A vehicle tire positioning method, device, equipment and storage medium

Abstract

This application discloses a vehicle tire positioning method, apparatus, device, and storage medium, relating to the field of vehicle engineering. The method includes: utilizing the onboard ultra-wideband radar of a target vehicle to transmit status detection signals and acquiring echo signals corresponding to the tire pressure sensors of each target tire of the target vehicle; processing the echo signals using a preset feature extraction algorithm to determine the target features corresponding to the echo signals; the target features include first-arrival path time, micro-Doppler features, and echo amplitude features; obtaining the target feature vector corresponding to the echo signals based on the target features, and using a target inference model to determine the tire position and tire pressure status of the target tire corresponding to the target feature vector. This application achieves accurate positioning of vehicle tires.

Description

Technical Field

[0001] This invention relates to the field of vehicle engineering, and in particular to a method, apparatus, device, and storage medium for vehicle tire positioning. Background Technology

[0002] In automotive engineering, sensors are commonly used to locate vehicle tires. Traditional sensors calculate their position using algorithms such as acceleration, a process that consumes a significant amount of power and shortens the sensor's lifespan. Furthermore, the tire position calculated using acceleration by traditional sensors has an error margin of about 10°, resulting in inaccurate positioning. Summary of the Invention

[0003] In view of this, the purpose of this invention is to provide a vehicle tire positioning method, apparatus, device, and storage medium capable of accurately positioning vehicle tires. The specific solution is as follows: In a first aspect, this application discloses a vehicle tire positioning method, including: The vehicle's onboard ultra-wideband radar transmits status detection signals, and the echo signals corresponding to the tire pressure sensors of each target tire of the target vehicle are obtained. The echo signal is processed using a preset feature extraction algorithm to determine the target features corresponding to the echo signal; the target features include first-arrival path time, micro-Doppler features, and echo amplitude features; Based on the target features, the target feature vector corresponding to the echo signal is obtained, and the tire position and tire pressure state of the target tire corresponding to the target feature vector are determined using the target inference model.

[0004] Optionally, processing the echo signal using a preset feature extraction algorithm to determine the target features corresponding to the echo signal includes: The echo signal is filtered using a preset noise reduction algorithm to obtain the filtered signal; The target features corresponding to the echo signal are determined based on the filtered signal using a preset feature extraction algorithm.

[0005] Optionally, determining the target features corresponding to the echo signal based on the filtered signal using a preset feature extraction algorithm includes: If the state detection signal is a modulated pulse signal, then the filtered signal is pulse compressed to obtain the corresponding compressed signal, and a preset feature extraction algorithm is used to determine the target features corresponding to the echo signal based on the compressed signal.

[0006] Optionally, determining the target features corresponding to the echo signal based on the filtered signal using a preset feature extraction algorithm includes: The first-arrival path time of the echo signal is determined based on the filtered signal using a first feature extraction algorithm; the first feature extraction algorithm includes a matched filtering algorithm and a leading edge detection algorithm. The filtered signal is subjected to time-frequency analysis using a second feature extraction algorithm to determine the micro-Doppler features corresponding to the echo signal.

[0007] Optionally, determining the tire position and tire pressure state of the target tire corresponding to the target feature vector using the target inference model includes: The target inference model is used to determine the spatial point cloud of the reflection point corresponding to each echo signal based on the target feature vector, and the tire position corresponding to each target tire is determined based on the vehicle coordinate system corresponding to the target vehicle and the spatial point cloud of the reflection point. The vibration mode of the target tire is determined based on the driving state of the target vehicle using the target inference model, and the tire pressure state of the target tire is determined based on the tire position corresponding to the target tire, the vibration mode, and the target feature vector.

[0008] Optionally, the target inference model includes a multilayer perceptron neural network constructed using a preset multi-task learning loss function.

[0009] Secondly, this application discloses a vehicle tire alignment device, comprising: The echo signal acquisition module is used to utilize the vehicle-mounted ultra-wideband radar of the target vehicle to transmit status detection signals and acquire the echo signals corresponding to the tire pressure sensors of each target tire of the target vehicle. The feature extraction module is used to process the echo signal using a preset feature extraction algorithm to determine the target features corresponding to the echo signal; the target features include first-arrival path time, micro-Doppler features, and echo amplitude features; The tire positioning module is used to obtain the target feature vector corresponding to the echo signal based on the target features, and to determine the tire position and tire pressure state of the target tire corresponding to the target feature vector using a target inference model.

