Tire explosion-proof early warning method and system based on wheel rotating speed recognition

Abstract

The invention relates to a tire explosion-proof early warning method and system based on wheel rotating speed recognition, and belongs to the technical field of automobile safety. The method comprises the following steps: acquiring a wheel rotating speed signal, and correcting the wheel rotating speed signal through a signal correction model to obtain a wheel rotating speed time domain signal; converting the wheel rotating speed time domain signal to obtain a wheel rotating speed frequency domain signal; extracting a wheel speed time domain signal and a wheel speed frequency domain signal through features to obtain a wheel speed signal feature; identifying and converting wheel speed signal characteristics through an indirect monitoring model to obtain a tire pressure state; and obtaining a current environment image, processing the tire pressure state and the current environment image through a multivariate coupling model, and outputting tire explosion-proof early warning information. Tire explosion-proof early warning based on wheel rotating speed identification is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of automobile safety, and in particular relates to a tire explosion prevention early warning method and system based on wheel speed recognition. Background Art

[0002] In recent years, with the continuous increase in car ownership, while people enjoy the convenience of automobiles, they are also paying more and more attention to driving safety. As the only contact component between the vehicle and the road, tires play a vital role in vehicle safety. Abnormal internal tire pressure can easily lead to serious traffic accidents such as tire blowouts. Therefore, how to achieve real-time monitoring of tire pressure and abnormal warning has become an important research direction for improving the level of active vehicle safety. In particular, indirect pressure measurement and warning through wheel speed signals have attracted widespread attention due to their advantages such as independence from external sensors and rapid response.

[0003] Traditional tire pressure monitoring methods based on radius estimation primarily indirectly determine tire pressure status by estimating the mapping relationship between tire radius, relative radius, or effective rolling radius and tire pressure. However, this method can typically only identify abnormal conditions in one to three tires, resulting in limited monitoring accuracy and adaptability. In contrast, tire pressure identification methods based directly on wheel speed signals are more accurate and real-time. However, due to the highly dynamic environment in which vehicles operate, effectively integrating environmental data with wheel speed characteristics to achieve rapid perception and accurate warning of tire blowout risks under complex operating conditions remains a key issue that needs to be addressed. Summary of the Invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides a tire explosion prevention warning method and system based on wheel speed recognition.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A tire explosion prevention warning method based on wheel speed recognition includes:

[0007] S1: Acquire a wheel speed signal, and correct the wheel speed signal using a signal correction model to obtain a wheel speed time domain signal;

[0008] S2: converting the wheel speed time domain signal to obtain a wheel speed frequency domain signal; extracting features from the wheel speed time domain signal and the wheel speed frequency domain signal to obtain wheel speed signal features;

[0009] S3: Obtaining a tire pressure state by identifying and converting the wheel speed signal characteristics through an indirect monitoring model;

[0010] S4: Acquire a current environment image, process the tire pressure state and the current environment image through a multivariate coupling model, and output tire explosion prevention warning information.

[0011] Preferably, the wheel speed signal correction process in step S1 includes:

[0012] S101: Calculating the wheel speed signal error by the least square method to obtain a ring gear error;

[0013] S102: Correcting the wheel speed signal using the ring gear error to obtain a reference wheel speed signal;

[0014] S103: Linearly interpolating the reference wheel speed signal to obtain a reference wheel speed reconstruction signal;

[0015] S104: Reconstructing the reference wheel speed reconstructed signal through adaptive weight gradient to obtain the wheel speed time domain signal.

[0016] Preferably, the mathematical expression converted in step S2 is:

[0017]

[0018] Wherein, X(k) is the wheel speed frequency domain signal, n represents the time domain index variable, i.e., the sampling point position currently being processed, N is the total number of sampling points, i.e., the length of the wheel speed time domain signal, x(n) is the wheel speed time domain signal, k represents the frequency index variable, and j is the imaginary unit.

