Vehicle control method, device and vehicle

Abstract

The present disclosure relates to the field of vehicle control, and provides a vehicle control method, device and vehicle. An echo signal corresponding to an obstacle detection signal is received, and a moving speed of an obstacle is determined according to the echo signal. Principal component analysis processing is performed on the echo signal to obtain a target principal component corresponding to the echo signal. The target principal component is input into a pre-trained obstacle classification model, and an obstacle type is output via obstacle classification model processing. Vehicle driving is controlled according to the moving speed of the obstacle and the obstacle type to avoid the obstacle. According to the determined moving speed of the obstacle and the obstacle type, accurate sensing and identification of the obstacle are realized, and vehicle driving is controlled to avoid the obstacle, thereby improving the safety of vehicle driving.

Description

Technical Field

[0001] This disclosure relates to the field of vehicle control, and more particularly to a vehicle control method, device, and vehicle. Background Technology

[0002] With the rapid development of vehicle technology, vehicles have become an important means of transportation in people's daily lives. To facilitate user travel and improve user experience, autonomous driving technology has emerged and is gradually maturing.

[0003] For autonomous driving technology, accurate perception of the surrounding environment is crucial to ensuring its safety and reliability. Factors such as ambient light and weather conditions can lead to lower accuracy in environmental recognition, thus creating safety hazards in autonomous driving systems. Summary of the Invention

[0004] In view of this, the purpose of this disclosure is to propose a vehicle control method, device and vehicle to solve the problem that the accuracy of environmental recognition is low due to the influence of factors such as ambient light and weather, which leads to safety hazards in autonomous driving.

[0005] To achieve the above objectives, a first aspect of this disclosure provides a vehicle control method, the method comprising:

[0006] Receive the echo signal corresponding to the obstacle detection signal, and determine the moving speed of the obstacle based on the echo signal;

[0007] Principal component analysis is performed on the echo signal to obtain the target principal components corresponding to the echo signal;

[0008] The target principal component is input into a pre-trained obstacle classification model, and after processing by the obstacle classification model, the obstacle type is output.

[0009] The vehicle is controlled to avoid obstacles based on their speed and type.

[0010] Specifically, the step of performing principal component analysis on the echo signal to obtain the target principal components corresponding to the echo signal includes:

[0011] The echo signal is converted from analog to digital to obtain the digital signal corresponding to the echo signal;

[0012] Determine the correlation coefficient matrix corresponding to the digital signal, and determine the projection matrix based on the correlation coefficient matrix;

[0013] The digital signal is reduced in dimensionality using the projection matrix to obtain the target principal component corresponding to the echo signal.

[0014] Specifically, determining the correlation coefficient matrix corresponding to the digital signal includes:

[0015] Determine the initial signal data matrix corresponding to the digital signal;

[0016] The initial signal data matrix is ​​preprocessed to obtain the signal data matrix;

[0017] Correlation calculation is performed on the signal data matrix to obtain the correlation coefficient matrix corresponding to the digital signal.

[0018] Specifically, determining the projection matrix based on the correlation coefficient matrix includes:

[0019] Determine multiple eigenvalues ​​corresponding to the correlation coefficient matrix and the eigenvector corresponding to each eigenvalue;

[0020] Sort all feature values ​​in descending order;

[0021] Obtain the preset dimension, select the feature values ​​that are in the first preset dimension as the target feature values, and take the feature vector corresponding to each target feature value as the target feature vector;

[0022] Construct a projection matrix based on all target feature vectors.

[0023] The above scheme ensures accurate signal conversion and transmission by performing analog-to-digital conversion on the echo signal. Simultaneously, by determining the projection matrix of the digital signal corresponding to the echo signal, dimensionality reduction is achieved using this matrix, revealing key feature information of the obstacle, i.e., the target principal components. Consequently, the data subsequently input into the obstacle classification model is more concise and clear, reducing interference from irrelevant features and improving the classification accuracy of the model.

[0024] Specifically, determining the moving speed of the obstacle based on the echo signal includes:

[0025] The echo signal is converted from analog to digital to obtain the digital signal corresponding to the echo signal;

[0026] The digital signal is windowed to obtain the signal to be processed;

[0027] Perform a Fourier transform on the signal to be processed to extract the Doppler frequency shift data;

[0028] The wavelength and center frequency of the obstacle detection signal are obtained, and the moving speed of the obstacle is determined based on the wavelength, the center frequency, and the Doppler frequency shift data.

[0029] The above scheme reduces spectral leakage by windowing the digital signal. Simultaneously, the parallel computing Fourier transform algorithm improves the extraction accuracy and computational efficiency of Doppler frequency shift data compared to sequential Fourier transform calculations, providing a reliable data foundation for subsequent calculations of obstacle movement speed.

[0030] Specifically, controlling the vehicle's movement based on the obstacle's moving speed and type to avoid the obstacle includes:

[0031] Obtain the time delay corresponding to the echo signal, and determine the relative distance between the obstacle and the vehicle based on the time delay;

[0032] The angle between the obstacle detection signal and the direction of obstacle movement is determined based on the Doppler frequency shift data.

[0033] The target driving state of the vehicle is determined based on the moving speed of the obstacle, the type of obstacle, the relative distance, and the included angle. The vehicle is then controlled to drive according to the target driving state to avoid the obstacle.

[0034] Specifically, the training process of the obstacle classification model includes:

[0035] Obtain a preset principal component dataset and an initial obstacle classification model. The preset principal component dataset includes a training principal component dataset, which includes training principal component data and a first actual obstacle type.

[0036] The initial obstacle classification model is trained using the training principal component dataset to obtain the training obstacle types;

[0037] The regularization objective function is determined based on the training obstacle type and the first actual obstacle type;

[0038] The training of the initial obstacle classification model is completed when the regularization objective function is minimized, and the obstacle classification model is obtained.

[0039] The above scheme introduces a regularization term when determining the regularization objective function based on the type of training obstacle and the first actual obstacle type. By introducing the regularization term, the complexity of the obstacle classification model is constrained, overfitting of the obstacle classification model is prevented, and the generalization ability of the model is improved.

[0040] Specifically, the preset principal component dataset further includes a validation principal component dataset, which includes validation principal component data and a second actual obstacle type; after obtaining the obstacle classification model, it also includes:

[0041] The validation principal component dataset is input into the obstacle classification model, processed by the obstacle classification model, and the validation obstacle type is output.

[0042] The accuracy of the obstacle classification model is determined based on the verified obstacle type and the second actual obstacle type;

[0043] In response to the accuracy being less than a preset accuracy threshold, the model parameters of the obstacle classification model and / or the dimension corresponding to the principal component analysis are adjusted, and the obstacle classification model is retrained until the accuracy is greater than or equal to the preset accuracy threshold.

[0044] With the above approach, after obtaining the obstacle classification model, the accuracy of the obstacle classification model can be calculated by validating the principal component dataset. When the accuracy is lower than the preset accuracy threshold, the obtained obstacle classification model can be further trained so that the obstacle types obtained by the subsequent application of the obstacle classification model are more consistent with the actual obstacle types. This enables the vehicle to accurately avoid obstacles when controlling the vehicle's movement based on the obstacle type and the obstacle's moving speed, thereby improving driving safety.

