Vehicle Motion State Detection Method, Device and Medium Based on Millimeter-Wave Radar

Abstract

The present invention proposes a method, device, and medium for detecting vehicle motion status based on millimeter-wave radar, relating to the field of vehicle safety technology. The method comprises: obtaining an echo signal transmitted back by a vehicle-mounted millimeter-wave radar; filtering the echo signal to remove abnormal values in the echo signal to obtain a filtered echo signal; extracting a coordinate transformation signal from the echo signal based on the return time node, performing a discrete wavelet transform on the coordinate transformation signal to extract high-frequency and low-frequency components from the vehicle coordinate parameters, and determining the coordinate transformation speed in the coordinate direction based on the high-frequency and low-frequency components; determining a correction parameter based on the vehicle's driving scenario, correcting the vehicle speed parameter to obtain the vehicle's current actual speed, and determining the vehicle's motion status based on the vehicle's current actual speed. The present invention has the advantages of high precision, low computational complexity, and high robustness, and solves the problem of certain errors in millimeter-wave radar measurement of vehicle motion status.

Description

Technical Field

[0001] The present invention relates to the field of vehicle safety technology, and in particular to a vehicle motion state detection method, device and medium based on millimeter wave radar. Background Art

[0002] In the field of vehicle safety, millimeter-wave radar (MMW Radar) can provide precise coordinates and speed information of objects around the vehicle. This information is crucial for realizing vehicle adaptive cruise control, collision warning, automatic emergency braking and other functions. Therefore, millimeter-wave radar is widely used in vehicle assisted driving systems due to its excellent environmental adaptability and measurement accuracy.

[0003] However, millimeter-wave radars exhibit certain errors when measuring vehicle speed. These errors can be caused by factors such as the radar's limited dynamic range, the instability of radar wave reflection intensity, and interference from obstacles in the surrounding environment. Furthermore, the dynamic nature of vehicles in motion places higher demands on the processing and analysis of radar measurement data. Existing methods for processing vehicle motion states based on millimeter-wave radars still have limitations in terms of processing speed and computational complexity.

[0004] In order to improve the accuracy of vehicle motion state detection, it is necessary to provide a new vehicle motion state detection method based on millimeter wave radar to ensure the accuracy and reliability of the vehicle warning system. Summary of the Invention

[0005] In view of this, the present invention proposes a vehicle motion state detection method, device and medium based on millimeter-wave radar to solve the problem that the current vehicle motion state processing method based on millimeter-wave radar still has limitations in processing speed and computational complexity.

[0006] The technical solution of the present invention is achieved as follows:

[0007] According to a first aspect, an embodiment of the present invention provides a method for detecting a vehicle motion state based on a millimeter-wave radar, the method comprising:

[0008] Acquire an echo signal transmitted back by a vehicle-mounted millimeter-wave radar; the echo signal includes a vehicle speed parameter, a vehicle coordinate parameter, a radar cross-sectional area parameter, and a transmission time node of the vehicle speed parameter, vehicle coordinate parameter, and radar cross-sectional area parameter;

[0009] The echo signal is filtered according to the vehicle speed parameter, the vehicle coordinate parameter and the radar reflection cross-sectional area parameter to filter out abnormal values in the echo signal and obtain a filtered echo signal;

[0010] extracting a coordinate transformation signal from the echo signal according to a return time node, performing a discrete wavelet transform on the coordinate transformation signal, extracting high-frequency and low-frequency components from the vehicle coordinate parameters, and determining a coordinate transformation speed in a coordinate direction based on the high-frequency and low-frequency components; the coordinate transformation signal includes the vehicle coordinate parameters of consecutive adjacent frames, the vehicle coordinate parameters of a preceding frame in the coordinate transformation signal differ from the vehicle coordinate parameters of a last frame by a first preset number of frames, and the vehicle coordinate parameters of the last frame in the coordinate transformation signal are the vehicle coordinate parameters of a frame in the echo signal that is closest to the current time node, and the value of the first preset number of frames is determined based on a driving scenario of the vehicle;

[0011] The correction parameters are determined according to the driving scenario of the vehicle, and the vehicle speed parameters are corrected using the coordinate transformation speed and the correction parameters to obtain the current actual speed of the vehicle at the current time node, and the vehicle motion state is determined according to the current actual speed.