[0010] Optionally, the feature extraction module includes: The signal filtering submodule is used to filter the echo signal using a preset noise reduction algorithm to obtain the filtered signal; The feature extraction submodule is used to determine the target features corresponding to the echo signal based on the filtered signal using a preset feature extraction algorithm.

[0011] Thirdly, this application discloses an electronic device, including: Memory, used to store computer programs; A processor is used to execute the computer program to implement the aforementioned vehicle tire positioning method.

[0012] Fourthly, this application discloses a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the aforementioned vehicle tire positioning method.

[0013] In this application, when performing vehicle tire positioning, the vehicle-mounted ultra-wideband radar of the target vehicle emits a status detection signal, and the echo signals corresponding to the tire pressure sensors of each target tire of the target vehicle are obtained; the echo signals are processed using a preset feature extraction algorithm to determine the target features corresponding to the echo signals; the target features include first-arrival path time, micro-Doppler features, and echo amplitude features; based on the target features, the target feature vector corresponding to the echo signals is obtained, and the tire position and tire pressure status of the target tire corresponding to the target feature vector are determined using a target inference model. As can be seen, this application employs onboard ultra-wideband radar for active detection, where the tire pressure sensor only passively reflects the echo signal. This eliminates the need for position calculation and active reporting via acceleration algorithms, fundamentally reducing power consumption from complex calculations, significantly lowering sensor power consumption, and extending lifespan. By extracting multi-dimensional features such as first-arrival path time, micro-Doppler, and echo amplitude, and combining them with an inference model for position determination, it replaces the traditional indirect acceleration calculation method, effectively eliminating inherent positioning errors and significantly improving tire position acquisition accuracy. Simultaneously, it synchronously identifies tire position and tire pressure status based on the same signal features, achieving integrated positioning and status monitoring. This results in more direct signal processing, stronger anti-interference capabilities, and improved accuracy, real-time performance, and system stability of tire monitoring. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0015] Figure 1 This is a flowchart of a vehicle tire positioning method disclosed in this application; Figure 2 This is a schematic diagram of the structure of a vehicle tire positioning device disclosed in this application; Figure 3 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] In the field of vehicle engineering, sensors are commonly used to locate vehicle tires. Traditional sensors transmit their own position through calculations using algorithms such as acceleration, a process that consumes a significant amount of power and shortens the sensor's lifespan. Furthermore, the tire position calculated by traditional sensors using acceleration has an error margin of approximately 10°, resulting in inaccurate positioning. To address these technical problems, this application discloses a vehicle tire positioning method capable of accurately locating vehicle tires.

[0018] See Figure 1 As shown, an embodiment of the present invention discloses a vehicle tire alignment method, including: Step S11: Use the vehicle-mounted ultra-wideband radar of the target vehicle to transmit status detection signals and obtain the echo signals corresponding to the tire pressure sensors of each target tire of the target vehicle.

[0019] In this embodiment, the vehicle's onboard ultra-wideband radar transmits status detection signals, and the echo signals corresponding to the tire pressure sensors of each target tire are acquired. Specifically, the UWB (Ultra Wide Band) module transmits ultra-wideband signals (e.g., 6.5–8 GHz) in pulse mode as status detection signals and receives the echoes. The TPMS (Tire Pressure Monitoring System) on each target tire of the target vehicle does not need to be modified, as its metal casing / valve naturally becomes a strong reflector for the UWB. In practical applications, the receiving antenna captures the UWB pulse echoes, forming a baseband equivalent signal r(t), with a sampling rate typically not lower than 8 GHz to match the GHz-level bandwidth.

[0020] Step S12: Process the echo signal using a preset feature extraction algorithm to determine the target features corresponding to the echo signal; the target features include first-arrival path time, micro-Doppler features, and echo amplitude features.