[0045] Based on the same inventive concept, a second aspect of this disclosure provides a vehicle control device, comprising:

[0046] The signal receiving module is configured to receive the echo signal corresponding to the obstacle detection signal and determine the moving speed of the obstacle based on the echo signal.

[0047] The principal component determination module is configured to perform principal component analysis on the echo signal to obtain the target principal components corresponding to the echo signal;

[0048] The type determination module is configured to input the target principal component into a pre-trained obstacle classification model, process it through the obstacle classification model, and output the obstacle type.

[0049] The obstacle avoidance control module is configured to control the vehicle's movement based on the obstacle's moving speed and the obstacle's type, so as to avoid the obstacle.

[0050] Based on the same inventive concept, a third aspect of this disclosure proposes an electronic device including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor, when executing the computer program, implements the vehicle control method as described above.

[0051] Based on the same inventive concept, a fourth aspect of this disclosure provides a non-transitory computer-readable storage medium that stores computer instructions for causing a computer to perform the vehicle control method as described above.

[0052] Based on the same inventive concept, the fifth aspect of this disclosure provides a vehicle including the vehicle control device described in the second aspect, the electronic device described in the third aspect, or the storage medium described in the fourth aspect.

[0053] As can be seen from the above, this disclosure proposes a vehicle control method, device, and vehicle that receives echo signals corresponding to obstacle detection signals and determines the moving speed of the obstacle based on the echo signals. Principal component analysis (PCA) is performed on the echo signals to obtain the target principal components corresponding to the echo signals. Because PCA can convert high-dimensional data into low-dimensional data, it reduces the data dimensionality, simplifies the data in the echo signals, and obtains the key feature information in the echo signals, i.e., the target principal components. This allows for subsequent classification processing based on the target principal components to determine the obstacle type, reducing the influence of irrelevant features on the subsequent determination of the obstacle type, making the determined obstacle type more accurate. The target principal components are input into a pre-trained obstacle classification model, which processes the data and outputs the obstacle type. Because the obstacle classification model is a model pre-trained with a large amount of data, compared to manually determining the obstacle type through PCA, inputting the target principal components into the obstacle classification model and having the trained obstacle classification model identify and classify the target principal components results in a more accurate obstacle type. By accurately sensing and identifying obstacles based on their movement speed and type, the vehicle can be controlled to avoid them, thereby improving driving safety. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in this disclosure or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is a flowchart of a vehicle control method according to an embodiment of the present disclosure;

[0056] Figure 2 This is a structural block diagram of a vehicle control device according to an embodiment of the present disclosure;

[0057] Figure 3 This is a structural block diagram of a vehicle obstacle detection and avoidance system according to another embodiment of the present disclosure;

[0058] Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0060] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this disclosure should have the ordinary meaning understood by one of ordinary skill in the art to which this disclosure pertains. The terms "first," "second," and similar terms used in the embodiments of this disclosure do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0061] The following are definitions of terms used in this disclosure:

[0062] Continuous wave (CW): A continuous wave is a wave that a laser outputs in a continuous manner rather than in a pulsed manner.

[0063] Frequency Modulated Continuous Wave (FMCW): Frequency Modulated Continuous Wave (FMCW) is a method that uses frequency modulation to achieve distance resolution between signals.

[0064] Analog-to-digital converter (ADC): An analog-to-digital converter, also known as an A / D converter or simply ADC, is an electronic component that converts analog signals into digital signals.

[0065] PCA: Principal Components Analysis (PCA) is a technique that aims to transform multiple indicators into a few comprehensive indicators by using the concept of dimensionality reduction.

[0066] SVM: Support Vector Machine (SVM) is a powerful supervised learning algorithm that is widely used in classification and regression problems.

[0067] Hanning Window: The Hanning window is a window function, a special case of the raised cosine window, which is the sum of the spectra of three rectangular time windows.

[0068] Hamming Window: The Hamming window is a commonly used window function in digital signal processing. It is widely used due to its significant sidelobe attenuation in the frequency domain, especially in applications requiring sidelobe suppression.

[0069] With the rapid development of vehicle technology, vehicles have become an important means of transportation in people's daily lives. To facilitate user travel and improve user experience, autonomous driving technology has emerged and is gradually maturing.

[0070] For autonomous driving technology, accurate perception of the surrounding environment is crucial to ensuring its safety and reliability. In obstacle detection, traditional visual sensors, LiDAR, and millimeter-wave radar each have their advantages and disadvantages. While visual sensors can provide rich image information, they are greatly affected by lighting conditions and have poor adaptability to concealed obstacles and inclement weather. LiDAR is known for its high precision and high resolution, but it still faces challenges in terms of cost, data processing complexity, and performance under adverse weather conditions. Millimeter-wave radar, on the other hand, has been widely used in the field of autonomous driving due to its advantages such as low cost, small size, strong penetration, and all-weather operation. However, traditional millimeter-wave radar has limitations in identifying obstacle types, sizes, and precise locations, making it difficult to meet the high-precision detection requirements in complex environments.

[0071] In other words, due to factors such as ambient light and weather, the accuracy of environmental recognition is low, which in turn poses safety risks to autonomous driving.

[0072] Based on the above description, this embodiment proposes a vehicle control method, such as... Figure 1 As shown, the method includes:

[0073] Step 101: Receive the echo signal corresponding to the obstacle detection signal, and determine the moving speed of the obstacle based on the echo signal.

[0074] In practice, the vehicle includes a radar signal transmitting unit and a signal receiving unit. The radar signal transmitting unit can generate and transmit obstacle detection signals for detecting obstacles. These obstacle detection signals are electromagnetic wave signals with specific frequencies and waveforms. According to user needs, parameters such as the frequency, power, and beam shape of the obstacle detection signals can be flexibly adjusted to achieve effective detection of obstacles within different distances and angles.

[0075] Specifically, the radar signal transmitting unit transmits obstacle detection signals according to preset parameters. The type of obstacle detection signal can be continuous wave (CW) or frequency modulated continuous wave (FMCW).

[0076] Obstacle detection signals propagate through space and are reflected upon encountering obstacles, forming echo signals. The signal receiving unit captures these echo signals, which contain information such as the obstacle's distance and speed. The echo signals are then processed for identification and extraction to determine the obstacle's moving speed.

[0077] Step 102: Perform principal component analysis on the echo signal to obtain the target principal components corresponding to the echo signal.

[0078] In practice, principal component analysis (PCA) is performed on the echo signal to obtain the target principal components corresponding to the echo signal. PCA is used to reduce the dimensionality of the data while retaining the main information in the signal, which is the most critical feature for obstacle identification. Through PCA, the original high-dimensional radar signal data is converted into a series of low-dimensional principal components, which represent the main variation patterns in the signal.

[0079] Step 103: Input the target principal component into the pre-trained obstacle classification model, process it through the obstacle classification model, and output the obstacle type.

[0080] In specific implementation, a pre-trained obstacle classification model is obtained. The obstacle classification model is preferably a support vector machine model. The support vector machine model is a powerful classifier, which is particularly suitable for processing high-dimensional data and small sample situations. It achieves classification by finding an optimal hyperplane in the feature space to separate data of different categories.

