IMU (Inertial Measurement Unit) sampling frequency self-adaption method

Abstract

The invention provides an IMU (Inertial Measurement Unit) sampling frequency self-adaption method. The IMU sampling frequency self-adaption method comprises the following steps: S1, data acquisition and spectrum analysis; s2, detecting aliasing frequency; s3, performing high-frequency backstepping; s4, sampling frequency design; s5, adjusting the sampling frequency; and S6, verifying and optimizing. By dynamically adjusting the sampling frequency, high-frequency signals can be effectively prevented from being aliasing to a low-frequency region, so that the accuracy of low-frequency signals is ensured. The system can automatically adjust the sampling frequency according to the actual characteristics of signals so as to adapt to variable flight environments and vibration conditions. Through the optimization of the sampling frequency, the quality of inertial measurement unit (IMU) data is improved, and more reliable data support is provided for the control and navigation of the unmanned aerial vehicle. In addition, the maximum limit of the sampling frequency is introduced, so that the method is compatible with the existing IMU hardware, and the wide application potential of the method is shown.

Description

Technical Field

[0001] The present invention relates to the technical field of human-machine inertial measurement unit (IMU) data processing, and specifically provides an IMU sampling frequency adaptive method. Background Art

[0002] During the flight of an unmanned aerial vehicle (UAV), the inertial measurement unit (IMU) plays a crucial role, responsible for collecting key data such as the attitude, acceleration, and angular velocity of the aircraft. This data is indispensable for the navigation and control of the UAV. The data provided by the IMU not only covers low-frequency signals (such as the attitude change of the aircraft), but may also include high-frequency signals (such as the vibration of the engine and rotor). Due to the limitation of the sampling frequency, high-frequency signals may be mapped to the low-frequency region through frequency aliasing, thus interfering with the analysis of low-frequency signals.

[0003] Traditional IMU data processing methods often use a fixed sampling frequency, which is largely unable to effectively avoid the aliasing problem of high-frequency signals. To solve this problem, researchers have proposed adaptive sampling techniques based on specific algorithms. For example, one method is to analyze the spectral characteristics of IMU data and dynamically adjust the sampling frequency to avoid potential aliasing regions. However, this method requires complex spectral analysis and real-time processing capabilities, has high requirements for hardware and software, and may not be able to respond in a timely manner in a dynamically changing flight environment.

[0004] Another method is to use multi-rate sampling technology, that is, to apply different sampling frequencies in different time periods. This method can reduce aliasing to a certain extent, but also faces problems such as high hardware requirements and large algorithm complexity. In addition, multi-rate sampling technology may introduce additional errors and discontinuities when switching sampling frequencies, affecting the accuracy and continuity of the data. Summary of the Invention

[0005] The purpose of the present invention is to provide an IMU sampling frequency adaptive method to solve the problems in the prior art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions: The present invention provides an IMU sampling frequency adaptive method, including the following steps:

[0007] S1. Data acquisition and spectral analysis:

[0008] Use two different sampling frequencies f1 and f2 to collect the original data of the IMU, and perform spectral analysis on the data to determine the main frequency components of the signal;

[0009] S2. Aliasing frequency detection:

[0010] Based on the spectrum analysis results, detect whether there is an aliasing phenomenon and determine which frequencies are aliased frequencies;

[0011] S3. High-frequency frequency back-calculation:

[0012] The high-frequency frequency f original is back-calculated through the formula f aliased = |n·f s - f original |;

[0013] S4. Sampling frequency design:

[0014] The sampling frequency design includes calculating the sampling frequency range and selecting an appropriate sampling frequency f s , satisfying the formula

[0015] S5. Sampling frequency adjustment:

[0016] According to the design results, dynamically adjust the sampling frequency of the IMU and re-collect data;

[0017] S6. Verification and optimization:

[0018] Perform spectrum analysis on the adjusted data to verify whether the aliased frequencies have moved out of the low-frequency range. If the expected effect is not achieved, further optimize the sampling frequency.

[0019] Preferably, in S3, by traversing possible n values, select the n that makes the back-calculated high-frequency frequency f original closest to the actual value.

[0020] Preferably, in S4, add a maximum limit to the sampling frequency to ensure that the sampling frequency does not exceed the maximum allowable value f max of the hardware or application.

[0021] Preferably, in S6, it includes performing spectrum analysis on the adjusted data and further optimizing the sampling frequency until the aliased frequencies have moved out of the low-frequency range.

[0022] Preferably, in S3, f s is the sampling frequency, f aliased is the aliased frequency, and n is an integer and is traversed to make the back-calculated high-frequency frequency closest to the actual value.

[0023] Preferably, in S4, is the back-calculated high-frequency frequency, f low is the upper limit of the set low-frequency range, and f max is the maximum sampling frequency.

[0024] Preferably, in S5, the sampling frequency adjustment includes dynamically adjusting the sampling frequency of the IMU.