[0021] In this embodiment, the echo signal is processed using a preset feature extraction algorithm to determine the target features corresponding to the echo signal. This includes: filtering the echo signal using a preset noise reduction algorithm to obtain a filtered signal; and determining the target features corresponding to the echo signal based on the filtered signal using the preset feature extraction algorithm. Specifically, determining the target features corresponding to the echo signal based on the filtered signal using the preset feature extraction algorithm includes: if the state detection signal is a modulated pulse signal, then pulse compression is performed on the filtered signal to obtain a compressed signal, and the target features corresponding to the echo signal are determined based on the compressed signal using the preset feature extraction algorithm.

[0022] In one specific implementation, the target features include first-arrival path time (TOA-FP), micro-Doppler characteristics, and echo amplitude characteristics. The target features corresponding to the echo signal are determined based on the filtered signal using a preset feature extraction algorithm. Specifically, this may include: determining the TOA-FP corresponding to the echo signal using a first feature extraction algorithm; the first feature extraction algorithm includes a matched filtering algorithm and a leading-edge detection algorithm; and performing time-frequency analysis on the filtered signal using a second feature extraction algorithm to determine the micro-Doppler characteristics corresponding to the echo signal. The TOA-FP is the time taken for the first line-of-sight (LOS) component of the UWB signal reflected from the tire pressure sensor to reach the receiving antenna. Even with multipath propagation (such as wheel hub or vehicle body reflections), the high temporal resolution of UWB can still distinguish the first arrival pulse, and the TOA-FP can be used to accurately calculate the distance from the sensor to the radar. When the tire rotates, the tire pressure sensor moves in a circular motion with the wheel hub, causing periodic frequency modulation of the echo (similar to the micro-Doppler effect in radar). In the time-frequency domain (such as Short-Time Fourier Transform (STFT) or Wavelet Transform), this micro-Doppler effect will present sinusoidal or comb-shaped spectral lines. This micro-Doppler characteristic can be used to distinguish stationary objects (such as wheel hub brackets) from rotating sensors, and to identify the specific wheel position of the sensor (left front, right rear, etc.), thus aiding in determining tire speed and direction. Furthermore, tire pressure sensors typically contain a metal casing or PCB, whose RCS (radar cross section) differs from that of rubber and the wheel hub. This causes periodic fluctuations in echo energy at specific angles (as the tire rotates). Extracting the echo amplitude or energy characteristics from the echo signal can assist in target identification and enhance the signal-to-noise ratio. Understandably, if a dual-polarized antenna is used, because the tire pressure sensor's installation posture is fixed, its reflected signal may have a specific polarization response. In this case, the polarization characteristics in the echo signal can be analyzed to suppress clutter from the vehicle body or other parts based on these polarization characteristics.

[0023] In this embodiment, when detecting the first-arrival path time (TOA), energy detection can be used, with a sliding window calculating the energy; exceeding a threshold indicates the signal start. Alternatively, matched filtering can be used, correlated with a known transmitted pulse template, with the peak position representing the TOA. Akaike Information Criterion (AIC) or Minimum Description Length (MDL) can also be employed to automatically detect the number of multipath components and the TOA. Leading Edge Detection can also be used to directly locate the rising edge when the signal-to-noise ratio is high. It should be noted that the strong reflection caused by the metal wheel hub may obscure the true TOA; therefore, prior knowledge (such as minimum physical distance) or a machine learning classifier is needed to achieve accurate TOA detection. When extracting micro-Doppler features, time-frequency analysis is performed on multiple consecutive pulse echoes, including short-time Fourier transform, continuous wavelet transform, and Wigner-Ville distribution (WVD) (high resolution but with cross terms), to extract feature parameters, including micro-Doppler frequency, modulation period, spectral symmetry and harmonic structure. Radial velocity can be determined based on micro-Doppler frequency, and the axle distance from the tire pressure sensor to the vehicle tire can be determined based on modulation period.

[0024] Step S13: Based on the target features, obtain the target feature vector corresponding to the echo signal, and use the target inference model to determine the tire position and tire pressure state of the target tire corresponding to the target feature vector.

[0025] In this embodiment, the input to the target inference model is typically a structured feature vector, that is, the target feature vector corresponding to the echo signal obtained based on the target features, including distance features, micro-Doppler features, temporal features, motion context, historical states, etc. Specifically, the TOA distance, the micro-Doppler period corresponding to the micro-Doppler features, and the energy pattern are input to the rule engine. For example, if the micro-Doppler period is approximately equal to the tire rotation period, it is determined to be a TPMS, or a machine learning model (such as SVM, random forest, lightweight CNN) is used for classification to determine which tire pressure sensor the echo signal corresponds to. A typical input vector may be a 20-50 dimensional floating-point array.