[0012] In combination with the first aspect, in a first implementation of the first aspect, the correction parameter is determined according to a driving scenario of the vehicle;

[0013] Accordingly, the method of determining the correction parameter according to the driving scene of the vehicle, correcting the vehicle speed parameter using the coordinate transformation speed and the correction parameter to obtain the current actual speed of the vehicle at the current time node, and determining the vehicle motion state according to the current actual speed specifically includes:

[0014] Determine the vehicle's driving scenario;

[0015] According to the driving scenario, a first correction parameter and a second correction parameter are matched to the coordinate transformation speed and the vehicle speed parameter, respectively; a mapping relationship is formed between the first correction parameter of the coordinate transformation speed, the second correction parameter of the vehicle speed parameter, and the driving scenario;

[0016] The coordinate transformation speed and the vehicle speed parameter are weighted according to the first correction parameter and the second correction parameter to obtain the current actual speed, and the vehicle motion state is determined according to the direction and magnitude of the vehicle longitudinal speed in the current actual speed.

[0017] In combination with the first embodiment of the first aspect, in the second embodiment of the first aspect, the correction parameter is determined by:

[0018] Acquire a historical echo signal transmitted back by a vehicle-mounted millimeter-wave radar, and establish a first queue to store historical vehicle speed parameters; the historical echo signal includes a historical vehicle speed parameter, a historical vehicle coordinate parameter, and a time node at which the historical vehicle speed parameter and the historical vehicle coordinate parameter were transmitted back; the first queue stores the historical vehicle speed parameters of consecutive adjacent frames; the historical vehicle speed parameter of the first frame in the first queue differs from the historical vehicle speed parameter of the last frame by a second preset number of frames, and the historical vehicle speed parameter of the last frame in the first queue is the historical vehicle speed parameter of the last frame transmitted back in the historical echo signal;

[0019] determining a driving scenario of the vehicle, and establishing an association between the first queue and the driving scenario;

[0020] The historical actual speed of the vehicle is obtained, and the correction parameter is obtained by fitting according to the association between the first queue and the driving scene and the historical actual speed.

[0021] In combination with the first aspect, in a third embodiment of the first aspect, filtering the echo signal according to the vehicle speed parameter, the vehicle coordinate parameter, and the radar reflection cross-sectional area parameter to filter out abnormal values in the echo signal to obtain a filtered echo signal specifically includes:

[0022] Determine the angle and position of the vehicle relative to the onboard millimeter-wave radar based on the vehicle speed and coordinate parameters;

[0023] Determine the adaptive threshold according to the radar reflection cross-sectional area, angle information and position information;

[0024] Adaptive threshold is used to filter out abnormal values in the echo signal to obtain a filtered echo signal.

[0025] In combination with the first aspect, in a fourth implementation of the first aspect, the method further includes the following steps:

[0026] Obtain the corresponding warning interval based on driving scenario matching;

[0027] Based on the direction and magnitude of the vehicle's longitudinal speed, the warning interval of the vehicle speed is determined, and warning information of corresponding levels is generated according to the warning interval; different warning intervals correspond to different levels of warning information, and the higher the level of warning information, the more dangerous the vehicle driving is.

[0028] In combination with the fourth implementation of the first aspect, in the fifth implementation of the first aspect, the warning interval is determined by:

[0029] Obtain accident information and the driving scenario in which each accident occurred, and summarize accident information for the same driving scenario;

[0030] According to the vehicle speed information in the accident information, a corresponding warning interval is divided for each driving scenario.

[0031] In combination with the first aspect, in a sixth implementation of the first aspect, the discrete wavelet transform uses Dobesi wavelet to perform discrete wavelet transform. Accordingly, the calculation formula for the discrete wavelet transform using Dobesi wavelet is:

[0032]

[0033]

[0034] in, Indicates the low-frequency part; Represents the high frequency part, represents the first coefficient, which is the coefficient corresponding to the scaling function; represents the second coefficient, which is the coefficient corresponding to the Dobesi wavelet function; Indicates the input signal.

