Braking state identification method, device, electronic device and storage medium

By pre-processing and characterizing the three-axis acceleration data, combined with the analysis of Z-axis, Y-axis and X-axis acceleration data, the noise problem of the bicycle taillight brake detection system when riding on slope sections is solved, achieving more accurate Brake status identification and reduce false braking misjudgment, reducing costs.

CN114964805BActive Publication Date: 2024-12-06QINGDAO MAGENE INTELLIGENCE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210441328.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-04-25
Publication Date
2024-12-06
Estimated Expiration
2042-04-25

AI Technical Summary

Technical Problem

The existing bicycle taillight brake detection system based on an accelerometer has system noise and frequency domain noise when riding on a slope section, resulting in inaccurate determination of the brake status and the inability to effectively distinguish pseudo-braking signals.

Method used

By obtaining three-axis acceleration data for preprocessing, the pseudo-brake state is judged using Z-axis and Y-axis acceleration data, and when it is not in the pseudo-brake state, the brake state point is judged based on the X-axis acceleration data, and the band pass baseline processing and valley are used. Value detection and other methods can improve the signal-to-noise ratio and reduce false braking misjudgment.

Benefits of technology

It improves the accuracy and robustness of brake status recognition, reduces costs, eliminates the static calibration process, and enhances the system's slope adaptability and noise immunity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114964805B_ABST
    Figure CN114964805B_ABST
Patent Text Reader

Abstract

The present invention discloses a brake state identification method, device, electronic device and storage medium, the method comprising: obtaining three-axis acceleration data and preprocessing the three-axis acceleration data; judging whether the current state is in a pseudo-brake state based on the preprocessed Z-axis acceleration data and Y-axis acceleration data, wherein the Y-axis represents the acceleration direction perpendicular to the riding direction, and the Z-axis represents the acceleration direction collinear with the gravity acceleration; judging the brake state point in the X-axis acceleration data based on the preprocessed X-axis acceleration data, and taking the timestamp corresponding to the brake state point as the brake moment, wherein the X-axis represents the acceleration direction collinear with the riding direction. The present invention solves the technical problem that the brake state of a bicycle cannot be accurately judged at present.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of riding equipment, and in particular to a brake state recognition method, device, electronic equipment and storage medium. Background Art

[0002] Traditional accelerometer-based bicycle taillight brake detection systems require static calibration before performing brake detection. Although this improves the adaptability of the taillight installation angle to a certain extent, it introduces system noise caused by gravity acceleration when riding on sloped roads. However, static calibration is not adaptive to slope. In addition, the accelerometer itself has a large amount of frequency domain noise, which affects brake detection. False brake signals can also cause brake identification to diverge.

[0003] Therefore, the existing brake state determination method cannot accurately determine the brake state of the bicycle. Summary of the Invention

[0004] The purpose of the present invention is to overcome the above technical deficiencies and provide a brake status recognition method, device, electronic device and storage medium to solve the technical problem that the existing technology cannot accurately judge the brake status of a bicycle.

[0005] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present invention provides a method for identifying a braking state, comprising the following steps:

[0007] Acquiring three-axis acceleration data and preprocessing the three-axis acceleration data;

[0008] Based on the pre-processed Z-axis acceleration data and Y-axis acceleration data, determine whether the current state is pseudo-braking, where the Y-axis represents the acceleration direction perpendicular to the riding direction, and the Z-axis represents the acceleration direction collinear with the acceleration due to gravity;

[0009] When the vehicle is not currently in a pseudo-braking state, a braking state point in the X-axis acceleration data is determined based on the preprocessed X-axis acceleration data, and a timestamp corresponding to the braking state point is used as the braking moment, where the X-axis represents an acceleration direction collinear with the riding direction.

[0010] Preferably, in the braking state identification method, the acquiring of three-axis acceleration data and preprocessing of the three-axis acceleration data include:

[0011] Acquire triaxial acceleration data, and perform bandpass baseline removal processing on the triaxial acceleration data to obtain triaxial acceleration data within a window.

[0012] Preferably, in the braking state identification method, the following formula is used to perform bandpass baseline removal processing on the three-axis acceleration data:

[0013]

[0014] Where Acc represents the input three-axis acceleration data, where the acceleration data of each axis consists of the accelerations at different times in n corresponding directions. y represents the output filtered value, a and b represent the denominator and numerator vectors of the corresponding Butterworth filter, the order of the bandpass filter is M, and n represents the length of the sequence.