[0026] In this embodiment, the architecture of the target inference model can be replaced according to requirements such as real-time performance, embedded deployment, and accuracy. Lightweight neural networks, tree models, and hybrid models are commonly used, specifically including the following architectures: 1. Lightweight neural networks are the mainstream choice, including multilayer perceptrons (MLP) and 1D convolutional neural networks (1D-CNN). The former has a simple structure and fast inference, making it suitable for MCUs or automotive SoCs. The input is passed through 2 to 3 fully connected layers (ReLU) to obtain a dual-head output of position and state. The latter can extract temporal patterns when the input is a raw or short time-series signal (such as a TOA sequence of 10 consecutive frames). It can also flatten the micro-Doppler time-frequency map before input.

[0027] 2. Tree models, used for interpretability scenarios, including XGBoost, LightGBM, and Random Forest, are often used for feature importance analysis during the R&D phase. They can be deployed to edge devices (such as the NXP S32K series). Their advantages are that they do not require a GPU, have a small memory footprint, and strong deterministic inference.

[0028] 3. Hybrid model: A combination of graph neural network (GNN) and Transformer encoder is adopted. The graph neural network treats the four wheels as graph nodes and uses geometric constraints for modeling. The Transformer encoder handles the time fusion of multiple sensors.

[0029] When implementing industrial applications, small MLPs or XGBoosts are usually the first choice to meet millisecond-level inference requirements and have a small memory footprint. They can be integrated into automotive ECUs via AUTOSAR or ROS2.

[0030] In one specific implementation, the target inference model includes a multilayer perceptron neural network constructed using a preset multi-task learning loss function. During the training phase, data is first collected by installing TPMS sensors at known locations on a real vehicle or drum test bench, simultaneously recording raw UWB radar echoes or extracted feature vectors, real wheel labels (manually labeled or mechanically located), and real tire pressure values ​​(read via a high-precision pressure gauge or CAN bus). Then, a dataset is constructed. This process can be augmented by adding Gaussian noise to simulate channel variations, artificially increasing TOA deviation to simulate NLOS (non-line-of-sight) scenarios, and simulating micro-Doppler variations at different vehicle speeds. Finally, multi-task learning is implemented using the loss function, with the Adam optimizer and a learning rate set to 1e-3 to 1e-4 to train the model. Validation metrics may include wheel position recognition accuracy (accuracy threshold set to greater than 98%) and tire pressure status F1-score (target score greater than 95%). Finally, the trained inference model is quantized (e.g., converted from FP32 to INT8) and pruned (removing redundant neurons), and then converted into C / C++ code to achieve the compression and deployment of the target inference model.

[0031] In one specific implementation, the tire position and tire pressure state of the target tire corresponding to the target feature vector are determined using a target inference model. This includes: using the target inference model to determine the spatial point cloud of reflection points corresponding to each echo signal based on the target feature vector, and determining the tire position of each target tire based on the vehicle coordinate system corresponding to the target vehicle and the spatial point cloud of reflection points; using the target inference model to determine the vibration mode of the target tire based on the driving state of the target vehicle, and determining the tire pressure state of the target tire based on the tire position, vibration mode, and target feature vector. UWB calculates the spatial coordinates of each reflection point using Time of Flight (ToF) or Angle of Arrival (AoA); each tire region forms a unique reflection cluster; the reflection point clouds of the four tires are separated using a clustering algorithm (such as DBSCAN); and automatically labeled as front left (FL), front right (FR), etc., in conjunction with the vehicle coordinate system. This yields a relatively accurate position with an accuracy of ±2cm. It can also be used to indirectly monitor tire pressure. The physical principle can be expressed as follows: decreased tire pressure leads to an increase in the tire's contact patch and sidewall curvature, causing a slight change in the spatial attitude of the rim / TPMS shell. This deformation leads to changes in reflection intensity and RCS (radar cross section); slight changes in path length cause phase differences, resulting in multipath phase shift; and during driving, the different vibration modes of the deformed tire lead to Doppler spectrum broadening. Using a target inference model based on the target feature vector, the tire pressure level, including normal / low / severely insufficient, can be obtained and output, thereby determining whether the target tire has a slow leaking trend.