[0035] In combination with the sixth embodiment of the first aspect, in the seventh embodiment of the first aspect, extracting a coordinate transformation signal from the echo signal according to the return time node, performing discrete wavelet transform using the coordinate transformation signal, extracting high-frequency components and low-frequency components from the vehicle coordinate parameters, and determining the coordinate transformation speed in the coordinate direction according to the high-frequency components and the low-frequency components, specifically includes:

[0036] Extracting a coordinate transformation signal from the echo signal according to a return time node;

[0037] Using the coordinate transformation signal and Dobesi wavelet to perform discrete wavelet transformation, extracting high-frequency components and low-frequency components in the vehicle coordinate parameters;

[0038] Accumulating the high-frequency component and the low-frequency component to obtain a high-frequency total component and a low-frequency total component respectively;

[0039] The coordinate transformation acceleration is obtained by dividing the total high-frequency component by the total low-frequency component, and the coordinate transformation speed is obtained according to the coordinate transformation speed.

[0040] According to a second aspect, an embodiment of the present invention further provides a vehicle motion state detection device based on millimeter wave radar, the device comprising:

[0041] A parameter acquisition module is used to acquire the echo signal returned by the vehicle-mounted millimeter-wave radar; the echo signal includes vehicle speed parameters, vehicle coordinate parameters, radar cross-sectional area parameters, and the return time nodes of the vehicle speed parameters, vehicle coordinate parameters, and radar cross-sectional area parameters;

[0042] The target filtering module is used to filter the echo signal according to the vehicle speed parameter, the vehicle coordinate parameter and the radar reflection cross-sectional area parameter, filter out the abnormal values in the echo signal, and obtain the filtered echo signal;

[0043] a wavelet transform module for extracting a coordinate transformation signal from the echo signal according to a return time node, performing a discrete wavelet transform using the coordinate transformation signal, extracting high-frequency and low-frequency components from the vehicle coordinate parameters, and determining a coordinate transformation speed in a coordinate direction based on the high-frequency and low-frequency components; the coordinate transformation signal comprises the vehicle coordinate parameters of consecutive adjacent frames, the vehicle coordinate parameters of the first frame in the coordinate transformation signal differ from the vehicle coordinate parameters of the last frame by a first preset number of frames, and the vehicle coordinate parameters of the last frame in the coordinate transformation signal are the vehicle coordinate parameters of a frame in the echo signal that is closest to the current time node, and the value of the first preset number of frames is determined based on a driving scenario of the vehicle;

[0044] The state detection module is used to determine the correction parameters according to the vehicle's driving scenario, use the coordinate transformation speed and the correction parameters to correct the vehicle speed parameters, obtain the current actual speed of the vehicle at the current time node, and determine the vehicle's motion state based on the current actual speed.

[0045] According to the third aspect, an embodiment of the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor, wherein when the processor executes the program, the steps of the vehicle motion state detection method based on millimeter-wave radar as described above are implemented.

[0046] According to a fourth aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-described methods for detecting vehicle motion states based on millimeter-wave radar.

[0047] According to the fifth aspect, an embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the vehicle motion state detection method based on millimeter wave radar as described in any one of the above items.

[0048] The vehicle motion state detection method, device, and medium based on millimeter-wave radar of the present invention have the following beneficial effects compared with the prior art:

[0049] Adaptive threshold and filtering are implemented for the target through radar reflection cross-sectional area, which can effectively filter out the presence of interference caused by trees or multipath interference, reduce the amount of calculation for subsequent calculations, improve the response speed of the system, reduce terminal information redundancy, and use the vehicle coordinate parameters of the first preset number of frames that have been recently acquired and are continuously adjacent to each other for discrete wavelet transform. The high-frequency and low-frequency components in the vehicle coordinate parameters are extracted and the coordinate transformation speed in the coordinate direction is determined based on the high-frequency and low-frequency components. By performing discrete wavelet transform on the echo signal returned to the vehicle by the millimeter-wave radar, the motion frequency of the target object can be separated and its speed can be calculated. Discrete wavelet transform can also perform signal denoising and feature extraction to further improve the performance of vehicle motion state detection. The correction parameters are then determined according to the driving scene of the vehicle for different driving scenes. The correction parameters can be adaptively adjusted to fit the changes in the environment and can respond to environmental changes in real time, such as weather, road conditions, etc., to ensure safety and stability under various conditions. The coordinate transformation speed and correction parameters are then used to correct the vehicle speed parameters, which can eliminate noise and other interference factors, obtain the current actual speed of the vehicle, and determine the vehicle motion state based on the current actual speed of the vehicle, to achieve the correction of the measured speed, and obtain more accurate speed parameters and vehicle motion state. This method has the advantages of high precision, low computational complexity, and high robustness. It solves the problem of certain errors in the millimeter-wave radar when measuring the vehicle motion state. At the same time, according to the real-time determination of the vehicle motion state, early warning information can be generated in time during the vehicle driving process and early warning can be issued, ensuring the safe driving of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0051] Figure 1 This is a flow chart of a method for detecting vehicle motion status based on millimeter-wave radar according to the present invention;