[0019] Preferably, the identification conversion process of the indirect monitoring model in step S3 is:

[0020] S301: Identifying the wheel speed signal characteristics through a tire pressure recognition network to obtain normal wheel speed signal characteristics;

[0021] S302: Predict and convert the normal wheel speed signal characteristics to obtain the tire pressure state.

[0022] Preferably, the identification process of the tire pressure identification network in step S301 is:

[0023] S301-1: Processing the wheel speed signal characteristics through adaptive wavelet threshold denoising to obtain a reference frequency band resonance characteristic;

[0024] S301 - 2 : Classifying the reference frequency band resonance characteristics through a decision tree to obtain the normal wheel speed signal characteristics.

[0025] Preferably, the specific process of the prediction conversion in step S302 is:

[0026] S302-1: Obtaining an initial tire pressure state by predicting the normal wheel speed signal characteristics;

[0027] S302-2: Preset a tire stiffness compensation factor and an environment compensation weight and correct the initial tire pressure state to obtain a tire pressure reference state;

[0028] S302-3: Adding a timestamp to the tire pressure reference state to obtain a time series tire pressure reference state; obtaining a verification coefficient by linearly calculating the time series tire pressure reference state;

[0029] S302-4: Obtain the tire pressure state by verifying the verification coefficient through multi-time domain consistency.

[0030] Preferably, the method for multi-time domain consistency verification in step S302-4 is:

[0031] A verification threshold is preset, and whether the verification coefficient is less than the verification threshold is determined based on the verification coefficient and the verification threshold:

[0032] If yes, the time series tire pressure reference state is the tire pressure state;

[0033] If not, the agent confirms the time series tire pressure reference state.

[0034] Preferably, the modeling process of the multivariate coupling model in step S4 is:

[0035] S401: Obtaining environmental features by feature extraction of the current environmental image;

[0036] S402: Classify the environmental features using a classifier to obtain an environmental scene category;

[0037] S403: Mapping the environmental scene categories to obtain an environmental risk factor matrix, and calculating the environmental risk factor matrix through dot product fusion to obtain a weight adjustment factor;

[0038] S404: Obtaining a risk assessment coefficient by combining the weight adjustment factor and the tire pressure state risk assessment;

[0039] S405: Preset a risk threshold, determine the risk threshold and the risk assessment coefficient through multi-element coupling to output the tire explosion prevention warning information, wherein the risk threshold includes an upper threshold limit and a lower threshold limit.

[0040] Preferably, the multi-coupling determination method in step S405 is:

[0041] Determine whether the risk assessment coefficient is greater than the upper threshold:

[0042] If yes, the tire explosion warning information is "red warning";

[0043] If not, the risk assessment coefficient is further compared to see whether it is greater than the lower limit of the threshold. If it is greater, the tire explosion-proof warning information is "yellow warning"; if it is less than, the tire explosion-proof warning information is "normal".

[0044] A tire explosion warning system based on wheel speed recognition includes a signal correction module, a feature extraction module, a tire pressure conversion module, and a coupling evaluation module, including:

[0045] The signal correction module is used to obtain a wheel speed signal, and correct the wheel speed signal using a signal correction model to obtain a wheel speed time domain signal;

[0046] The feature extraction module is used to convert the wheel speed time domain signal into a wheel speed frequency domain signal; and obtain wheel speed signal features by feature extraction of the wheel speed time domain signal and the wheel speed frequency domain signal;

[0047] The tire pressure conversion module is used to obtain the tire pressure state by identifying and converting the wheel speed signal characteristics through an indirect monitoring model;

[0048] The coupling evaluation module is used to obtain a current environment image, process the tire pressure state and the current environment image through a multivariate coupling model, and output tire explosion prevention warning information.

[0049] The beneficial effects of the present invention are:

[0050] (1) The wheel speed signal is corrected by the signal correction model, and the frequency domain and time domain feature extraction are combined to effectively improve the stability and accuracy of the wheel speed feature, thereby achieving more accurate indirect tire pressure identification and compensating for the delay and drift problems of traditional tire pressure sensors.