[0032] As can be seen, this application employs onboard ultra-wideband radar for active detection, where the tire pressure sensor only passively reflects the echo signal. This eliminates the need for position calculation and active reporting via acceleration algorithms, fundamentally reducing power consumption from complex calculations, significantly lowering sensor power consumption, and extending lifespan. By extracting multi-dimensional features such as first-arrival path time, micro-Doppler, and echo amplitude, and combining them with an inference model for position determination, it replaces the traditional indirect acceleration calculation method, effectively eliminating inherent positioning errors and significantly improving tire position acquisition accuracy. Simultaneously, it synchronously identifies tire position and tire pressure status based on the same signal features, achieving integrated positioning and status monitoring. This results in more direct signal processing, stronger anti-interference capabilities, and improved accuracy, real-time performance, and system stability of tire monitoring.

[0033] See Figure 2 As shown, this application discloses a vehicle tire alignment device, comprising: The echo signal acquisition module 11 is used to utilize the vehicle-mounted ultra-wideband radar of the target vehicle to transmit status detection signals and acquire the echo signals corresponding to the tire pressure sensors of each target tire of the target vehicle. Feature extraction module 12 is used to process the echo signal using a preset feature extraction algorithm to determine the target features corresponding to the echo signal; the target features include first-arrival path time, micro-Doppler features, and echo amplitude features; The tire positioning module 13 is used to obtain the target feature vector corresponding to the echo signal based on the target features, and to determine the tire position and tire pressure state of the target tire corresponding to the target feature vector using the target inference model.

[0034] As can be seen, this application employs onboard ultra-wideband radar for active detection, where the tire pressure sensor only passively reflects the echo signal. This eliminates the need for position calculation and active reporting via acceleration algorithms, fundamentally reducing power consumption from complex calculations, significantly lowering sensor power consumption, and extending lifespan. By extracting multi-dimensional features such as first-arrival path time, micro-Doppler, and echo amplitude, and combining them with an inference model for position determination, it replaces the traditional indirect acceleration calculation method, effectively eliminating inherent positioning errors and significantly improving tire position acquisition accuracy. Simultaneously, it synchronously identifies tire position and tire pressure status based on the same signal features, achieving integrated positioning and status monitoring. This results in more direct signal processing, stronger anti-interference capabilities, and improved accuracy, real-time performance, and system stability of tire monitoring.

[0035] In one specific embodiment, the feature extraction module 12 may include: The signal filtering submodule is used to filter the echo signal using a preset noise reduction algorithm to obtain the filtered signal; The feature extraction submodule is used to determine the target features corresponding to the echo signal based on the filtered signal using a preset feature extraction algorithm.

[0036] In one specific implementation, the feature extraction submodule may include: The compression feature extraction unit is used to perform pulse compression on the filtered signal to obtain the corresponding compressed signal if the state detection signal is a modulated pulse signal, and to determine the target feature corresponding to the echo signal based on the compressed signal using a preset feature extraction algorithm.

[0037] In one specific implementation, the feature extraction submodule may include: The first feature extraction unit is used to determine the first-arrival path time of the echo signal based on the filtered signal using a first feature extraction algorithm; the first feature extraction algorithm includes a matched filtering algorithm and a leading edge detection algorithm. The second feature extraction unit is used to perform time-frequency analysis on the filtered signal using a second feature extraction algorithm to determine the micro-Doppler features corresponding to the echo signal.

[0038] In one specific embodiment, the tire positioning module 13 may include: The tire positioning submodule is used to determine the spatial point cloud of the reflection point corresponding to each echo signal based on the target feature vector using the target inference model, and to determine the tire position corresponding to each target tire based on the vehicle coordinate system corresponding to the target vehicle and the spatial point cloud of the reflection point. The tire pressure determination submodule is used to determine the vibration mode of the target tire based on the driving state of the target vehicle using the target inference model, and to determine the tire pressure state of the target tire based on the tire position corresponding to the target tire, the vibration mode and the target feature vector.

[0039] Furthermore, embodiments of this application also disclose an electronic device, Figure 3 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.