[0025] The present invention has at least the following beneficial effects:

[0026] An IMU sampling frequency adaptive method provided by the present invention can effectively prevent high-frequency signals from aliasing into the low-frequency region by dynamically adjusting the sampling frequency, thereby ensuring the accuracy of low-frequency signals. The system can automatically adjust the sampling frequency according to the actual characteristics of the signal to adapt to the changing flight environment and vibration conditions. By optimizing the sampling frequency, the quality of the data of the inertial measurement unit (IMU) is improved, providing more reliable data support for the control and navigation of the unmanned aerial vehicle. In addition, the introduction of the maximum limit of the sampling frequency ensures the compatibility of this method with the existing IMU hardware, demonstrating its broad application potential. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a flowchart provided by the present invention;

[0028] Figure 2 is a spectrum comparison diagram before and after the sampling frequency adjustment of the present invention;

[0029] Figure 3 is an example spectrum diagram. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as commonly understood by those skilled in the technical field to which this invention belongs. The terms used in this specification in the description of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used in this specification includes any and all combinations of one or more of the related listed items.

[0031] Embodiment

[0032] As Figures 1-3 shown, an IMU sampling frequency adaptive method includes the following steps:

[0033] S1. Data acquisition and spectrum analysis:

[0034] Use two different sampling frequencies f1 and f2 to acquire the original data of the IMU, and perform spectrum analysis on the data to determine the main frequency components of the signal;

[0035] Specifically, use the IMU to collect the attitude, acceleration, and angular velocity data of the unmanned aerial vehicle, and perform a fast Fourier transform (FFT) on the data to obtain a spectrum diagram.

[0036] S2. Aliasing frequency detection:

[0037] Detect whether there is an aliasing phenomenon according to the spectrum analysis result, and determine which frequencies are aliased frequencies;

[0038] Determine the aliased frequencies in the spectrogram

[0039] S3. High-frequency frequency back-calculation:

[0040] High-frequency frequency f original Back-calculation is achieved through the formula f aliased =|n·f s -f original |;

[0041] Specifically, according to the aliased frequencies obtained from the two sampling results and the sampling frequencies f1, f2, through the formula

[0042]

[0043] Back-calculate to obtain the estimated value of the original high-frequency frequency where n1, n2 are integer multiples closest to the original frequency f original ;

[0044] According to the aliased frequencies and the current sampling frequencies f1 = 500Hz, f2 = 1000Hz, by traversing the n values, back-calculate the original high-frequency frequency

[0045] S4. Sampling frequency design:

[0046] Sampling frequency design includes calculating the sampling frequency range and selecting an appropriate sampling frequency f s , satisfying the formula

[0047]

[0048] Specifically, according to the back-calculated high-frequency frequency and the set upper limit f of the low-frequency interval low , calculate the sampling frequency f s that satisfies the condition:

[0049]

[0050] Select an appropriate sampling frequency f s , ensure that the aliased frequency f aliased does not fall into the low-frequency interval [0, f low , and the sampling frequency f s is less than the maximum sampling frequency f max

[0051] According to the back-calculated high-frequency frequency and the upper limit f of the set low-frequency range low = 200 Hz, calculate the sampling frequency range, and select an appropriate sampling frequency f s = 700 Hz

[0052] S5. Sampling frequency adjustment:

[0053] According to the design results, dynamically adjust the sampling frequency of the IMU and re-collect the data;

[0054] S6. Verification and optimization:

[0055] Perform spectral analysis on the adjusted data to verify whether the aliasing frequency has been moved out of the low-frequency range. If the expected effect is not achieved, further optimize the sampling frequency.

[0056] Among them, in S3, by traversing the possible n values, select the n that makes the deduced high-frequency frequency f original closest to the actual value.

[0057] Among them, in S4, add a maximum limit to the sampling frequency to ensure that the sampling frequency does not exceed the maximum allowable value f of the hardware or application max .

[0058] Among them, in S6, it includes performing spectral analysis on the adjusted data and further optimizing the sampling frequency until the aliasing frequency has been moved out of the low-frequency range.

[0059] Among them, in S3, f s is the sampling frequency, f aliased is the aliasing frequency, and n is an integer and is traversed to make the deduced high-frequency frequency closest to the actual value.

[0060] Among them, in S4, is the deduced high-frequency frequency, f low is the upper limit of the set low-frequency range, f max is the maximum sampling frequency.

[0061] Among them, in S5, the sampling frequency adjustment includes dynamically adjusting the sampling frequency of the IMU.

[0062] Install the IMU device on the drone to ensure that it can accurately collect the attitude, acceleration, and angular velocity data during flight. Start the drone and start recording the original data of the IMU. Collect data samples at two preset different sampling frequencies, such as 50 Hz and 100 Hz, respectively. Use the fast Fourier transform (FFT) to perform spectral analysis on the collected data to identify the main frequency components of the signal and their distribution.