[0081] The target principal components obtained through principal component analysis are input into the obstacle classification model. The model processes the data and outputs the obstacle type corresponding to each obstacle. The obstacle type represents the classification category to which the obstacle belongs. The obstacle type includes static obstacles and dynamic obstacles. Static obstacles are those that remain stationary for a relatively long period, while dynamic obstacles are those that move over a relatively long period. The "relatively long period" is a preset time frame; to accurately distinguish between static and dynamic obstacles, the "relatively long period" can be six months, a year, or even longer.

[0082] For example, the static obstacle includes at least one of the following: walls, utility poles, trees, curbs, etc. The dynamic obstacle includes at least one of the following: vehicles, pedestrians, pets, birds, etc.

[0083] Step 104: Control the vehicle's movement according to the obstacle's speed and type to avoid the obstacle.

[0084] In practice, different types of obstacles have different common movement characteristics. For example, moving vehicles generally travel along lanes and typically maintain a constant speed for a certain period of time. Pedestrians, on the other hand, move more freely, resulting in more complex routes. Therefore, vehicle movement can be controlled based on the obstacle's speed and type to avoid it.

[0085] For example, if the obstacle is a moving vehicle, the distance between the obstacle and the vehicle is determined based on the obstacle's moving speed, and the vehicle's speed is controlled based on the distance to ensure that the distance between the vehicle and the obstacle is greater than a preset safe distance.

[0086] In another example, if the obstacle is a moving pedestrian, assuming the pedestrian is in front of the vehicle and is crossing the road in front of the vehicle, the in-vehicle camera will monitor the pedestrian in real time to obtain the distance between the pedestrian and the vehicle. Based on the distance between the pedestrian and the vehicle and the pedestrian's moving speed, the vehicle's target moving speed will be determined, and the vehicle will be controlled to travel at the target moving speed to avoid a collision with the pedestrian.

[0087] The above scheme receives the echo signal corresponding to the obstacle detection signal and determines the obstacle's moving speed based on the echo signal. Principal component analysis (PCA) is performed on the echo signal to obtain the target principal components. Since PCA can convert high-dimensional data into low-dimensional data, it reduces the data dimensionality, simplifies the data in the echo signal, and obtains the key feature information in the echo signal, i.e., the target principal components. This is used for subsequent classification based on the target principal components to determine the obstacle type, reducing the influence of irrelevant features on the subsequent obstacle type determination, resulting in a more accurate obstacle type. The target principal components are input into a pre-trained obstacle classification model, which processes the data and outputs the obstacle type. Because the obstacle classification model is pre-trained on a large amount of data, compared to manually determining the obstacle type through PCA, inputting the target principal components into the obstacle classification model and having the trained obstacle classification model identify and classify the target principal components results in a more accurate obstacle type. By accurately sensing and identifying obstacles based on their movement speed and type, the vehicle can be controlled to avoid them, thereby improving driving safety.

[0088] In some embodiments, when determining the target principal component corresponding to the echo signal, the digital signal corresponding to the echo signal can be determined first, and then the projection matrix corresponding to the digital signal can be determined. Dimensionality reduction is achieved using the projection matrix. Therefore, step 102 involves performing principal component analysis on the echo signal to obtain the target principal component corresponding to the echo signal, specifically including:

[0089] Step 1021: Perform analog-to-digital conversion on the echo signal to obtain the digital signal corresponding to the echo signal;

[0090] Step 1022: Determine the correlation coefficient matrix corresponding to the digital signal, and determine the projection matrix based on the correlation coefficient matrix;

[0091] Step 1023: Use the projection matrix to perform dimensionality reduction processing on the digital signal to obtain the target principal component corresponding to the echo signal.

[0092] In practice, to ensure accurate signal conversion and transmission, the echo signal is converted from analog to digital to obtain the corresponding digital signal. In this embodiment, a high-precision analog-to-digital converter is used to perform the analog-to-digital conversion of the echo signal.

[0093] The correlation coefficient matrix corresponding to the digital signal is determined, and the specific method for determining the correlation coefficient matrix includes:

[0094] Step a: Determine the initial signal data matrix corresponding to the digital signal;

[0095] Step b: Perform data preprocessing on the initial signal data matrix to obtain the signal data matrix;

[0096] Step c: Perform correlation calculation on the signal data matrix to obtain the correlation coefficient matrix corresponding to the digital signal.

[0097] Specifically, an initial signal data matrix corresponding to the digital signal is determined. The initial signal data matrix includes multiple features measured multiple times, that is, the observation values ​​of multiple samples in multiple feature dimensions.

[0098] The initial signal data matrix undergoes data preprocessing operations to obtain a signal data matrix. This data preprocessing includes steps such as data cleaning, noise reduction, and normalization. Correlation calculations are then performed on the signal data matrix to obtain the correlation coefficient matrix corresponding to the digital signal.

[0099] In this embodiment, the specific process of performing correlation calculation on the signal data matrix to obtain the correlation coefficient matrix corresponding to the digital signal includes:

[0100] First, the covariance matrix corresponding to the signal data matrix is ​​calculated. The covariance matrix is ​​a matrix composed of the covariances between random variables. After obtaining the covariance matrix, the corresponding correlation coefficient matrix is ​​calculated based on the covariance matrix. The correlation coefficient matrix is ​​used to statistically determine the degree of linear correlation between two random variables. The correlation coefficient matrix represents the Pearson correlation coefficient between the two variables, defined as the product of the covariances of the two variables and their standard deviations.

[0101] After determining the correlation coefficient matrix, the projection matrix is ​​determined based on the correlation coefficient matrix. The specific method for determining the projection matrix includes:

[0102] Step A: Determine multiple eigenvalues ​​corresponding to the correlation coefficient matrix and the eigenvector corresponding to each eigenvalue;

[0103] Step B: Sort all feature values ​​in descending order;

[0104] Step C: Obtain the preset dimension, select the feature values ​​that are in the first preset dimension as the target feature values, and take the feature vector corresponding to each target feature value as the target feature vector;

[0105] Step D: Construct a projection matrix based on all target feature vectors.

[0106] In practice, the correlation coefficient matrix is ​​decomposed into eigenvalues ​​to obtain multiple eigenvalues ​​corresponding to the correlation coefficient matrix and an eigenvector corresponding to each of the multiple eigenvalues.

[0107] All feature values ​​are counted and sorted in descending order. A preset dimension is obtained, which is the dimension after dimensionality reduction. Feature values ​​that rank first in the preset dimension from the sorted feature values ​​are selected as target feature values. The feature vector corresponding to each target feature value is then used as the target feature vector, and a projection matrix is ​​constructed based on all target feature vectors.

[0108] In this embodiment, the preset dimension can be determined by setting a dimension, or by retaining a certain percentage of variance (such as 95%) or by using the elbow rule of eigenvalues.

[0109] After constructing the projection matrix, the digital signal is reduced in dimension using the projection matrix. That is, the signal data matrix is ​​projected onto a lower-dimensional space according to the projection matrix to obtain the number of principal components after dimension reduction. The number of principal components is the target principal component corresponding to the echo signal.

[0110] For example, assuming the initial signal data matrix is ​​X, a specific example of the principal component analysis process is as follows:

[0111] The initial signal data matrix contains multiple features from multiple measurements. The initial signal data matrix is ​​standardized so that the mean of each feature is 0 and the variance is 1, resulting in the signal data matrix, which is expressed by the formula:

[0112]

[0113] Where X′ is the signal data matrix, μ is the mean of the features, and σ is the standard deviation of the features.