[0052] Figure 2 This is a second flow chart of the vehicle motion state detection method based on millimeter wave radar of the present invention;

[0053] Figure 3 This is a third flow chart of the vehicle motion state detection method based on millimeter wave radar of the present invention;

[0054] Figure 4A logic diagram of the vehicle motion state detection method based on millimeter wave radar of the present invention when performing vehicle motion state detection;

[0055] Figure 5 The figure shows a schematic structural diagram of a vehicle motion state detection system based on millimeter wave radar provided by the present invention;

[0056] Figure 6 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0058] In the field of vehicle safety, millimeter-wave radar can provide precise coordinates and speed information of objects around the vehicle. This information is crucial for realizing vehicle adaptive cruise control, collision warning, automatic emergency braking and other functions. Therefore, millimeter-wave radar is widely used in vehicle assisted driving systems due to its excellent environmental adaptability and measurement accuracy.

[0059] However, millimeter-wave radars can exhibit certain errors when measuring vehicle speed. These errors can be caused by factors such as the radar's dynamic range, the intensity of the radar wave's reflection, and interference from obstacles in the surrounding environment. Furthermore, the dynamic nature of vehicles while driving places higher demands on the processing and analysis of radar measurement data.

[0060] Existing methods for processing vehicle motion states based on millimeter-wave radar have proposed various data processing and analysis methods. For example, filtering algorithms are used to process radar data to reduce the impact of noise and interference, or attempts are made to improve system robustness by fusing data from multiple sensors. However, these methods still have limitations in terms of processing speed and computational complexity.

[0061] In order to improve the accuracy of vehicle motion state detection, it is necessary to provide a new vehicle motion state detection method based on millimeter wave radar to ensure the accuracy and reliability of the vehicle warning system.

[0062] The millimeter-wave radar-based vehicle motion state detection method provided in this specification can be applied to electronic devices with onboard data processing capabilities. Such electronic devices may include laptops, desktop computers, smartphones, smart wearable devices (such as virtual reality glasses and smart watches), tablet computers, and the like. Of course, the millimeter-wave radar-based vehicle motion state detection method provided in this specification can also be applied within applications running on these electronic devices. For example, the millimeter-wave radar-based vehicle motion state detection method can be applied to browsers with onboard data processing capabilities, or to processing software with onboard data processing capabilities.

[0063] See also Figure 1 , Figure 1 A flow chart of a method for detecting vehicle motion status based on millimeter-wave radar according to an embodiment of the present invention is shown. The method may include the following steps:

[0064] S101. Acquire an echo signal from a vehicle-mounted millimeter-wave radar. In this embodiment, the vehicle speed, vehicle coordinates, and radar cross-sectional area parameters returned by the vehicle-mounted millimeter-wave radar are used. Specifically, the echo signal includes the vehicle speed, vehicle coordinates, and radar cross-sectional area parameters, as well as the corresponding transmission times. The radar cross-sectional area parameter, also known as a radar signature, is an indicator of the millimeter-wave radar's ability to detect an object, as well as its motion state and size.

[0065] When detecting vehicle motion, millimeter-wave radar determines a parameter, the radar cross-section (RCS), based on the amount of electromagnetic waves reflected back along the original path after the radar wave hits the surface of an object. The RCS reflects the relative size of an object. The smaller the object, the fewer electromagnetic waves are reflected back, the smaller the radar's signal signature to the target, and the shorter the detection range. Specifically, the RCS depends on the object's ability to reflect a limited amount of radar radio frequency energy back to the millimeter-wave radar. Factors influencing this include the material of the target, its absolute size, and its reflection and incident angles.

[0066] By continuously receiving the echo signals returned by the vehicle-mounted millimeter-wave radar in real time as perception data, it provides massive data support for the subsequent detection of the vehicle's motion status.

