial octagonal manned multirotor aircraft may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to programs stored in ROM (Read Only Memory) 1002 or programs loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the detection device for the coaxial octagonal manned multirotor aircraft. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, LCDs (Liquid Crystal Displays), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the testing equipment of the coaxial octocopter manned multirotor aircraft to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows testing equipment for a coaxial octocopter manned multirotor aircraft with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems may be implemented alternatively.

[0128] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0129] The testing equipment for a coaxial eight-rotor manned multirotor aircraft provided in this application adopts the testing method for coaxial eight-rotor manned multirotor aircraft in the above embodiments, which can solve the technical problems of low testing efficiency and insufficient fault location accuracy in traditional methods. Compared with the prior art, the beneficial effects of the testing equipment for a coaxial eight-rotor manned multirotor aircraft provided in this application are the same as the beneficial effects of the testing method for a coaxial eight-rotor manned multirotor aircraft provided in the above embodiments, and other technical features in the testing equipment for a coaxial eight-rotor manned multirotor aircraft are the same as those disclosed in the method of the previous embodiment, and will not be repeated here.

[0130] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0131] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0132] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the detection method for the coaxial eight-rotor manned multi-rotor aircraft in the above embodiments.

[0133] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable Read Only Memory or Flash Memory), optical fibers, CD-ROM (CD-Read Only Memory), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0134] The aforementioned computer-readable storage medium may be included in the testing equipment for a coaxial octagonal manned multirotor aircraft; or it may exist independently and not be installed in the testing equipment for a coaxial octagonal manned multirotor aircraft.

[0135] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by the testing equipment of the coaxial eight-rotor manned multi-rotor aircraft, the testing equipment performs the following actions: within a preset time period after the aircraft is powered on, it concurrently initiates integrity checks on the motor drivers, three-phase current sensors, speed encoders, inertial measurement units, and flight control communication links of the four power arms, obtaining self-test results; after the self-test results indicate that the self-test has passed, it applies preset amplitude excitation signals to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification, obtaining the results of each... The system generates pre-flight baseline data for each powered arm; constructs a five-dimensional residual vector based on the difference between the currently collected operational characteristic data and the pre-flight baseline data, and determines the comprehensive health score for each powered arm based on the five-dimensional residual vector; updates the comprehensive health score during the real-time flight of the aircraft to obtain an updated comprehensive health score; when the updated comprehensive health score is less than a preset health score threshold, determines the five-dimensional residual vector of the target powered arm whose updated comprehensive health score is less than the preset health score threshold; and performs fault detection based on the five-dimensional residual vector of the target powered arm.

[0136] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including LAN (Local Area Network) or WAN (Wide Area Network)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0137] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0138] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0139] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the detection method for the aforementioned coaxial eight-rotor manned multi-rotor aircraft. This solves the technical problems of low detection efficiency and insufficient fault location accuracy in traditional methods. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the detection method for the coaxial eight-rotor manned multi-rotor aircraft provided in the above embodiments, and will not be elaborated upon here.

[0140] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the detection method for a coaxial eight-rotor manned multi-rotor aircraft as described above.

[0141] The computer program product provided in this application can solve the technical problems of low detection efficiency and insufficient fault location accuracy in traditional methods. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the detection method for a coaxial eight-rotor manned multi-rotor aircraft provided in the above embodiments, and will not be repeated here.

[0142] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for detecting a coaxial eight-rotor manned multi-rotor aircraft, characterized in that, The detection method for the coaxial eight-rotor manned multi-rotor aircraft includes: Within a preset time period after the aircraft is powered on, the integrity of the upper motor driver, lower motor driver, three-phase current sensor, speed encoder, inertial measurement unit and flight control communication link corresponding to each of the four power arms is started in parallel to obtain the self-test results; wherein, each of the four power arms is equipped with upper and lower coaxial motor assemblies to form a coaxial eight-propeller configuration; After the self-test result is that the self-test is passed, a preset amplitude excitation signal is applied to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification and obtain the pre-flight reference data corresponding to each power arm. A five-dimensional residual vector is constructed based on the difference between the currently collected operational characteristic data and the pre-flight baseline data, and the comprehensive health score corresponding to each power arm is determined based on the five-dimensional residual vector. During the real-time flight of the aircraft, the comprehensive health score is updated to obtain an updated comprehensive health score; When the updated comprehensive health score is less than a preset health score threshold, the five-dimensional residual vector of the target power arm corresponding to the updated comprehensive health score being less than the preset health score threshold is determined; Fault detection is performed based on the five-dimensional residual vector of the target power arm.