[0114] Solve for the covariance matrix corresponding to the signal data matrix. The covariance matrix is ​​used to describe the correlation between various features in the data. The covariance matrix is ​​expressed by the formula:

[0115]

[0116] Where C is the covariance matrix and N is the number of samples.

[0117] The covariance matrix is ​​decomposed into eigenvalues ​​to obtain eigenvalues ​​and eigenvectors. The eigenvalue solving process is expressed by the following formula:

[0118] Cv i =λ i v i

[0119] Among them, v i Let λ be the eigenvector. i Let i be the eigenvalue, and i be the number of eigenvalues.

[0120] All eigenvalues ​​are sorted from largest to smallest. Based on the magnitude of the eigenvalues, the eigenvectors corresponding to the k largest eigenvalues ​​are selected as target eigenvectors. A projection matrix is ​​constructed based on all target eigenvectors. This projection matrix is ​​then used to project the signal data matrix into a lower-dimensional space, resulting in a dimensionality-reduced data matrix. This dimensionality-reduced data matrix contains the target principal components. The projection process is expressed by the following formula:

[0121] Y = X′P

[0122] Where Y is the data matrix after dimensionality reduction, and P is the projection matrix.

[0123] The above scheme ensures accurate signal conversion and transmission by performing analog-to-digital conversion on the echo signal. Simultaneously, by determining the projection matrix of the digital signal corresponding to the echo signal, dimensionality reduction is achieved using this matrix, revealing key feature information of the obstacle, i.e., the target principal components. Consequently, the data subsequently input into the obstacle classification model is more concise and clear, reducing interference from irrelevant features and improving the classification accuracy of the model.

[0124] In some embodiments, determining the moving speed of the obstacle based on the echo signal in step 101 specifically includes:

[0125] Step 1011: Perform analog-to-digital conversion on the echo signal to obtain the digital signal corresponding to the echo signal;

[0126] Step 1012: Window the digital signal to obtain the signal to be processed;

[0127] Step 1013: Perform Fourier transform on the signal to be processed to extract Doppler frequency shift data;

[0128] Step 1014: Obtain the wavelength and center frequency of the obstacle detection signal, and determine the moving speed of the obstacle based on the wavelength, the center frequency and the Doppler frequency shift data.

[0129] In practice, to ensure accurate signal conversion and transmission, the echo signal is converted from analog to digital to obtain the corresponding digital signal. In this embodiment, a high-precision analog-to-digital converter is used to perform the analog-to-digital conversion of the echo signal.

[0130] The digital signal is windowed using a preset window function to obtain the signal to be processed, thereby reducing spectral leakage. In this embodiment, a window function such as the Hanning window or the Hamming window is used for windowing. The form of the window function is expressed by the formula:

[0131]

[0132] Where w(n) is the window function, n is the sample index, and N is the window length.

[0133] Because the Fourier transform algorithm is inherently decomposable and suitable for parallel processing, parallel computing techniques (such as multithreading and GPU acceleration) are used to accelerate the Fourier transform operation. The signal to be processed is subjected to a Fourier transform to extract the Doppler frequency shift data.

[0134] The Doppler effect causes a change in the frequency of the radar signal, and this change is proportional to the relative velocity of the obstacle. The wavelength and center frequency of the obstacle detection signal are obtained, and the moving speed of the obstacle is determined based on the wavelength, the center frequency, and the Doppler frequency shift data. This process can be expressed by the following formula:

[0135]

[0136] Among them, f d Here, v is the moving speed of the obstacle, λ is the wavelength of the obstacle detection signal, and f0 is the center frequency of the obstacle detection signal.

[0137] In this embodiment, to improve the accuracy of the obstacle's moving speed, the moving speed of the obstacle obtained from multiple periods of spectrum data can be averaged and the average value obtained can be used as the moving speed of the obstacle to reduce noise interference.

[0138] In this embodiment, before windowing the digital signal, data preprocessing can be performed. This preprocessing includes filtering and noise reduction, as well as signal enhancement. Filtering aims to eliminate random noise and interference components in the signal. Signal enhancement improves the signal-to-noise ratio and clarity through signal amplification, phase compensation, and other methods.

[0139] The above scheme reduces spectral leakage by windowing the digital signal. Simultaneously, the parallel computing Fourier transform algorithm improves the extraction accuracy and computational efficiency of Doppler frequency shift data compared to sequential Fourier transform calculations, providing a reliable data foundation for subsequent calculations of obstacle movement speed.

[0140] In some embodiments, step 104 specifically includes:

[0141] Step 1041: Obtain the time delay corresponding to the echo signal, and determine the relative distance between the obstacle and the vehicle based on the time delay;

[0142] Step 1042: Determine the angle between the obstacle detection signal and the obstacle's direction of motion based on the Doppler frequency shift data;

[0143] Step 1043: Determine the target driving state of the vehicle based on the moving speed of the obstacle, the type of obstacle, the relative distance and the included angle, and control the vehicle to drive according to the target driving state to avoid the obstacle.

[0144] In practice, the time delay corresponding to the echo signal is obtained, and the relative distance between the obstacle and the vehicle is calculated based on the time delay. The relative distance is expressed by the formula:

[0145]

[0146] Where d is the relative distance between the obstacle and the vehicle, c is the speed of light, and Δt is the time delay of the echo signal.

[0147] Based on the Doppler frequency shift data, the moving speed of the obstacle, and the wavelength of the obstacle detection signal, the angle between the obstacle detection signal and the direction of obstacle movement is calculated. This process is expressed by the following formula:

[0148]

[0149] Where v is the moving speed of the obstacle, and f d The data is Doppler frequency shift data, λ is the wavelength of the obstacle detection signal, and θ is the angle between the obstacle detection signal and the direction of obstacle movement.

[0150] The target driving state of the vehicle is determined based on the moving speed of the obstacle, the type of obstacle, the relative distance, and the included angle. The target driving state includes driving parameter information of the vehicle, such as the vehicle's driving speed and driving direction.

[0151] In this embodiment, a specific method for determining the target driving state of the vehicle based on the obstacle's moving speed, the obstacle type, the relative distance, and the included angle includes:

[0152] Based on the obstacle's moving speed and type, a database is searched to determine the corresponding target vehicle speed. The database pre-stores the correspondence between obstacle moving speeds, obstacle types, and target vehicle speeds. The vehicle's current speed is obtained, and the vehicle is controlled to adjust from its current speed to the target speed.

[0153] During the adjustment process, the relative distance and angle between the vehicle and the obstacle are constantly monitored. If the relative distance is less than a preset distance threshold and / or the angle is less than a preset angle threshold, it indicates that the distance between the vehicle and the obstacle is too close. In this case, the target steering wheel angle value of the vehicle is determined by searching the database based on the current relative distance and / or the current angle, and the steering wheel angle of the vehicle is adjusted from the current angle value to the target steering wheel angle value.

[0154] In this embodiment, another specific method for determining the target driving state of the vehicle based on the obstacle's moving speed, the obstacle type, the relative distance, and the included angle includes:

[0155] A pre-trained obstacle avoidance path planning model and the vehicle's current position are obtained. The obstacle's moving speed, obstacle type, and the vehicle's current position are input into the obstacle avoidance path planning model. The model processes the data and outputs a target obstacle avoidance path. The vehicle then travels according to the target obstacle avoidance path to avoid the obstacle.