[0067] Preferably, in this embodiment, a 77GHz FMCW radar is selected as the vehicle-mounted millimeter-wave radar and the required echo signal is transmitted through the vehicle-mounted millimeter-wave radar. The 77GHz FMCW radar emits scanning waves approximately 17 times per second, and can achieve higher accuracy in object position detection.

[0068] S102 , filtering the echo signal according to the vehicle speed parameter, the vehicle coordinate parameter, and the radar reflection cross-sectional area parameter, filtering out abnormal values in the echo signal, and obtaining a filtered echo signal.

[0069] In this embodiment, combining the radar reflection cross-sectional area parameter with the vehicle coordinate parameter and vehicle speed parameter information can effectively filter out irrelevant targets, and by performing a preliminary screening of objects based on the radar reflection cross-sectional area, not only can the impact of interference on the terminal interface be greatly reduced, but the number of objects in subsequent calculations can also be greatly reduced, thereby improving calculation efficiency.

[0070] S103: Extract a coordinate transformation signal from the filtered echo signal based on the return time node, perform a discrete wavelet transform on the coordinate transformation signal, extract high-frequency and low-frequency components from the vehicle coordinate parameters, and determine the coordinate transformation speed in the coordinate direction based on the high-frequency and low-frequency components. In this embodiment, the coordinate transformation signal includes vehicle coordinate parameters of consecutive adjacent frames. The vehicle coordinate parameters of the first frame in the coordinate transformation signal differ from the vehicle coordinate parameters of the last frame by a first preset number of frames, and the vehicle coordinate parameters of the last frame in the coordinate transformation signal are the vehicle coordinate parameters of the frame closest to the current time node in the echo signal. That is, the coordinate transformation signal is the vehicle coordinate parameters of the most recently acquired continuous echo signal that include the first preset number of frames. The value of the first preset number of frames is determined based on the driving scenario of the vehicle. The position change rate of the vehicle coordinate parameters over time is obtained through discrete wavelet transform, that is, the speed information is extracted. Then, the extracted speeds are averaged to obtain the real-time coordinate transformation speed.

[0071] This step specifically includes: first extracting a coordinate transformation signal from the echo signal according to the return time node, then using the coordinate transformation signal and adopting Dobesi wavelet to perform discrete wavelet transform to extract the high-frequency component and the low-frequency component in the vehicle coordinate parameter, then accumulating the high-frequency component and the low-frequency component to obtain the high-frequency total component and the low-frequency total component respectively, dividing the high-frequency total component by the low-frequency total component to obtain the coordinate transformation acceleration, and obtaining the coordinate transformation speed according to the coordinate transformation speed.

[0072] Wavelet Transform (WT), as an effective time-frequency analysis tool, is used in radar signal processing due to its advantages in processing non-stationary signals. The discrete wavelet transform (DWT) can decompose the signal into time-frequency components at different scales, effectively extracting local features from the signal, such as high-frequency information such as breakpoints and edges. This, in turn, extracts the high-frequency motion characteristics of the target object, providing more accurate velocity estimation than traditional methods. In this embodiment, by performing a continuous wavelet transform on the echo signal from the vehicle-mounted millimeter-wave radar, the target object's motion frequency can be separated and its velocity calculated. Furthermore, the DWT can be used for signal denoising and feature extraction, further improving the performance of subsequent vehicle motion state detection.

[0073] In this embodiment, the first preset frame number is determined based on the vehicle's driving scenario. The specific frame number of the first preset frame number needs to be determined based on the vehicle's actual driving scenario while ensuring the timeliness and accuracy of speed calculation. Since the first preset frame number is set to a large number, such as 10 frames, the time difference between the first frame and the tenth frame is large, but a more comprehensive consideration will be given in the subsequent calculation of the vehicle's current actual speed. Therefore, when the frame number is set to a large number, it is suitable for driving scenarios with smooth road conditions. Since the first preset frame number is set to a small number, such as 5 frames, the time difference between the first frame and the fifth frame is small, and the time error generated will also be smaller. Therefore, when the frame number is set to a large number, it is suitable for driving scenarios with more complex road conditions. The above-mentioned driving scenarios can be determined using various sensors on the vehicle, which will not be elaborated in more detail here.