2. The method as described in claim 1, characterized in that, The steps for performing parallel integrity checks on the motor drivers, three-phase current sensors, speed encoders, inertial measurement units, and flight control communication links of the four power arms within a preset time period after the aircraft is powered on, and obtaining self-test results, include: Within a preset time period after the aircraft is powered on, power-on enable signals are sent in parallel to the power management units of the four power arms, so that the power management units sequentially activate the power supply rails of the corresponding motor drivers, three-phase current sensors, speed encoders, inertial measurement units, and flight control communication links based on the power-on enable signals. The hardware handshake protocol is initiated, and read commands are sent to each motor driver, heartbeat handshake commands are sent to the three-phase current sensor, speed encoder and flight control communication link, and self-test commands are sent to the inertial measurement unit to perform integrity detection. The system also receives power supply status data fed back by the motor driver, communication link response signals fed back by the three-phase current sensor, speed encoder and flight control communication link, and basic function response flags fed back by the inertial measurement unit. The self-test result is obtained based on the power supply status data, the communication link response signal, and the basic function response flag.

3. The method as described in claim 2, characterized in that, The step of obtaining the self-test result based on the power supply status data, the communication link response signal, and the basic function response flag includes: The voltage monitoring register value of the motor driver is obtained based on the power supply status data; If the voltage monitoring register value is within a preset safe range, the first detection result is determined to be that the power supply is normal. If the communication link response signal is received within the preset time window, the second detection result is determined to be that the communication is normal; If the basic function response flag is set to a preset value, the third detection result is determined to be that the inertial measurement unit has no calibration deviation or hardware failure. The self-test result is obtained based on the first test result, the second test result, and the third test result.

4. The method as described in claim 1, characterized in that, The method further includes: Acceleration and angular velocity data are collected from multiple consecutive sampling points of the inertial measurement units corresponding to the four power arms under static conditions. The mean zero-bias ratio of each inertial measurement unit in the three-axis directions is calculated based on the acceleration data and the angular velocity data. Calculate the root mean square value between any two inertial measurement units based on the mean zero bias values ​​in the three-axis directions; The root mean square value is compared with a preset zero bias difference threshold. When the root mean square value is greater than the preset zero bias difference threshold, a verification result is generated indicating that the inertial measurement unit has installation stress or abnormal device drift. The verification results are sent to the user for re-inspection until the inertial measurement unit is free from installation stress or device drift abnormalities. Then, a pre-flight power link verification is performed by applying a preset amplitude excitation signal to the upper and lower coaxial motor assemblies of each power arm to obtain the pre-flight reference data corresponding to each power arm.

5. The method as described in claim 1, characterized in that, The step of applying a preset amplitude excitation signal to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification after the self-test result is a pass, and obtaining the pre-flight reference data corresponding to each power arm, includes: When the ground is stationary, pulse width modulation excitation signals with preset amplitude and preset frequency and preset duty cycle are sent sequentially to the upper and lower motors of each power arm. The instantaneous values ​​of the three-phase current, motor speed feedback, fuselage vibration acceleration, and flight control communication link delay data of each power arm under the action of the pulse width modulation excitation signal are collected simultaneously. The instantaneous values ​​of the three-phase current are filtered to calculate the mean current and total harmonic distortion rate; the motor speed feedback value is smoothed to extract the steady-state speed deviation; the fuselage vibration acceleration value is subjected to fast Fourier transform to calculate the proportion of high-frequency vibration energy, and the standard deviation of flight control communication link delay jitter is calculated based on the flight control communication link delay data; The mean current, total harmonic distortion rate, steady-state speed deviation, high-frequency vibration energy ratio, and standard deviation of flight control communication link delay jitter are used as five-dimensional initial feature data, and the five-dimensional initial feature data are saved as pre-flight reference data for each power arm.