[0156] Furthermore, after obtaining the target obstacle avoidance path, the relative distance and angle between the vehicle and the obstacle are first determined when the vehicle travels along the target obstacle avoidance path. If the relative distance is less than a preset distance threshold and / or the angle is less than a preset angle threshold, it indicates that the distance between the vehicle and the obstacle is too close. The target obstacle avoidance path can be adjusted according to optimization algorithms such as reinforcement learning to ensure driving safety.

[0157] In this embodiment, when controlling the vehicle's movement, intelligent obstacle avoidance decisions can be made by combining the vehicle's current motion state (such as speed, acceleration, direction, etc.) and surrounding environmental conditions (such as road type, traffic rules, etc.).

[0158] In the obstacle avoidance decision-making process, multiple aspects such as safety, efficiency, and comfort can also be considered. For example, actions that may cause passenger discomfort, such as sudden braking or sharp turns, should be avoided as much as possible. At the same time, the obstacle avoidance route that can reach the destination the fastest can be selected while ensuring safety.

[0159] Simultaneously, the vehicle control system executes corresponding obstacle avoidance maneuvers (such as steering, deceleration, or stopping). During the obstacle avoidance process, the system continuously monitors changes in the surrounding environment and the vehicle's motion status, making real-time adjustments as needed. It also collects feedback information during the obstacle avoidance process (such as obstacle avoidance effectiveness and passenger reactions) for subsequent optimization and improvement.

[0160] In some embodiments, the training process of the obstacle classification model used in step 103 specifically includes:

[0161] Step 10A: Obtain a preset principal component dataset and an initial obstacle classification model. The preset principal component dataset includes a training principal component dataset, which includes training principal component data and a first actual obstacle type.

[0162] Step 10B: Train the initial obstacle classification model using the training principal component dataset to obtain the training obstacle types;

[0163] Step 10C: Determine the regularization objective function based on the training obstacle type and the first actual obstacle type;

[0164] Step 10D continues until the regularization objective function is minimized, confirming that the initial obstacle classification model training is complete, and the obstacle classification model is obtained.

[0165] In specific implementation, a preset principal component dataset and an initial obstacle classification model are obtained. The preset principal component dataset includes a training principal component dataset and a validation principal component dataset. The training principal component dataset includes training principal component data and a first actual obstacle type. The validation principal component dataset includes validation principal component data and a second actual obstacle type.

[0166] The initial obstacle classification model is trained using the training principal component dataset to obtain training obstacle types. A regularization objective function is determined based on the training obstacle types and the first actual obstacle types. The regularization parameter is a penalty coefficient in the objective function, used to balance the classification margin and misclassified samples. Specifically, a larger regularization parameter results in a lower tolerance for error, making it more prone to overfitting. A smaller regularization parameter results in a higher tolerance for error, making it more prone to underfitting.

[0167] Specifically, the initial obstacle classification model is a support vector machine (SVM) model. This SVM model segments data of different categories by finding the optimal hyperplane in the feature space. The optimal hyperplane is defined by the normal vector w and the intercept b. For a linear SVM model, its decision function is expressed by the formula:

[0168] f(x) = sign(ω·x+b)

[0169] The optimal hyperplane is expressed by the following formula:

[0170] ω·x+b=0

[0171] Where ω is the normal vector of the hyperplane, x is the training principal component data, and b is the bias.

[0172] The regularization objective function is determined based on the type of training obstacle and the type of the first actual obstacle. The regularization objective function is expressed by the formula:

[0173]

[0174] Where, ξ i C is the regularization term, and C is the regularization parameter.

[0175] The initial obstacle classification model is trained using data from the training principal component dataset until the regularization objective function is minimized. At this point, the initial obstacle classification model training is complete, and the obstacle classification model is obtained.

[0176] The above scheme introduces a regularization term when determining the regularization objective function based on the type of training obstacle and the first actual obstacle type. By introducing the regularization term, the complexity of the obstacle classification model is constrained, overfitting of the obstacle classification model is prevented, and the generalization ability of the model is improved.

[0177] In some embodiments, the preset principal component dataset further includes a validation principal component dataset, which includes validation principal component data and a second actual obstacle type. After training the initial obstacle classification model to obtain the obstacle classification model, the obtained obstacle classification model can be validated using the validation principal component dataset to further determine whether the obstacle classification model meets the requirements. The method further includes:

[0178] Step 10a: Input the validation principal component dataset into the obstacle classification model, process it through the obstacle classification model, and output the validation obstacle type;

[0179] Step 10b: Determine the accuracy of the obstacle classification model based on the verified obstacle type and the second actual obstacle type;

[0180] Step 10c: In response to the accuracy being less than a preset accuracy threshold, adjust the model parameters of the obstacle classification model and / or the dimension corresponding to the principal component analysis, and retrain the obstacle classification model until the accuracy is greater than or equal to the preset accuracy threshold.

[0181] In practice, the validation principal component data in the validation principal component dataset is input into the obstacle classification model. After processing by the obstacle classification model, the validation obstacle type corresponding to the validation principal component data is output.

[0182] The accuracy of the obstacle classification model is determined based on the verified obstacle type and the second actual obstacle type. Specifically, a first number of verification principal component data is obtained, and a second number of verification principal component data whose verified obstacle type is the same as the second actual obstacle type is determined. The identical verification obstacle type indicates that the obstacle type output by the obstacle classification model is correct. The ratio between the second number and the first number is then calculated, and this ratio represents the accuracy of the obstacle classification model.

[0183] For example, if the first number of validating principal component data is 100, and the second number of validating principal component data that determines that the validating obstacle type is the same as the second actual obstacle type is 90, then the accuracy of the obstacle classification model is determined to be 90%.

[0184] The accuracy of the determined obstacle classification model is compared with a preset accuracy threshold. If the accuracy is less than the preset accuracy threshold, it indicates that the obstacle classification model still needs to be trained. The model parameters of the obstacle classification model are adjusted. The model parameters are the parameters in the obstacle classification model, including at least one of the following: regularization term parameters, kernel function parameters, etc.

[0185] In this embodiment, since the data input to the obstacle classification model is principal component data obtained after principal component analysis, when the accuracy of the obstacle classification model is lower than the preset accuracy threshold, in addition to adjusting the model parameters of the obstacle classification model, the preset dimension corresponding to the principal component analysis can also be adjusted, thereby causing the principal component data obtained after principal component analysis to change, and thus achieving the adjustment of the accuracy of the obstacle type output by the obstacle classification model.

[0186] Obtain a new training principal component dataset, use the new training principal component dataset to continue training the obstacle classification model, and recalculate the accuracy of the trained obstacle classification model until the accuracy is greater than or equal to the preset accuracy threshold.

[0187] With the above approach, after obtaining the obstacle classification model, the accuracy of the obstacle classification model can be calculated by validating the principal component dataset. When the accuracy is lower than the preset accuracy threshold, the obtained obstacle classification model can be further trained so that the obstacle types obtained by the subsequent application of the obstacle classification model are more consistent with the actual obstacle types. This enables the vehicle to accurately avoid obstacles when controlling the vehicle's movement based on the obstacle type and the obstacle's moving speed, thereby improving driving safety.