[0074] Among them, the calculation formula of discrete wavelet transform is:

[0075] (1)

[0076] In formula (1), Represents the preset wavelet transform coefficients; Indicates the Vehicle coordinate parameters of the frame; It is Conjugate basis functions of wavelet transform of frame vehicle coordinate parameters; Represents the sequence length of the input signal, i.e., the vehicle coordinate parameters.

[0077] In order to improve the accuracy of speed measurement, in this embodiment, the Dobesi wavelet (DB4 wavelet) is used to perform discrete wavelet transform. The corresponding calculation formula for discrete wavelet transform using the Dobesi wavelet (DB4 wavelet), that is, the calculation formula for the DB4 wavelet transform basis function, is:

[0078] (2)

[0079] (3)

[0080] In formulas (2) and (3), Indicates the low-frequency part; Represents the high frequency part, represents the first coefficient, which is the coefficient corresponding to the scaling function; The second coefficient in the table is the coefficient corresponding to the Dobechy wavelet function; In this embodiment, the input signal It is obtained by translating and scaling the original vehicle coordinate parameters. This operation is used to extract the characteristics of the signal at different sizes and positions. Represents the index of the original vehicle coordinate parameters in the time domain, Represents the sampling of the original vehicle coordinate parameters at discrete time points. Based on the high-frequency part and the low-frequency part, the high-frequency component and the low-frequency component can be obtained.

[0081] The calculation formula for determining the coordinate transformation speed in the coordinate direction includes:

[0082] (4)

[0083] In formula (4), Indicates the vehicle horizontal coordinate parameter of the first preset number of frames in the coordinate transformation signal that is recently acquired and continuously adjacent to each other ( ) is the horizontal high-frequency component of the coordinate transformation obtained after discrete wavelet transform.

[0084] (5)

[0085] In formula (5), Indicates the vehicle longitudinal coordinate parameters of the first preset number of frames in the coordinate transformation signal that are recently acquired and continuously adjacent ( ) is the vertical high-frequency component of the coordinate transformation obtained after discrete wavelet transform.

[0086] and The high-frequency part, also known as the high-frequency component, is obtained by performing discrete wavelet transform on the transverse coordinate and the longitudinal coordinate of the vehicle coordinate parameter.

[0087] (6)

[0088] In formula (6), Indicates the vehicle horizontal coordinate parameter of the first preset number of frames in the coordinate transformation signal that is recently acquired and continuously adjacent to each other ( ) is the horizontal low-frequency component of the coordinate transformation obtained after discrete wavelet transform.

[0089] (7)

[0090] In formula (7), Indicates the vehicle longitudinal coordinate parameters of the first preset number of frames in the coordinate transformation signal that are recently acquired and continuously adjacent ( ) is the longitudinal low-frequency component of the coordinate transformation obtained after discrete wavelet transform.

[0091] and The low-frequency part is obtained by performing discrete wavelet transform on the horizontal coordinate and the longitudinal coordinate of the vehicle coordinate parameter, that is, the low-frequency component.

[0092] The high-frequency component obtained by the change contains information about the object's motion speed. Based on this, in this embodiment, the total high-frequency energy and the total low-frequency energy of the most recently acquired and consecutive first preset number of frames are calculated respectively, and the magnitude of the high-frequency energy relative to the low-frequency energy is calculated to obtain the acceleration to reflect the magnitude of the velocity at this time. The calculation formula for the acceleration is:

[0093] (8)

[0094] (9)

[0095] In formulas (8) and (9), It represents the lateral acceleration of the coordinate transformation obtained by dividing the total lateral high-frequency component of the coordinate transformation by the total lateral low-frequency component of the coordinate transformation; It represents the longitudinal acceleration of the coordinate transformation obtained by dividing the total longitudinal high-frequency component of the coordinate transformation by the total longitudinal low-frequency component of the coordinate transformation.

[0096] and Together they constitute the coordinate transformation acceleration. In this embodiment, after obtaining the coordinate transformation acceleration, the coordinate transformation acceleration is used to integrate the data frame time corresponding to the coordinate transformation signal, and then added to the previous corrected speed to obtain the time ( The coordinate transformation speed of the time is calculated as follows:

[0097] (10)

[0098] (11)

[0099] In formulas (10) and (11), express Coordinate transformation lateral velocity at time; express Coordinate transformation longitudinal velocity at time; express The previous moment adjacent to the moment Coordinate transformation lateral velocity at time; express The previous moment adjacent to the moment Coordinate transformation longitudinal velocity at time. and Both of these parameters have been corrected.