[0015] Preferably, in the braking state identification method, judging whether the vehicle is currently in a pseudo-braking state based on the pre-processed Z-axis acceleration data and Y-axis acceleration data includes:

[0016] Based on the pre-processed Z-axis acceleration data, determine whether the current state is riding;

[0017] Based on the pre-processed Y-axis acceleration data, determine whether the vehicle is currently in a shaking state;

[0018] When the vehicle is currently in a riding state and not in a shaking state, it is determined that the vehicle is not currently in a pseudo-braking state; otherwise, it is determined that the vehicle is currently in a pseudo-braking state.

[0019] Preferably, in the braking state identification method, judging whether the vehicle is currently in a riding state based on the pre-processed Z-axis acceleration data includes:

[0020] The mean of the pre-processed Z-axis acceleration data is obtained, and after taking the absolute value, it is determined whether the absolute value is greater than a first threshold. If so, it is determined that the current state is riding; otherwise, it is determined that the current state is pseudo-riding.

[0021] Preferably, in the braking state identification method, judging whether the vehicle is currently in a shaking state based on the pre-processed Y-axis acceleration data includes:

[0022] Obtain the mean error of the preprocessed Y-axis acceleration data, and determine whether the mean error is greater than a second threshold value. If so, determine that the vehicle is currently in a shaking state; otherwise, determine that the vehicle is currently in a normal riding state.

[0023] Preferably, in the braking state identification method, when the vehicle is not currently in a pseudo-braking state, based on the pre-processed X-axis acceleration data, determining a braking state point in the X-axis acceleration data, and using a timestamp corresponding to the braking state point as the braking moment, the method includes:

[0024] A valley detection is performed on the preprocessed X-axis acceleration data. If a valley exists and is less than a third threshold, the valley point is determined to be a braking state point, and a timestamp corresponding to the braking state point is used as the braking moment. Otherwise, the next round of iteration is performed.

[0025] In a second aspect, the present invention further provides a brake state detection device, comprising:

[0026] A data processing module, used for acquiring three-axis acceleration data and preprocessing the three-axis acceleration data;

[0027] A pseudo-braking state detection module is used to determine whether the vehicle is currently in a pseudo-braking state based on pre-processed Z-axis acceleration data and Y-axis acceleration data, where the Y-axis represents the acceleration direction perpendicular to the riding direction and the Z-axis represents the acceleration direction collinear with the acceleration due to gravity;

[0028] The brake state detection module is used to determine the brake state point in the X-axis acceleration data based on the preprocessed X-axis acceleration data when the vehicle is not currently in a pseudo-brake state, and use the timestamp corresponding to the brake state point as the braking moment, wherein the X-axis represents the acceleration direction collinear with the riding direction.

[0029] In a third aspect, the present invention further provides an electronic device, comprising a processor and a memory;

[0030] The memory stores a computer-readable program executable by the processor;

[0031] When the processor executes the computer-readable program, the steps in the brake state detection method described above are implemented.

[0032] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the brake status detection method as described above.

[0033] Compared to the prior art, the brake state identification method, device, electronic device, and storage medium provided by the present invention first obtain three-axis acceleration data, then pre-process the three-axis acceleration data. Then, based on the acceleration data in the acceleration direction perpendicular to the riding direction and the acceleration data in the acceleration direction collinear with the gravitational acceleration, the method determines whether the current state is a pseudo-braking state. If not, the method determines the braking state point in the X-axis acceleration data based on the pre-processed X-axis acceleration data, thereby detecting the braking state. By characterizing the accelerometer data, the present invention can improve the signal-to-noise ratio of the feature data, thereby reducing data input and improving efficiency without reducing the accuracy of brake state identification. Furthermore, the static calibration process is eliminated. Furthermore, by performing pseudo-braking state identification, it can effectively reduce brake misjudgments caused by some pseudo-braking signals, making the brake determination more robust. Furthermore, the present invention does not require the connection of other hardware devices and only requires an accelerometer to identify the brake state, thus reducing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of a preferred embodiment of the brake status identification method provided by the present invention;

[0035] Figure 2 This is a comparison diagram of the X-axis acceleration data before and after preprocessing of the present invention;

[0036] Figure 3 Schematic diagram of Z-axis acceleration data and Y-axis acceleration data under various states of the present invention;