[0188] In this application, when determining obstacle types, a combination of principal component analysis and an obstacle classification model is used. Compared to directly inputting the echo signal corresponding to the obstacle detection signal into the obstacle classification model, the key feature information corresponding to the echo signal is extracted through principal component analysis. This reduces the amount of data input to the obstacle classification model, reduces interference from irrelevant information, and improves the accuracy of the determined obstacle type. Simultaneously, when controlling the vehicle, in addition to considering the obstacle's moving speed, the obstacle type is also considered. This results in a more accurate analysis of obstacle movement, leading to more precise vehicle control, improved vehicle safety, and reduced safety hazards.

[0189] Based on the same inventive concept, and addressing the problem that the accuracy of environmental recognition is low due to factors such as ambient light and weather, thus posing safety hazards to autonomous driving, corresponding to the above embodiments, another embodiment of this disclosure provides a vehicle control method, the method specifically including:

[0190] The receiver transmits electromagnetic wave signals, such as continuous wave (CW) or frequency modulated continuous wave (FMCW), according to preset parameters. These signals propagate in space and are reflected upon encountering obstacles, forming echo signals. The receiver captures the echo signals reflected from the obstacles and converts them into digital signals. To ensure accurate signal conversion and transmission, the receiving unit must have a high-precision analog-to-digital converter (ADC) and a stable clock source.

[0191] The received digital signal undergoes preprocessing. The preprocessing stage mainly includes steps such as filtering and noise reduction, and signal enhancement. Filtering aims to eliminate random noise and interference components in the signal; signal enhancement improves the signal-to-noise ratio and clarity through signal amplification, phase compensation, and other methods.

[0192] Based on the preprocessed signal, an improved Fast Fourier Transform (FFT) algorithm is used to extract Doppler frequency shift information. Doppler frequency shift is a frequency change phenomenon caused by the relative motion between the obstacle and the radar. By analyzing the Doppler frequency shift information, the velocity information of the obstacle can be calculated. To improve computational accuracy and efficiency, the FFT algorithm is optimized and improved, such as using a window function to reduce spectral leakage and utilizing parallel computing to accelerate FFT operations.

[0193] Specifically, window functions such as the Hanning window or Hamming window are used to window the signal to reduce spectral leakage. The basic form of the window function is:

[0194]

[0195] Where n is the sample index and N is the window length.

[0196] Parallel computing techniques (such as multithreading and GPU acceleration) can be used to accelerate FFT operations. The FFT algorithm itself is decomposable and suitable for parallel processing.

[0197] The wavelength and center frequency of the obstacle detection signal are acquired, and the moving speed of the obstacle is determined based on the wavelength, the center frequency, and the Doppler frequency shift data. The above process can be expressed by the following formula:

[0198]

[0199] Among them, f d Here, v is the moving speed of the obstacle, λ is the wavelength of the obstacle detection signal, and f0 is the center frequency of the obstacle detection signal.

[0200] A combination of principal component analysis (PCA) and obstacle classification models was used to extract and classify features from signal intensity distribution. PCA was used to reduce the dimensionality of the data while retaining the main information in the signal, namely the features most critical for obstacle identification. Through PCA processing, the original high-dimensional radar signal data was transformed into a series of low-dimensional principal components, which represent the main variation patterns in the signal.

[0201] The obstacle classification model is a Support Vector Machine (SVM) model, which classifies these principal components. SVM is a powerful classifier, particularly suitable for handling high-dimensional data and small sample sizes. It achieves classification by finding an optimal hyperplane in the feature space to separate data of different categories. In the system of this invention, the SVM is trained to recognize different types of obstacles (such as vehicles, pedestrians, trees, etc.) and makes classification decisions based on the principal components extracted by principal component analysis.

[0202] Specifically, the original high-dimensional radar signal data matrix X (containing observations of multiple samples across multiple feature dimensions) is first used as input for principal component analysis (PCA). PCA calculates the correlation coefficient matrix R, then solves for its eigenvalues ​​and eigenvectors, selecting the eigenvectors corresponding to the k largest eigenvalues ​​to form the projection matrix P. Here, k is a key parameter that determines the dimensionality after dimensionality reduction, typically determined by preserving a certain percentage of variance (e.g., 95%) or according to the elbow rule of eigenvalues. Using the projection matrix P, the original data X is projected into a low-dimensional space, yielding the dimensionality-reduced principal components Y = XP. These principal components not only reduce the dimensionality of the data but also preserve the most important variation patterns in the signal—those features most critical for obstacle identification.

[0203] Furthermore, the process of principal component analysis is described in detail, specifically including:

[0204] First, data preprocessing is performed on the original high-dimensional radar signal data matrix X, including data cleaning, noise reduction, and normalization, to ensure data quality and consistency.

[0205] Calculate the correlation coefficient matrix, specifically the correlation coefficient matrix R of the preprocessed data matrix X. This step aims to understand the correlation between the features, providing a foundation for subsequent feature selection.

[0206] By solving for the eigenvalues ​​and eigenvectors of the correlation coefficient matrix R, the main directions of change in the data can be identified. These eigenvectors will be used to construct the projection matrix P.

[0207] Construct a projection matrix P by selecting the eigenvectors corresponding to the k largest eigenvalues. The choice of k is crucial, as it determines the dimension of the reduced matrix. Typically, the value of k is determined by preserving a certain proportion of the variance (e.g., 95%) or by using the elbow rule of eigenvalues.

[0208] Finally, data projection is performed. Using the projection matrix P, the original data X is projected into a low-dimensional space, resulting in the dimensionality-reduced principal components Y = XP. These principal components not only reduce the dimensionality of the data but also preserve the most important change patterns in the signal, namely those features most critical for obstacle recognition.

[0209] Furthermore, the process of processing the obstacle classification model to obtain obstacle types is described in detail, specifically including:

[0210] Input the principal component data, and use the principal component data Y extracted by principal component analysis as input to train the obstacle classification model.

[0211] Optimal hyperplane search: Obstacle classification models separate data of different classes in the feature space by finding an optimal hyperplane (defined by the normal vector w and the intercept b). For a linear SVM, its decision function is expressed as f(x) = sign(w·x+b).

[0212] Objective function optimization, during training, utilizes the training dataset (including labeled obstacle types) to optimize the objective function. This typically involves minimizing the regularized error term to find the optimal w and b. The introduction of the regularization term helps prevent overfitting and improves the model's generalization ability.

[0213] Parameter tuning further optimizes the obstacle classification model's classification performance by adjusting its parameters (such as the penalty parameter C and kernel function parameters). These parameter adjustments need to be made based on the specific dataset and classification task.

[0214] Model training and evaluation: After parameter tuning, the obstacle classification model is trained using the training dataset, and its performance is evaluated using the validation dataset. If the model performance does not meet the requirements, the k-value of the principal component analysis or the parameters of the obstacle classification model need to be adjusted until satisfactory classification results are achieved.

[0215] After training, the obstacle classification model can make classification decisions on new radar signal data based on the principal components extracted by principal component analysis, thereby achieving rapid and accurate identification of obstacle types.