[0100] Among them, when the millimeter wave radar is just started, since there is no moment, in this case That is the echo signal The vehicle speed parameter corresponding to the moment, that is, the initial for The vehicle speed parameters corresponding to the moment, after which the millimeter-wave radar will continuously send back the vehicle echo signal, and use the calculated coordinate transformation acceleration to continuously correct the coordinate transformation speed of the previous moment to obtain the coordinate transformation speed of the current moment.

[0101] S103. Determine a correction parameter based on the vehicle's driving scenario, use the coordinate transformation speed and the correction parameter to correct the vehicle speed parameter to obtain the current actual speed of the vehicle at the current time node, and determine the vehicle's motion state based on the current actual speed of the vehicle.

[0102] In this embodiment, the coordinate transformation velocity and vehicle speed parameters are also fused to correct the vehicle speed parameters returned by the millimeter-wave radar. During this correction process, corresponding correction parameters are assigned to the coordinate transformation velocity and vehicle speed parameters, and these correction parameters are determined based on the vehicle's driving scenario. Since the vehicle speed parameters returned by the millimeter-wave radar are calculated using the Doppler effect, and different driving scenarios will cause different interference with the Doppler effect, matching different correction parameters to different driving scenarios can optimize the factors that interfere with the Doppler effect, resulting in a more effective correction of the speed during the subsequent fusion process, thereby obtaining a more accurate current actual vehicle speed.

[0103] The vehicle motion state detection method based on millimeter-wave radar of the present invention realizes adaptive threshold and filtering of the target through radar reflection cross-sectional area, which can effectively filter out the presence of interference caused by trees or multipath interference, reduce the amount of calculation for subsequent calculations, improve the response speed of the system, and reduce terminal information redundancy.

[0104] The vehicle coordinate parameters of the first preset number of frames that have been recently acquired and are continuously adjacent are used for discrete wavelet transform, and the high-frequency and low-frequency components in the vehicle coordinate parameters are extracted. The coordinate transformation speed in the coordinate direction is determined based on the high-frequency and low-frequency components. By performing discrete wavelet transform on the echo signal returned to the vehicle by the millimeter-wave radar, the motion frequency of the target object can be separated, and then its speed can be calculated. The discrete wavelet transform can also perform signal denoising and feature extraction to further improve the performance of vehicle motion state detection. The correction parameters are then determined based on the vehicle's driving scenario. For different driving scenarios, the correction parameters can be adaptively adjusted to fit the changes in the environment and can respond to environmental changes in real time, such as weather conditions. Air quality, road conditions, etc., ensuring safety and stability under various conditions. The coordinate transformation speed and correction parameters are then used to correct the vehicle speed parameters, which can eliminate noise and other interference factors, obtain the current actual speed of the vehicle, and determine the vehicle motion state based on the current actual speed of the vehicle, realize the correction of the measured speed, and obtain more accurate speed parameters and vehicle motion state. This method has the advantages of high precision, small computational complexity, and high robustness. It solves the problem of certain errors in the measurement of vehicle motion state by millimeter-wave radar. At the same time, according to the real-time determination of the vehicle motion state, early warning information can be generated in time during the vehicle driving process and early warning can be issued, ensuring the safe driving of the vehicle.

[0105] See also Figure 2 , the method may further comprise the following steps:

[0106] S201: Acquire the echo signal transmitted by the vehicle-mounted millimeter-wave radar. For details, refer to step S101.

[0107] S2021. Determine the angle information and position information of the vehicle relative to the on-board millimeter-wave radar based on the vehicle speed parameters and the vehicle coordinate parameters.

[0108] In this embodiment, an adaptive threshold is set using radar cross-sectional area, target coordinates, movement direction and other information. The adaptive threshold is used to filter out target locations that do not meet the conditions in the echo signal.

[0109] S2022: Determine an adaptive threshold based on the radar reflection cross-sectional area, angle information, and position information.