[0051] (2) Introduce a multivariate coupling model, combine it with environmental image recognition, dynamically extract environmental scene risk factors, form a weight adjustment mechanism that adapts to different driving scenarios, and improve the adaptability and sensitivity of the early warning system to complex road conditions.

[0052] (3) Through the multi-level coupling judgment of risk assessment coefficient and upper and lower thresholds, the red warning, yellow warning and normal status are clearly distinguished, and the tire status is managed in a refined manner, the accuracy of tire explosion warning is improved, and the driving safety level and driving decision-making assistance capabilities are enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0054] Figure 1The figure is a flow chart of a tire explosion prevention warning method based on wheel speed recognition according to the present invention. DETAILED DESCRIPTION

[0055] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0056] See also Figure 1 A tire explosion prevention warning method based on wheel speed recognition includes:

[0057] S1: Acquire a wheel speed signal, and correct the wheel speed signal using a signal correction model to obtain a wheel speed time domain signal;

[0058] S2: Converting the wheel speed time domain signal to obtain a wheel speed frequency domain signal; extracting the wheel speed time domain signal and the wheel speed frequency domain signal to obtain wheel speed signal features, wherein the wheel speed signal features include wheel speed signal time domain features and wheel speed signal frequency domain features;

[0059] S3: Obtaining a tire pressure state by identifying and converting the wheel speed signal characteristics through an indirect monitoring model;

[0060] S4: Acquire a current environment image, process the tire pressure state and the current environment image through a multivariate coupling model, and output tire explosion prevention warning information.

[0061] Specifically, in step S1, the process of correcting the wheel speed signal using the signal correction model includes:

[0062] S101: Calculating the wheel speed signal error by the least square method to obtain a ring gear error;

[0063] S102: Correcting the wheel speed signal using the ring gear error to obtain a reference wheel speed signal;

[0064] S103: Linearly interpolating the reference wheel speed signal to obtain a reference wheel speed reconstruction signal;

[0065] S104: Reconstructing the reference wheel speed reconstructed signal through adaptive weight gradient to obtain the wheel speed time domain signal.

[0066] In this embodiment, the signal correction model obtains a discrete and uniformly distributed wheel speed time domain signal, so the wheel speed time domain signal needs to be converted into the frequency domain for subsequent feature analysis of the frequency domain wheel speed signal.

[0067] Specifically, the mathematical expression converted in step S2 is:

[0068]

[0069] Wherein, X(k) is the wheel speed frequency domain signal, n represents the time domain index variable, i.e., the sampling point position currently being processed, N is the total number of sampling points, i.e., the length of the wheel speed time domain signal, x(n) is the wheel speed time domain signal, k represents the frequency index variable, and j is the imaginary unit.

[0070] In this embodiment, normal wheel speed signal characteristics refer to wheel speed signal characteristics suitable for indirect monitoring, and problematic wheel speed signal characteristics refer to wheel speed signal characteristics unsuitable for indirect monitoring. Normal wheel speed signal characteristics and problematic wheel speed signal characteristics can be distinguished by the magnitude of the fluctuation of the wheel speed time domain signal characteristics and the amplitude-frequency characteristics of the wheel speed frequency domain signal characteristics. Problematic wheel speed signal characteristics include: wheel speed is too low, wheel speed jitter is too large, and periodic noise is present. The wheel speed average of the normal wheel speed signal characteristic is 51 km / h, and the fluctuation amplitude is approximately 0.5 km / h. The problematic wheel speed signal characteristic of too low wheel speed undergoes sudden changes at regular intervals, and there is no resonance peak reflecting the wheel speed signal characteristic within the entire frequency domain; the problematic wheel speed signal characteristic of excessive jitter has a fluctuation amplitude of 0.8 km / h. The problematic wheel speed signal characteristic with periodic noise is caused by the failure to completely eliminate the wheel speed sensor error. Therefore, the problematic wheel speed signal characteristic is not suitable for indirect tire pressure monitoring.