[0216] The time delay corresponding to the echo signal is obtained, and the relative distance between the obstacle and the vehicle is calculated based on the time delay. The relative distance is expressed by the formula:

[0217]

[0218] Where d is the relative distance between the obstacle and the vehicle, c is the speed of light, and Δt is the time delay of the echo signal.

[0219] Based on the Doppler frequency shift data, the moving speed of the obstacle, and the wavelength of the obstacle detection signal, the angle between the obstacle detection signal and the direction of obstacle movement is calculated. This process is expressed by the following formula:

[0220]

[0221] Where v is the moving speed of the obstacle, and f d The data is Doppler frequency shift data, λ is the wavelength of the obstacle detection signal, and θ is the angle between the obstacle detection signal and the direction of obstacle movement.

[0222] The target driving state of the vehicle is determined based on the moving speed of the obstacle, the type of obstacle, the relative distance, and the included angle. The target driving state includes driving parameter information of the vehicle, such as the vehicle's driving speed and driving direction.

[0223] In this embodiment, when controlling the vehicle's movement, intelligent obstacle avoidance decisions can be made by combining the vehicle's current motion state (such as speed, acceleration, direction, etc.) and surrounding environmental conditions (such as road type, traffic rules, etc.).

[0224] In the obstacle avoidance decision-making process, multiple aspects such as safety, efficiency, and comfort can also be considered. For example, actions that may cause passenger discomfort, such as sudden braking or sharp turns, should be avoided as much as possible. At the same time, the obstacle avoidance route that can reach the destination the fastest can be selected while ensuring safety.

[0225] Simultaneously, the vehicle control system executes corresponding obstacle avoidance maneuvers (such as steering, deceleration, or stopping). During the obstacle avoidance process, the system continuously monitors changes in the surrounding environment and the vehicle's motion status, making real-time adjustments as needed. It also collects feedback information during the obstacle avoidance process (such as obstacle avoidance effectiveness and passenger reactions) for subsequent optimization and improvement.

[0226] It should be noted that the method of this disclosure embodiment can be executed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method of this disclosure embodiment, and the multiple devices will interact with each other to complete the method described.

[0227] It should be noted that the above description describes some embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0228] Based on the same inventive concept, corresponding to any of the above-described embodiments, this disclosure also provides a vehicle control device.

[0229] refer to Figure 2 , Figure 2 The vehicle control device, as described in this embodiment, includes:

[0230] The signal receiving module 201 is configured to receive the echo signal corresponding to the obstacle detection signal and determine the moving speed of the obstacle based on the echo signal;

[0231] Principal component determination module 202 is configured to perform principal component analysis on the echo signal to obtain the target principal components corresponding to the echo signal;

[0232] The type determination module 203 is configured to input the target principal component into a pre-trained obstacle classification model, process it through the obstacle classification model, and output the obstacle type.

[0233] The obstacle avoidance control module 204 is configured to control the vehicle's movement based on the obstacle's moving speed and the obstacle's type, so as to avoid the obstacle.

[0234] In some embodiments, the principal component determination module 202 specifically includes:

[0235] An analog-to-digital conversion unit is configured to perform analog-to-digital conversion on the echo signal to obtain a digital signal corresponding to the echo signal;

[0236] The projection matrix determination unit is configured to determine the correlation coefficient matrix corresponding to the digital signal and determine the projection matrix based on the correlation coefficient matrix.

[0237] The target principal component determination unit is configured to perform dimensionality reduction processing on the digital signal using the projection matrix to obtain the target principal component corresponding to the echo signal.

[0238] In some embodiments, the projection matrix determination unit is specifically configured as follows:

[0239] Determine the initial signal data matrix corresponding to the digital signal;

[0240] The initial signal data matrix is ​​preprocessed to obtain the signal data matrix;

[0241] Correlation calculation is performed on the signal data matrix to obtain the correlation coefficient matrix corresponding to the digital signal.

[0242] In some embodiments, the projection matrix determining unit is further configured to:

[0243] Determine multiple eigenvalues ​​corresponding to the correlation coefficient matrix and the eigenvector corresponding to each eigenvalue;

[0244] Sort all feature values ​​in descending order;

[0245] Obtain the preset dimension, select the feature values ​​that are in the first preset dimension as the target feature values, and take the feature vector corresponding to each target feature value as the target feature vector;

[0246] Construct a projection matrix based on all target feature vectors.

[0247] In some embodiments, the signal receiving module 201 is specifically configured as follows:

[0248] The echo signal is converted from analog to digital to obtain the digital signal corresponding to the echo signal;

[0249] The digital signal is windowed to obtain the signal to be processed;

[0250] Perform a Fourier transform on the signal to be processed to extract the Doppler frequency shift data;

[0251] The wavelength and center frequency of the obstacle detection signal are obtained, and the moving speed of the obstacle is determined based on the wavelength, the center frequency, and the Doppler frequency shift data.

[0252] In some embodiments, the obstacle avoidance control module 204 is specifically configured as follows:

[0253] Obtain the time delay corresponding to the echo signal, and determine the relative distance between the obstacle and the vehicle based on the time delay;

[0254] The angle between the obstacle detection signal and the direction of obstacle movement is determined based on the Doppler frequency shift data.

[0255] The target driving state of the vehicle is determined based on the moving speed of the obstacle, the type of obstacle, the relative distance, and the included angle. The vehicle is then controlled to drive according to the target driving state to avoid the obstacle.

[0256] In some embodiments, the apparatus further includes a model training module, which is specifically configured to:

[0257] Obtain a preset principal component dataset and an initial obstacle classification model. The preset principal component dataset includes a training principal component dataset, which includes training principal component data and a first actual obstacle type.

[0258] The initial obstacle classification model is trained using the training principal component dataset to obtain the training obstacle types;

[0259] The regularization objective function is determined based on the training obstacle type and the first actual obstacle type;

[0260] The training of the initial obstacle classification model is completed when the regularization objective function is minimized, and the obstacle classification model is obtained.

[0261] In some embodiments, the preset principal component dataset further includes a validation principal component dataset, which includes validation principal component data and a second actual obstacle type. The model training module is further configured to:

[0262] The validation principal component dataset is input into the obstacle classification model, processed by the obstacle classification model, and the validation obstacle type is output.

[0263] The accuracy of the obstacle classification model is determined based on the verified obstacle type and the second actual obstacle type;

[0264] In response to the accuracy being less than a preset accuracy threshold, the model parameters of the obstacle classification model and / or the dimension corresponding to the principal component analysis are adjusted, and the obstacle classification model is retrained until the accuracy is greater than or equal to the preset accuracy threshold.

[0265] Based on the same inventive concept, another embodiment of this disclosure provides a vehicle obstacle detection and avoidance system, such as... Figure 3 As shown, the vehicle obstacle detection and avoidance system is connected to the vehicle control system. The vehicle obstacle detection and avoidance system includes a radar signal transmitting unit, a signal receiving unit, a signal processing unit, and an obstacle avoidance decision unit.

[0266] The radar signal transmitting unit is configured to generate and transmit electromagnetic wave signals of specific frequencies and waveforms. Depending on application requirements, parameters such as the frequency, power, and beam shape of the transmitted signal can be flexibly adjusted to achieve effective detection of obstacles within different distances and angles.