[0037] Figure 4 It is a schematic diagram of the X-axis acceleration data and the corresponding braking moment;

[0038] Figure 5 is a schematic diagram of an embodiment of a brake state detection device provided by the present invention;

[0039] Figure 6 2 is a schematic diagram of the operating environment of an embodiment of the brake status detection program of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0041] See also Figure 1 , an embodiment of the present invention provides a brake state identification method, comprising the following steps:

[0042] S100, acquiring three-axis acceleration data and preprocessing the three-axis acceleration data;

[0043] S200: Determine whether the vehicle is currently in a pseudo-braking state based on the pre-processed Z-axis acceleration data and the Y-axis acceleration data, where the Y-axis represents an acceleration direction perpendicular to the riding direction and the Z-axis represents an acceleration direction collinear with the acceleration due to gravity.

[0044] S300: When the vehicle is not currently in a pseudo-braking state, determine a braking state point in the preprocessed X-axis acceleration data, and use a timestamp corresponding to the braking state point as a braking moment, wherein the X-axis represents an acceleration direction collinear with the riding direction.

[0045] In this embodiment, three-axis acceleration data is first acquired and then pre-processed. A pseudo-braking state is then determined based on the acceleration data in the acceleration direction perpendicular to the riding direction and the acceleration data in the acceleration direction collinear with the gravitational acceleration. If the pseudo-braking state is not present, the braking state point in the pre-processed X-axis acceleration data is determined, thereby detecting the braking state. The present invention improves the signal-to-noise ratio of the feature data by characterizing the accelerometer data, thereby reducing data input and improving efficiency without compromising the accuracy of identifying the braking state. This eliminates the need for static calibration. Furthermore, by performing pseudo-braking state determination, false braking errors caused by some pseudo-braking signals can be effectively reduced, resulting in more robust braking determination. Furthermore, the present invention requires no additional hardware and only requires an accelerometer to identify the braking state, reducing costs.

[0046] In some embodiments, the conventional accelerometer-based bicycle taillight brake detection system needs to solve a conversion matrix based on the static three-axis acceleration data before performing brake detection, so that the acceleration output corresponding to the initial state of riding is [0, 0, g]. If a certain route has multiple sections with different slopes (the corresponding acceleration of gravity on each axis will be different), the data processed by the conversion matrix will cause the parking judgment to not converge. Therefore, after obtaining the three-axis acceleration data, the present invention pre-processes the three-axis acceleration data. Specifically, step S100 specifically includes:

[0047] Acquire triaxial acceleration data, and perform bandpass baseline removal processing on the triaxial acceleration data to obtain triaxial acceleration data within a window.

[0048] In this embodiment, the input data is de-baselined to achieve a de-biasing effect under various slope conditions. By performing bandpass sampling on the newly acquired acceleration data, the baseline error of the data sequence is eliminated (i.e., the baseline response of gravity acceleration to each axis under various slopes is adaptively eliminated), eliminating the static calibration process.

[0049] In some embodiments, the following formula is used to perform bandpass baseline removal processing on the triaxial acceleration data:

[0050]

[0051] Where Acc represents the input three-axis acceleration data, where the acceleration data of each axis consists of the accelerations at different times in n corresponding directions. y represents the output filtered value, a and b represent the denominator and numerator vectors of the corresponding Butterworth filter, the order of the bandpass filter is M, and n represents the length of the sequence.

[0052] In this embodiment, the accelerometer itself has high-frequency noise and power-frequency noise, which will interfere with or even cover the acceleration characteristics of the brake. To address this problem, the present invention designs a 0.01-0.5Hz bandpass filter, which can greatly eliminate high-frequency and power-frequency noise and eliminate the baseline of the acceleration sequence. In specific implementation, the batch acceleration data Acc is stored in a queue to form an acceleration sequence [Acc i , i=1~n], perform bandpass baseline removal on the acceleration, the acceleration sequence [Acc i , i=1~n] is brought into the bandpass filter to obtain the filtered acceleration data. Figure 2 As shown in FIG, it is a comparison diagram of the X-axis acceleration data before and after preprocessing.