[0040] Figure 3 This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the vehicle tire alignment method disclosed in any of the foregoing embodiments. Alternatively, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0041] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0042] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk, or optical disk, etc. The resources stored thereon can include an operating system 221, computer programs 222, etc., and the storage method can be temporary storage or permanent storage.

[0043] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the vehicle tire alignment method executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include a computer program capable of performing other specific tasks.

[0044] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned vehicle tire alignment method. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0045] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0046] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0047] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0048] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0049] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for aligning vehicle tires, characterized in that, include: The vehicle's onboard ultra-wideband radar transmits status detection signals, and the echo signals corresponding to the tire pressure sensors of each target tire of the target vehicle are obtained. The echo signal is processed using a preset feature extraction algorithm to determine the target features corresponding to the echo signal; the target features include first-arrival path time, micro-Doppler features, and echo amplitude features; Based on the target features, the target feature vector corresponding to the echo signal is obtained, and the tire position and tire pressure state of the target tire corresponding to the target feature vector are determined using the target inference model.

2. The vehicle tire positioning method according to claim 1, characterized in that, The step of processing the echo signal using a preset feature extraction algorithm to determine the target features corresponding to the echo signal includes: The echo signal is filtered using a preset noise reduction algorithm to obtain the filtered signal; The target features corresponding to the echo signal are determined based on the filtered signal using a preset feature extraction algorithm.

3. The vehicle tire positioning method according to claim 2, characterized in that, The step of determining the target features corresponding to the echo signal based on the filtered signal using a preset feature extraction algorithm includes: If the state detection signal is a modulated pulse signal, then the filtered signal is pulse compressed to obtain the corresponding compressed signal, and a preset feature extraction algorithm is used to determine the target features corresponding to the echo signal based on the compressed signal.

4. The vehicle tire positioning method according to claim 2, characterized in that, The step of determining the target features corresponding to the echo signal based on the filtered signal using a preset feature extraction algorithm includes: The first-arrival path time of the echo signal is determined based on the filtered signal using a first feature extraction algorithm; the first feature extraction algorithm includes a matched filtering algorithm and a leading edge detection algorithm. The filtered signal is subjected to time-frequency analysis using a second feature extraction algorithm to determine the micro-Doppler features corresponding to the echo signal.

5. The vehicle tire positioning method according to claim 1, characterized in that, The step of determining the tire position and tire pressure state of the target tire corresponding to the target feature vector using the target inference model includes: The target inference model is used to determine the spatial point cloud of the reflection point corresponding to each echo signal based on the target feature vector, and the tire position corresponding to each target tire is determined based on the vehicle coordinate system corresponding to the target vehicle and the spatial point cloud of the reflection point. The vibration mode of the target tire is determined based on the driving state of the target vehicle using the target inference model, and the tire pressure state of the target tire is determined based on the tire position corresponding to the target tire, the vibration mode, and the target feature vector.

6. The vehicle tire alignment method according to any one of claims 1 to 5, characterized in that, The target reasoning model includes a multilayer perceptron neural network constructed using a preset multi-task learning loss function.

7. A vehicle tire alignment device, characterized in that, include: The echo signal acquisition module is used to utilize the vehicle-mounted ultra-wideband radar of the target vehicle to transmit status detection signals and acquire the echo signals corresponding to the tire pressure sensors of each target tire of the target vehicle. The feature extraction module is used to process the echo signal using a preset feature extraction algorithm to determine the target features corresponding to the echo signal; the target features include first-arrival path time, micro-Doppler features, and echo amplitude features; The tire positioning module is used to obtain the target feature vector corresponding to the echo signal based on the target features, and to determine the tire position and tire pressure state of the target tire corresponding to the target feature vector using a target inference model.

8. The vehicle tire alignment device according to claim 7, characterized in that, The feature extraction module includes: The signal filtering submodule is used to filter the echo signal using a preset noise reduction algorithm to obtain the filtered signal; The feature extraction submodule is used to determine the target features corresponding to the echo signal based on the filtered signal using a preset feature extraction algorithm.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the vehicle tire alignment method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, Used to store a computer program, wherein the computer program, when executed by a processor, implements the vehicle tire positioning method as described in any one of claims 1 to 6.

Images