[0071] Specifically, the identification conversion process of the indirect monitoring model in step S3 is:

[0072] S301: Identifying the wheel speed signal characteristics through a tire pressure recognition network to obtain normal wheel speed signal characteristics;

[0073] S302: Predict and convert the normal wheel speed signal characteristics to obtain the tire pressure state.

[0074] Specifically, the tire pressure recognition network in step S301 is specifically modeled as follows:

[0075] S301-1: Processing the wheel speed signal characteristics through adaptive wavelet threshold denoising to obtain a reference frequency band resonance characteristic;

[0076] S301 - 2 : Classifying the reference frequency band resonance characteristics through a decision tree to obtain the normal wheel speed signal characteristics.

[0077] Specifically, the specific process of the prediction conversion in step S302 is:

[0078] S302-1: Analyzing a characteristic curve of the normal wheel speed signal characteristic to obtain a characteristic drift; predicting the normal wheel speed signal characteristic using a deep regression network to obtain an initial tire pressure state;

[0079] S302-2: Preset a tire stiffness compensation factor and an environment compensation weight and correct the initial tire pressure state to obtain a tire pressure reference state;

[0080] The modified expression is:

[0081] P t =P0·(1+α s tanh(Δs)-β e ·cos(ω e t)),

[0082] Among them, P t is the tire pressure reference state, P0 is the tire pressure initial state, α s is the tire stiffness compensation factor, Δs is the characteristic drift, β e is the environmental compensation weight, ω e is the environmental fluctuation frequency coefficient, t represents the current sampling time, tanh represents the tangent function, and cos represents the cosine function;

[0083] S302-3: Adding a timestamp to the tire pressure reference state to obtain a time series tire pressure reference state; obtaining a verification coefficient by linearly calculating the time series tire pressure reference state;

[0084] The calculation expression of the verification coefficient is:

[0085]

[0086] Wherein, Δp is the verification coefficient, W is the number of time-series tire pressure reference states, i represents the i-th time-series tire pressure reference state, Pt(i) represents the i-th time-series tire pressure reference state, and Pt(i-1) represents the i-1-th time-series tire pressure reference state;

[0087] S302-4: Verifying the verification coefficient by multi-time domain consistency to obtain the tire pressure state;

[0088] The multi-time domain consistency verification method is as follows: presetting a verification threshold, and judging whether the verification coefficient is less than the verification threshold according to the verification coefficient and the verification threshold:

[0089] If yes, the time series tire pressure reference state is the tire pressure state;

[0090] If not, the agent confirms the time series tire pressure reference state.

[0091] In this embodiment, the deep regression network includes an input layer, a hidden layer, and an output layer. The hidden layer includes three layers of gated recurrent units (GRU) to process temporal features, with 64 neurons in each layer.

[0092] Specifically, the current environment image includes but is not limited to current temperature, road humidity, load weight, gestational age, and vehicle speed.

[0093] Specifically, the modeling process of the multivariate coupling model in step S4 is:

[0094] S401: Obtaining environmental features by feature extraction of the current environmental image;

[0095] S402: Obtaining an environmental scene category according to the environmental feature classification;

[0096] S403: Mapping the environmental scene categories to obtain an environmental risk factor matrix, and calculating the environmental risk factor matrix through dot product fusion to obtain a weight adjustment factor;

[0097] S404: Obtaining a risk assessment coefficient by combining the weight adjustment factor and the tire pressure state risk assessment;

[0098] The functional expression of the risk assessment is:

[0099] R f =σ(μ·|P ref -P est |+η·D f +γ·R env ),

[0100] Among them, R f is the risk assessment coefficient, σ represents normalization processing, μ, η, γ are adjustable weight coefficients, P ref is the reference tire pressure value, P est is the tire pressure state, D f is the characteristic drift, R env is the weight adjustment factor;

[0101] S405: Preset a risk threshold, determine the risk threshold and the risk assessment coefficient through multi-element coupling to output the tire explosion warning information, the risk threshold includes an upper threshold and a lower threshold, and the tire explosion warning information includes red warning, yellow warning, and normal.