[0053] Step S200 is to determine the pseudo-braking state. After filtering the acceleration data, the filtered three-axis acceleration data is extracted and the pseudo-braking state is determined based on the three-axis acceleration data. In some embodiments, step S200 specifically includes:

[0054] Based on the pre-processed Z-axis acceleration data, determine whether the current state is riding;

[0055] Based on the pre-processed Y-axis acceleration data, determine whether the vehicle is currently in a shaking state;

[0056] When the vehicle is currently in a riding state and not in a shaking state, it is determined that the vehicle is not currently in a pseudo-braking state; otherwise, it is determined that the vehicle is currently in a pseudo-braking state.

[0057] In this embodiment, the X axis represents the acceleration direction collinear with the riding direction, the Y axis represents the acceleration direction perpendicular to the riding direction, and the Z axis represents the acceleration direction collinear with the acceleration due to gravity.

[0058] Since the main false braking state comes from the riding state such as shaking the car and removing the taillight, and the Z-axis data and Y-axis data in each state are obviously different, such as Figure 3 As shown, the present invention determines the riding state based on the Z-axis acceleration data and the vehicle shaking state based on the Y-axis acceleration data. In other words, the embodiment of the present invention first determines whether the vehicle is in the riding state based on the accelerometer that is collinear with gravity; then determines whether the vehicle is in the shaking state based on the accelerometer that is parallel to the normal plane of the riding direction; and if the judgment conditions are met, the braking state is determined.

[0059] In some embodiments, determining whether the vehicle is currently in a riding state based on the pre-processed Z-axis acceleration data includes:

[0060] The mean of the pre-processed Z-axis acceleration data is obtained, and after taking the absolute value, it is determined whether the absolute value is greater than a first threshold. If so, it is determined that the current state is riding; otherwise, it is determined that the current state is pseudo-riding.

[0061] In this embodiment, the z-axis acceleration data in the window is averaged and the absolute value is taken, which is recorded as |Z_mean|. The threshold 1 is set to 0.8g. If |Z_mean|>0.8g, it is considered to be a riding state, otherwise it is considered to be a pseudo-riding state.

[0062] In some embodiments, determining whether the vehicle is currently in a shaking state based on the preprocessed Y-axis acceleration data includes:

[0063] Obtain the mean error of the preprocessed Y-axis acceleration data, and determine whether the mean error is greater than a second threshold value. If so, determine that the vehicle is currently in a shaking state; otherwise, determine that the vehicle is currently in a normal riding state.

[0064] In this embodiment, the data of the Y axis is extracted, and the mean error is solved, recorded as Y_std, and the threshold 2 is set to 0.1g. If Y_std>0.1g, it is considered that the state is a shaking state, otherwise it is normal riding.

[0065] When both levels of judgment are passed, it is considered that there is no false braking state at present, and the braking state can be identified and judged at this time.

[0066] In some embodiments, step S300 is to extract the braking state point. Specifically, step S300 includes:

[0067] A valley detection is performed on the preprocessed X-axis acceleration data. If a valley exists and is less than a third threshold, the valley point is determined to be a braking state point, and a timestamp corresponding to the braking state point is used as the braking moment. Otherwise, the next round of iteration is performed.

[0068] In this embodiment, the accelerometer data sequence that is collinear with the riding direction is used for braking, a sliding window is used to obtain batches of acceleration data, and the peak detection and mean error judgment method is used to identify whether the current state is a braking state. Specifically, after the pseudo-braking state judgment is completed, the braking state detection is performed, and the x-axis data in the window (after filtering) is subjected to valley value detection. If a valley value exists and the valley value is less than the threshold value 3 (-0.2g), it is considered that the point is a braking state point. Otherwise, the next round of iteration is skipped. Figure 4 As shown, it is a schematic diagram of X-axis acceleration data and the corresponding braking moment.

[0069] The present invention eliminates the baseline error of the data sequence by bandpass sampling the newly acquired acceleration data (i.e., adaptively eliminating the baseline response of gravity acceleration to each axis under various slopes), thus eliminating the static calibration process. In addition, the designed bandpass filter can not only eliminate the baseline of the acceleration data, but also eliminate some high-frequency and power-frequency noise based on the different characteristics of the acceleration frequency and noise frequency in the braking state. In addition, the addition of a pseudo-braking algorithm can effectively identify pseudo-braking signals in non-riding states, making the entire recognition process more robust. Moreover, the present invention does not require the access of other hardware devices, only an accelerometer, which saves costs.

[0070] Based on the above brake state detection method, the present invention also provides a brake state detection device 600. Figure 5 The braking state detection device includes a data processing module 610 , a pseudo-braking state detection module 620 and a braking state detection module 630 .