[0267] The signal receiving unit is configured to receive electromagnetic wave signals reflected from obstacles. The received signals contain crucial information such as the obstacle's position, speed, shape, and material. To ensure signal integrity and accuracy, the receiving unit must possess high sensitivity and low noise characteristics, and effectively suppress clutter interference.

[0268] Signal Processing Unit: This unit is the core of the system, configured to preprocess, extract features, and classify the received radar signals. The preprocessing stage mainly includes filtering and noise reduction, signal enhancement, and other steps to improve signal quality. The feature extraction stage utilizes advanced signal processing algorithms (such as Fast Fourier Transform and Wavelet Transform) to extract key feature information from the signal. The classification stage combines machine learning algorithms (such as Support Vector Machines and Neural Networks) to classify and identify the extracted features to distinguish different types of obstacles.

[0269] The obstacle avoidance decision unit is configured to calculate the optimal obstacle avoidance path in real time using optimization algorithms based on obstacle information provided by the signal processing unit, combined with the vehicle's current motion state and environmental conditions, and then send commands to the vehicle control system to execute the obstacle avoidance action. To improve the system's intelligence level, the obstacle avoidance decision unit can also incorporate reinforcement learning algorithms to optimize the obstacle avoidance strategy through continuous trial and error and learning.

[0270] For ease of description, the above apparatus is described in terms of its functions, divided into various modules. Of course, in implementing this disclosure, the functions of each module can be implemented in one or more software and / or hardware.

[0271] The apparatus of the above embodiments is used to implement the corresponding vehicle control method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0272] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle control method described in any of the above embodiments.

[0273] Figure 4 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0274] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0275] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0276] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0277] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0278] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0279] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0280] The electronic devices described above are used to implement the corresponding vehicle control methods in any of the foregoing embodiments and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0281] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides a non-transitory computer-readable storage medium that stores computer instructions for causing the computer to execute the vehicle control method as described in any of the above embodiments.

[0282] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0283] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the vehicle control method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0284] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a vehicle, including the vehicle control device, the electronic device, and the computer-readable storage medium in the above embodiments, wherein the vehicle device implements the vehicle control method described in any of the above embodiments.

[0285] The vehicles described in the above embodiments are used to implement the vehicle control method described in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0286] It is understood that before using the technical solutions of the various embodiments in this disclosure, users will be informed of the type, scope of use, and usage scenarios of the personal information involved in an appropriate manner, and user authorization will be obtained.

[0287] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose, based on the prompt message, whether to provide personal information to the software or hardware such as electronic devices, applications, servers, or storage media performing the operations of this disclosed technical solution.

[0288] As an optional but not limited implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0289] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0290] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this disclosure (including the claims) is limited to these examples; within the framework of this disclosure, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this disclosure as described above, which are not provided in detail for the sake of brevity.

[0291] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this disclosure, the provided drawings may or may not show well-known power / ground connections to integrated circuit (IC) chips and other components. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this disclosure, and this also takes into account the fact that the details of implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this disclosure will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuitry) have been set forth to describe exemplary embodiments of this disclosure, it will be apparent to those skilled in the art that the embodiments of this disclosure may be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0292] Although this disclosure has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0293] This disclosure is intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A vehicle control method characterized by, The method comprises the following steps: receiving an echo signal corresponding to an obstacle detection signal, determining a moving speed of the obstacle according to the echo signal; performing principal component analysis on the echo signal to obtain a target principal component corresponding to the echo signal; inputting the target principal component into a pre-trained obstacle classification model, performing processing via the obstacle classification model, and outputting an obstacle type; controlling vehicle driving according to the moving speed of the obstacle and the obstacle type to avoid the obstacle; determining the moving speed of the obstacle according to the echo signal comprises: performing analog-to-digital conversion on the echo signal to obtain a digital signal corresponding to the echo signal; performing windowing processing on the digital signal to obtain a to-be-processed signal; performing Fourier transform on the to-be-processed signal to extract Doppler shift data; obtaining a wavelength and a center frequency of the obstacle detection signal, and determining the moving speed of the obstacle according to the wavelength, the center frequency and the Doppler shift data.

2. The method of claim 1, wherein, The principal component analysis processing on the echo signal to obtain the target principal component corresponding to the echo signal comprises: performing analog-to-digital conversion on the echo signal to obtain a digital signal corresponding to the echo signal; determining a correlation coefficient matrix corresponding to the digital signal, and determining a projection matrix according to the correlation coefficient matrix; performing dimension reduction processing on the digital signal by using the projection matrix to obtain the target principal component corresponding to the echo signal.

3. The method of claim 2, wherein, The determination of the correlation coefficient matrix corresponding to the digital signal comprises: determining an initial signal data matrix corresponding to the digital signal; performing data preprocessing on the initial signal data matrix to obtain a signal data matrix; performing correlation calculation on the signal data matrix to obtain the correlation coefficient matrix corresponding to the digital signal.

4. The method of claim 2, wherein, The determination of the projection matrix according to the correlation coefficient matrix comprises: determining a plurality of eigenvalues corresponding to the correlation coefficient matrix and an eigenvector corresponding to each eigenvalue; sorting all eigenvalues in descending order; obtaining a preset dimension, selecting eigenvalues located in the front preset dimension as target eigenvalues, and selecting an eigenvector corresponding to each target eigenvalue as a target eigenvector; constructing a projection matrix according to all target eigenvectors.

5. The method of claim 1, wherein, The control of vehicle driving according to the moving speed of the obstacle and the obstacle type to avoid the obstacle comprises: obtaining a time delay corresponding to the echo signal, and determining a relative distance between the obstacle and the vehicle according to the time delay; determining an included angle between the obstacle detection signal and a movement direction of the obstacle according to the Doppler shift data; determining a target driving state of the vehicle according to the moving speed of the obstacle, the obstacle type, the relative distance and the included angle, and controlling vehicle driving according to the target driving state to avoid the obstacle.

6. The method of claim 1, wherein, The training process of the obstacle classification model comprises: obtaining a preset principal component data set and an initial obstacle classification model, the preset principal component data set comprising a training principal component data set, the training principal component data set comprising training principal component data and a first actual obstacle type; training the initial obstacle classification model by using the training principal component dataset to obtain a training obstacle type; determining a regularization objective function according to the training obstacle type and the first actual obstacle type; until the regularization objective function is minimum, determining that the initial obstacle classification model is trained to obtain an obstacle classification model.

7. The method of claim 6, wherein, The preset principal component dataset further includes a verification principal component dataset, and the verification principal component dataset includes verification principal component data and a second actual obstacle type; after obtaining the obstacle classification model, further comprising: inputting the verification principal component dataset into the obstacle classification model, processing via the obstacle classification model, and outputting a verification obstacle type; determining an accuracy rate of the obstacle classification model according to the verification obstacle type and the second actual obstacle type; in response to the accuracy rate being less than a preset accuracy rate threshold, adjusting model parameters of the obstacle classification model and / or a corresponding dimension of principal component analysis processing, retraining the obstacle classification model until the accuracy rate is greater than or equal to the preset accuracy rate threshold.

8. An electronic device, comprising: The electronic device comprises a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the program to implement the method of any one of claims 1 to 7.

9. A vehicle characterized by comprising: The vehicle comprises the electronic device of claim 8.

Images