[0102] Specifically, the multi-coupling judgment method is:

[0103] Determine whether the risk assessment coefficient is greater than the upper threshold:

[0104] If yes, the tire explosion warning information is "red warning";

[0105] If not, the risk assessment coefficient is further compared to see whether it is greater than the lower limit of the threshold. If it is greater, the tire explosion-proof warning information is "yellow warning"; if it is less than, the tire explosion-proof warning information is "normal".

[0106] Specifically, the environmental scene categories include slippery road, gravel road, urban road, and highway; taking slippery road as an example, when the environmental scene category is the slippery road, the environmental risk factor matrix is mapped as W env ={β t =1.3,ω t =1.1}, where W env is the environmental risk factor matrix, β t is the tire deformation threshold, ω t Enhancement item for environmental factors.

[0107] In Example 1, a vehicle traveling at a constant speed of 60 km / h on a highway, with normal wheel speed signals, an ambient temperature of 33°C, and a load of 80%, was used as an example. The present invention collected the wheel speed signals and extracted a frequency domain peak frequency of 3.2 Hz with a characteristic drift of 0.03. The initial predicted tire pressure, P0, was 2.25 bar, and the corrected Pt was 2.18 bar. The early warning model outputted a risk level of 0.72, triggering a red alert.

[0108] Example 2, a tire explosion prevention warning system based on wheel speed recognition, includes a signal correction module, a feature extraction module, a tire pressure conversion module, and a coupling evaluation module, including:

[0109] The signal correction module is used to obtain a wheel speed signal, and correct the wheel speed signal using a signal correction model to obtain a wheel speed time domain signal;

[0110] The feature extraction module is used to convert the wheel speed time domain signal into a wheel speed frequency domain signal; and obtain wheel speed signal features by feature extraction of the wheel speed time domain signal and the wheel speed frequency domain signal;

[0111] The tire pressure conversion module is used to obtain the tire pressure state by identifying and converting the wheel speed signal characteristics through an indirect monitoring model;

[0112] The coupling evaluation module is used to obtain a current environment image, process the tire pressure state and the current environment image through a multivariate coupling model, and output tire explosion prevention warning information.

[0113] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A tire explosion prevention warning method based on wheel speed recognition, characterized in that: The following steps are involved: S1: Acquire a wheel speed signal, and correct the wheel speed signal using a signal correction model to obtain a wheel speed time domain signal; S2: converting the wheel speed time domain signal to obtain a wheel speed frequency domain signal; extracting features from the wheel speed time domain signal and the wheel speed frequency domain signal to obtain wheel speed signal features; S3: Obtaining a tire pressure state by identifying and converting the wheel speed signal characteristics through an indirect monitoring model; S4: Acquire a current environment image, process the tire pressure state and the current environment image through a multivariate coupling model, and output tire explosion prevention warning information.

2. The tire explosion prevention warning method based on wheel speed recognition according to claim 1, characterized in that: The wheel speed signal correction process in step S1 includes: S101: Calculating the wheel speed signal error by the least square method to obtain a ring gear error; S102: Correcting the wheel speed signal using the ring gear error to obtain a reference wheel speed signal; S103: Linearly interpolating the reference wheel speed signal to obtain a reference wheel speed reconstruction signal; S104: Reconstructing the reference wheel speed reconstructed signal through adaptive weight gradient to obtain the wheel speed time domain signal.

3. The tire explosion prevention warning method based on wheel speed recognition according to claim 1, characterized in that: The mathematical expression converted in step S2 is: Wherein, X(k) is the wheel speed frequency domain signal, n represents the time domain index variable, i.e., the sampling point position currently being processed, N is the total number of sampling points, i.e., the length of the wheel speed time domain signal, x(n) is the wheel speed time domain signal, k represents the frequency index variable, and j is the imaginary unit.