6. The method as described in claim 1, characterized in that, The step of constructing a five-dimensional residual vector based on the difference between the currently collected operational characteristic data and the pre-flight baseline data, and determining the comprehensive health score corresponding to each power arm based on the five-dimensional residual vector, includes: Real-time acquisition of the mean current, harmonic distortion rate, speed deviation, high-frequency vibration energy ratio, and standard deviation of delay jitter of each power arm at the current moment. The differences between the real-time average current, the real-time harmonic distortion rate, the real-time speed deviation, the real-time high-frequency vibration energy ratio, and the standard deviation of the real-time flight control communication link delay jitter and the corresponding items in the pre-flight reference data are calculated respectively to obtain the first residual component, the second residual component, the third residual component, the fourth residual component, and the fifth residual component. The first residual component to the fifth residual component are combined in sequence to construct the five-dimensional residual vector at the current time. After standardizing the five-dimensional residual vector at the current moment, calculate the probability density value of belonging to the healthy state; Based on the probability density value, a comprehensive health score corresponding to each power arm is obtained by mapping it through a preset scoring mapping function.

7. The method as described in claim 1, characterized in that, The steps for fault detection based on the five-dimensional residual vector of the target power arm include: The five-dimensional residual vector of the target power arm is reduced in dimension by principal component analysis to obtain the dimension-reduced residual vector. The Mahalanobis distance between the dimension-reduced residual vector and various fault prototype vectors in the pre-stored fault prototype library is calculated, and a matching score set is obtained based on the Mahalanobis distance through a preset distance-score conversion function. The pre-stored fault prototype library stores at least one or more fault types, such as propeller damage, motor bearing wear, ESC response lag, sensor drift, motor winding abnormality, and propeller loosening, and the fault prototype vector is associated with the power arm number, upper / lower coaxial position identifier, and component type identifier. The fault type corresponding to the maximum matching score is selected from the matching score set as the fault detection result of the target boom. When the maximum matching score is greater than the preset matching threshold and the difference between the maximum matching score and the second largest matching score is greater than the preset confidence threshold, the fault detection result is locked as the final fault type.

8. The method as described in claim 7, characterized in that, The method further includes: Based on the final fault type, the target power arm, and the upper / lower coaxial position identifier and component type identifier associated with the fault prototype vector, the specific component identifier where the fault occurred is parsed out; Based on the specific component identifier, query the preset control allocation matrix adjustment table to obtain the weight suppression parameters for the target power arm and the torque compensation parameters for the other normal power arms; The output commands of the faulty motor in the target power arm are reduced or cut off based on the weight suppression parameter, and the motor speed commands of the remaining normal power arms are redistributed using the torque compensation parameter to maintain the attitude stability of the aircraft.

9. A testing device for a coaxial eight-rotor manned multi-rotor aircraft, characterized in that, The device includes: The detection module is used to perform parallel integrity checks on the motor drivers, three-phase current sensors, speed encoders, inertial measurement units, and flight control communication links of the four power arms within a preset time period after the aircraft is powered on, and obtain self-test results. The verification module is used to apply a preset amplitude excitation signal to the upper and lower coaxial motor assemblies of each power arm to perform pre-flight power link verification after the self-test result is that the self-test is passed, so as to obtain the pre-flight reference data corresponding to each power arm. The construction module is used to construct a five-dimensional residual vector based on the difference between the currently collected operational feature data and the pre-flight baseline data, and to determine the comprehensive health score corresponding to each power arm based on the five-dimensional residual vector. The update module is used to update the comprehensive health score during the real-time flight of the aircraft to obtain an updated comprehensive health score. The determination module is used to determine the five-dimensional residual vector of the target power arm corresponding to the updated comprehensive health score being less than the preset health score threshold when the updated comprehensive health score is less than the preset health score threshold. The detection module is also used for fault detection based on the five-dimensional residual vector of the target power arm.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the detection method for a coaxial eight-rotor manned multi-rotor aircraft as described in any one of claims 1 to 8.

Images