[0071] The data processing module 610 is used to obtain three-axis acceleration data and pre-process the three-axis acceleration data;

[0072] The pseudo-braking state detection module 620 is configured to determine whether the vehicle is currently in a pseudo-braking state based on the pre-processed Z-axis acceleration data and the Y-axis acceleration data, where the Y-axis represents the acceleration direction perpendicular to the riding direction and the Z-axis represents the acceleration direction collinear with the acceleration due to gravity.

[0073] The braking state detection module 630 is used to determine the braking state point in the X-axis acceleration data based on the preprocessed X-axis acceleration data when the vehicle is not currently in a pseudo-braking state, and use the timestamp corresponding to the braking state point as the braking moment, wherein the X-axis represents the acceleration direction collinear with the riding direction.

[0074] In this embodiment, three-axis acceleration data is first acquired and then pre-processed. A pseudo-braking state is then determined based on the acceleration data in the acceleration direction perpendicular to the riding direction and the acceleration data in the acceleration direction collinear with the gravitational acceleration. If the pseudo-braking state is not present, the braking state point in the pre-processed X-axis acceleration data is determined, thereby detecting the braking state. The present invention improves the signal-to-noise ratio of the feature data by characterizing the accelerometer data, thereby reducing data input and improving efficiency without compromising the accuracy of identifying the braking state. This eliminates the need for static calibration. Furthermore, by performing pseudo-braking state determination, false braking errors caused by some pseudo-braking signals can be effectively reduced, resulting in more robust braking determination. Furthermore, the present invention requires no additional hardware and only requires an accelerometer to identify the braking state, reducing costs.

[0075] Since the brake state detection method has been described in detail above, the technical effects of the brake state detection method are also possessed by the brake state detection device, so they will not be described again here.

[0076] like Figure 6 As shown, based on the above brake state detection method, the present invention also provides an electronic device, which can be a computing device such as a mobile terminal, a desktop computer, a notebook, a palmtop computer, and a server. The electronic device includes a processor 10, a memory 20, and a display 30. Figure 6 Only some of the components of the electronic device are shown, but it should be understood that it is not required to implement all of the shown components, and more or fewer components may be implemented instead.

[0077] In some embodiments, the memory 20 may be an internal storage unit of the electronic device, such as a hard disk or memory of the electronic device. In other embodiments, the memory 20 may also be an external storage device of the electronic device, such as a plug-in hard disk equipped on the electronic device, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Furthermore, the memory 20 may also include both an internal storage unit of the electronic device and an external storage device. The memory 20 is used to store application software and various types of data installed in the electronic device, such as program codes installed in the electronic device. The memory 20 can also be used to temporarily store data that has been output or is to be output. In one embodiment, a brake status detection program 40 is stored on the memory 20, and the brake status detection program 40 can be executed by the processor 10, thereby realizing the brake status detection method of each embodiment of the present invention.

[0078] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor, or other data processing chip, configured to execute program codes stored in the memory 20 or process data, such as executing a brake status detection method.

[0079] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display 30 is used to display information on the electronic device and to display a visual user interface. The components 10-30 of the electronic device communicate with each other via a system bus.

[0080] In some embodiments, when the processor 10 executes the brake state detection program 40 in the memory 20 , the steps of the brake state detection method described in the above embodiments are implemented.

[0081] In summary, the brake state identification method, device, electronic device and storage medium provided by the present invention first obtain three-axis acceleration data, then pre-process the three-axis acceleration data, and then determine whether the current state is a pseudo-braking state based on the acceleration data in the acceleration direction perpendicular to the riding direction and the acceleration data in the acceleration direction collinear with the gravity acceleration. If it is not a pseudo-braking state, the brake state point in the X-axis acceleration data is determined based on the pre-processed X-axis acceleration data, thereby realizing the detection of the brake state. The present invention can improve the signal-to-noise ratio of the feature data by characterizing the accelerometer data, achieving the purpose of reducing data input and improving efficiency without reducing the accuracy of identifying the brake state, and eliminating the static calibration process. In addition, by performing pseudo-braking state judgment, it can effectively reduce the brake misjudgment caused by some pseudo-braking signals, making the brake judgment more robust. In addition, the present invention does not need to be connected to other hardware devices, and only needs to be equipped with an accelerometer to realize the recognition of the brake state, thereby reducing costs.