4. The tire explosion prevention warning method based on wheel speed recognition according to claim 1, characterized in that: The identification and conversion process of the indirect monitoring model in step S3 is: S301: Identifying the wheel speed signal characteristics through a tire pressure recognition network to obtain normal wheel speed signal characteristics; S302: Predict and convert the normal wheel speed signal characteristics to obtain the tire pressure state.

5. The tire explosion prevention warning method based on wheel speed recognition according to claim 4, characterized in that: The identification process of the tire pressure identification network in step S301 is as follows: S301-1: Processing the wheel speed signal characteristics through adaptive wavelet threshold denoising to obtain a reference frequency band resonance characteristic; S301 - 2 : Classifying the reference frequency band resonance characteristics through a decision tree to obtain the normal wheel speed signal characteristics.

6. The tire explosion prevention warning method based on wheel speed recognition according to claim 4, characterized in that: The specific process of the prediction conversion in step S302 is as follows: S302-1: Obtaining an initial tire pressure state by predicting the normal wheel speed signal characteristics; S302-2: Preset a tire stiffness compensation factor and an environment compensation weight and correct the initial tire pressure state to obtain a tire pressure reference state; S302-3: Adding a timestamp to the tire pressure reference state to obtain a time series tire pressure reference state; obtaining a verification coefficient by linearly calculating the time series tire pressure reference state; S302-4: Obtain the tire pressure state by verifying the verification coefficient through multi-time domain consistency.

7. The tire explosion prevention warning method based on wheel speed recognition according to claim 6, characterized in that: The method for multi-time domain consistency verification in step S302-4 is: A verification threshold is preset, and whether the verification coefficient is less than the verification threshold is determined based on the verification coefficient and the verification threshold: If yes, the time series tire pressure reference state is the tire pressure state; If not, the agent confirms the time series tire pressure reference state.

8. The tire explosion prevention warning method based on wheel speed recognition according to claim 1, characterized in that: The modeling process of the multivariate coupling model in step S4 is: S401: Obtaining environmental features by feature extraction of the current environmental image; S402: Classify the environmental features using a classifier to obtain an environmental scene category; S403: Mapping the environmental scene categories to obtain an environmental risk factor matrix, and calculating the environmental risk factor matrix through dot product fusion to obtain a weight adjustment factor; S404: Obtaining a risk assessment coefficient by combining the weight adjustment factor and the tire pressure state risk assessment; S405: Preset a risk threshold, determine the risk threshold and the risk assessment coefficient through multi-element coupling to output the tire explosion prevention warning information, wherein the risk threshold includes an upper threshold limit and a lower threshold limit.

9. The tire explosion prevention warning method based on wheel speed recognition according to claim 8, characterized in that: The multi-coupling determination method in step S405 is: Determine whether the risk assessment coefficient is greater than the upper threshold: If yes, the tire explosion warning information is "red warning"; If not, the risk assessment coefficient is further compared to see whether it is greater than the lower limit of the threshold. If it is greater than, the tire explosion-proof warning information is "yellow warning"; if it is less than, the tire explosion-proof warning information is "normal".

10. A tire explosion warning system based on wheel speed recognition, comprising a signal correction module, a feature extraction module, a tire pressure conversion module, and a coupling evaluation module, characterized in that: include: The signal correction module is used to obtain a wheel speed signal, and correct the wheel speed signal using a signal correction model to obtain a wheel speed time domain signal; The feature extraction module is used to convert the wheel speed time domain signal into a wheel speed frequency domain signal; and obtain wheel speed signal features by feature extraction of the wheel speed time domain signal and the wheel speed frequency domain signal; The tire pressure conversion module is used to obtain the tire pressure state by identifying and converting the wheel speed signal characteristics through an indirect monitoring model; The coupling evaluation module is used to obtain a current environment image, process the tire pressure state and the current environment image through a multivariate coupling model, and output tire explosion prevention warning information.

Images