[0063] Next, according to the spectrum analysis results, carefully check whether there is frequency aliasing. The specific operation is to observe whether there are high-frequency signal components in the low-frequency region of the spectrogram and record the specific values of these aliasing frequencies. For example, if at a sampling frequency of 50 Hz, an unexpected signal component appears at 30 Hz, then mark 30 Hz as the aliasing frequency.

[0064] Then, enter the high-frequency frequency back-calculation stage. Based on the known aliasing frequency and sampling frequency, perform back-calculation using the formula. Assume that aliasing frequencies of 30 Hz and 60 Hz are detected at sampling frequencies of 50 Hz and 100 Hz respectively. By traversing possible integer multiples, find the estimated value closest to the actual high-frequency frequency. For example, through calculation, it is found that the original high-frequency frequency may be 120 Hz.

[0065] Subsequently, conduct sampling frequency design. According to the back-calculated high-frequency frequency and the set upper limit of the low-frequency range, calculate the sampling frequency range that meets the conditions. Assume that the upper limit of the low-frequency range is 40 Hz. To ensure that the aliasing frequency does not fall into this range, select a new sampling frequency, such as 150 Hz, which not only meets the conditions but also does not exceed the maximum sampling frequency allowed by the hardware.

[0066] Next, according to the design result, dynamically adjust the sampling frequency of the IMU to 150 Hz and restart data acquisition. Use FFT for spectrum analysis again to verify whether the aliasing frequency has been successfully shifted out of the low-frequency range. If it is found that there is still aliasing, further optimize the sampling frequency and repeat the above steps until the expected effect is achieved.

[0067] Finally, apply the optimized sampling frequency and data processing method to the actual flight control of the drone, and adjust the sampling frequency of the IMU in real time to ensure the accuracy and reliability of the flight data. Through multiple flight tests, verify the effectiveness of this method under different flight environments and vibration conditions, and ensure its stability and robustness in practical applications.

[0068] An IMU sampling frequency adaptive method provided by the present invention can effectively prevent high-frequency signals from aliasing into the low-frequency region by dynamically adjusting the sampling frequency, thereby ensuring the accuracy of low-frequency signals. The system can automatically adjust the sampling frequency according to the actual characteristics of the signal to adapt to the changing flight environment and vibration conditions. Through the optimization of the sampling frequency, the quality of the data of the inertial measurement unit (IMU) is improved, providing more reliable data support for the control and navigation of the drone. In addition, introducing the maximum limit of the sampling frequency ensures the compatibility of this method with the existing IMU hardware, demonstrating its broad application potential.

[0069] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting them; under the idea of the present invention, the technical features in the above embodiments or different embodiments can also be combined, and the steps can be implemented in any order, and there are many other variations in different aspects of the present invention as above. For the sake of brevity, they are not provided in detail; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An IMU sampling frequency adaptive method, characterized in that Including the following steps: S1. Data acquisition and spectrum analysis: Use two different sampling frequencies f1 and f2 to acquire the raw data of the IMU, and perform spectrum analysis on the data to determine the main frequency components of the signal; S2. Aliasing frequency detection: According to the spectrum analysis results, detect whether there is an aliasing phenomenon and determine which frequencies are aliasing frequencies; S3. High-frequency frequency back-calculation: The high-frequency frequency f original is inversely derived through the formula f aliased = |n·f s - f original |; S4. Sampling frequency design: The sampling frequency design includes calculating the sampling frequency range and selecting an appropriate sampling frequency f s , satisfying the formula S5. Sampling frequency adjustment: According to the design results, dynamically adjust the sampling frequency of the IMU and re-acquire the data; S6. Verification and optimization: Perform spectrum analysis on the adjusted data to verify whether the aliasing frequency has moved out of the low-frequency range. If the expected effect is not achieved, further optimize the sampling frequency.

2. The IMU sampling frequency adaptive method according to claim 1, characterized in that In S3, by traversing possible n values, select the n that makes the deduced high-frequency original frequency f closest to the actual value.

3. An IMU sampling frequency adaptive method according to claim 1, characterized in that, In S4, add a maximum limit to the sampling frequency to ensure that the sampling frequency does not exceed the maximum allowable value f of the hardware or application max .

4. An IMU sampling frequency adaptive method according to claim 1, characterized in that In S6, it includes performing spectrum analysis on the adjusted data and further optimizing the sampling frequency until the aliasing frequency has moved out of the low-frequency range.

5. The IMU sampling frequency adaptive method according to claim 1, characterized in that In S3, f s is the sampling frequency, and f aliased is the aliasing frequency. n is an integer and is traversed so that the deduced high-frequency frequency is closest to the actual value.

6. The IMU sampling frequency adaptive method according to claim 1, characterized in that In the said S4, is the high-frequency frequency of reverse thrust, f low is the upper limit of the set low-frequency range, f max is the maximum sampling frequency.

7. A method for adaptively adjusting the sampling frequency of an IMU according to claim 1, characterized in that In S5, the sampling frequency adjustment includes dynamically adjusting the sampling frequency of the IMU.

Images