[0082] Of course, those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware (such as a processor, controller, etc.) through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a memory, a magnetic disk, an optical disk, etc.

[0083] The specific embodiments of the present invention described above do not limit the scope of protection of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. A method for identifying a braking state, characterized in that: The steps include: Acquire three-axis acceleration data, and pre-process the three-axis acceleration data, wherein the pre-processing of the three-axis acceleration data includes: performing band-pass baseline removal processing on the three-axis acceleration data to obtain the three-axis acceleration data within the window; Based on the preprocessed Z-axis acceleration data, determine whether the vehicle is currently in a riding state; based on the preprocessed Y-axis acceleration data, determine whether the vehicle is currently in a shaking state, wherein the Y-axis represents the acceleration direction perpendicular to the riding direction, and the Z-axis represents the acceleration direction collinear with the gravity acceleration. When the vehicle is currently in a riding state and not in a shaking state, determine that the vehicle is currently not in a pseudo-braking state; otherwise, determine that the vehicle is currently in a pseudo-braking state; When the vehicle is not currently in a pseudo-braking state, the braking state point in the X-axis acceleration data is determined based on the preprocessed X-axis acceleration data, and the timestamp corresponding to the braking state point is used as the braking moment, wherein the X-axis represents the acceleration direction collinear with the riding direction.

2. The brake status recognition method according to claim 1, characterized in that: The following formula is used to perform bandpass baseline removal processing on the triaxial acceleration data: , Among them, Acc represents the input three-axis acceleration data, where the acceleration data of each axis is composed of the accelerations at different times in n corresponding directions, y represents the output filtered value, a and b represent the denominator vector and numerator vector of the corresponding Butterworth filter, the order of the bandpass filter is M, and n represents the length of the sequence.

3. The brake status recognition method according to claim 1, characterized in that: The determining whether the vehicle is currently in a riding state based on the preprocessed Z-axis acceleration data includes: The mean of the preprocessed Z-axis acceleration data is calculated, and after taking the absolute value, it is determined whether the absolute value is greater than a first threshold value. If so, it is determined that the current state is riding, otherwise it is determined that the current state is pseudo riding.

4. The brake status recognition method according to claim 1, characterized in that: The method of judging whether the vehicle is currently in a shaking state based on the preprocessed Y-axis acceleration data includes: Obtain the mean error of the preprocessed Y-axis acceleration data, and determine whether the mean error is greater than a second threshold value. If so, determine that the vehicle is currently in a shaking state; otherwise, determine that the vehicle is currently in a normal riding state.

5. The brake status recognition method according to claim 1, characterized in that: When the vehicle is not currently in a pseudo-braking state, based on the pre-processed X-axis acceleration data, a braking state point in the X-axis acceleration data is determined, and a timestamp corresponding to the braking state point is used as the braking moment, including: The preprocessed X-axis acceleration data is subjected to valley detection. If a valley exists and the valley is less than a third threshold, the valley point is determined to be a braking state point, and a timestamp corresponding to the braking state point is used as the braking moment. Otherwise, the next round of iteration is performed.

6. A brake state detection device, characterized in that: include: A data processing module, used for acquiring three-axis acceleration data and preprocessing the three-axis acceleration data, wherein the preprocessing of the three-axis acceleration data includes: performing bandpass baseline removal processing on the three-axis acceleration data to obtain the three-axis acceleration data within the window; The pseudo-brake state detection module is used to determine whether the vehicle is currently in a riding state based on the pre-processed Z-axis acceleration data, and to determine whether the vehicle is currently in a shaking state based on the pre-processed Y-axis acceleration data, wherein the Y-axis represents the acceleration direction perpendicular to the riding direction, and the Z-axis represents the acceleration direction collinear with the gravity acceleration. When the vehicle is currently in a riding state and not in a shaking state, it is determined that the vehicle is not currently in a pseudo-brake state, otherwise it is determined that the vehicle is currently in a pseudo-brake state; The brake state detection module is used to determine the brake state point in the X-axis acceleration data based on the preprocessed X-axis acceleration data when the vehicle is not currently in a pseudo-brake state, and use the timestamp corresponding to the brake state point as the braking moment, wherein the X-axis represents the acceleration direction collinear with the riding direction.

7. An electronic device, characterized in that: including a processor and a memory; The memory stores a computer-readable program executable by the processor; When the processor executes the computer-readable program, the steps in the brake status identification method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the brake status identification method according to any one of claims 1-5.