[0110] Due to the different target speed directions and the different distances between the target and the radar, the corresponding radar beam reflection angle / incident angle and the reflected radar beam intensity will also change, so the radar reflection cross-sectional area range corresponding to the target will also be different. In this embodiment, the adaptive threshold is a dynamically adjusted value.

[0111] Specifically, assuming that the radar cross-sectional area of a vehicle such as an ordinary car is orthogonal to the radar at a standard passing distance (2m) is ,Right now:

[0112] (12)

[0113] In formula (12), Indicates the radar reflection cross-sectional area of the vehicle at the standard passing distance.

[0114] At the same time, the vehicle's direction of motion is obtained based on the transformation relationship between the current moment and the previous moment's coordinates. Assume that the angle between the vehicle's direction of motion and the radar normal direction is , the projection of the vehicle on the normal direction of the radar is obtained as , the distance between the radar and the vehicle is , then the relationship between the radar cross-sectional area and distance when the vehicle angle is orthogonal to the radar radio frequency direction is:

[0115] (13)

[0116] In formula (13), Indicates the standard radar cross-sectional area of the vehicle.

[0117] Combined with the distance and angle information between the vehicle and the radar, the standard radar cross-sectional area of the vehicle should be:

[0118] (14)

[0119] S2023. Filter out abnormal values in the echo signal using an adaptive threshold to obtain a filtered echo signal.

[0120] Furthermore, since there are many interference factors in the process of driving, such as road conditions, which may cause fluctuations in the target radar cross-sectional area, it is necessary to set a threshold range to filter the interference and enhance the robustness of the system. Principle, filter out abnormal values in the echo signal that deviate too much from the normal value, thereby retaining targets such as vehicles and enhancing the stability of the system. The corresponding calculation formula is:

[0121] (15)

[0122] In formula (15), is the standard radar cross-sectional area of the vehicle calculated in formula (14); Indicates the preset error value, The specific value can be configured by the user according to the needs, for example, The value of is set to 1; It is the actual radar reflection cross-section obtained by millimeter-wave radar measurement.

[0123] S203: Extract a coordinate transformation signal from the filtered echo signal based on the return time node. Perform a discrete wavelet transform on the coordinate transformation signal to extract the high-frequency and low-frequency components of the vehicle coordinate parameters. Determine the coordinate transformation speed in the coordinate direction based on the high-frequency and low-frequency components. For details, refer to step S103.

[0124] S204: Determine a correction parameter based on the vehicle's driving scenario, use the coordinate transformation speed and the correction parameter to correct the vehicle's speed parameter, obtain the vehicle's current actual speed at the current time point, and determine the vehicle's motion state based on the vehicle's current actual speed. For details, refer to step S104.

[0125] See also Figure 3 , the method may further comprise the following steps:

[0126] S301: Acquire the echo signal transmitted back by the vehicle-mounted millimeter-wave radar. For details, refer to step S101.

[0127] S302: Filter the echo signal according to the vehicle speed parameter, the vehicle coordinate parameter, and the radar cross-sectional area parameter to remove abnormal values in the echo signal and obtain a filtered echo signal. For details, refer to step S102.

[0128] S303: Extract a coordinate transformation signal from the filtered echo signal based on the return time node. Perform a discrete wavelet transform on the coordinate transformation signal to extract the high-frequency and low-frequency components of the vehicle coordinate parameters. Determine the coordinate transformation speed in the coordinate direction based on the high-frequency and low-frequency components. For details, refer to step S103.

[0129] S3041. Determine the driving scenario of the vehicle. The driving scenario can be determined using various sensors on the vehicle, which will not be elaborated in more detail here.

[0130] S3042. Match the first correction parameter and the second correction parameter to the coordinate transformation speed and the vehicle speed parameter, respectively, according to the driving scenario. There is a mapping relationship between the first correction parameter of the coordinate transformation speed, the second correction parameter of the vehicle speed parameter, and the driving scenario. Rapid matching can be performed through this mapping relationship. It can be understood that the above-mentioned first and second correction parameters together constitute the correction parameter. The first and second correction parameters are weights corresponding to the coordinate transformation speed and the vehicle speed parameter, respectively. The weight represents the proportion of the corresponding parameter in the subsequent speed correction processing.

[0131] In this embodiment, the correction parameters are determined in the following manner:

[0132] Obtain the historical echo signal returned by the vehicle-mounted millimeter-wave radar, and establish a first queue to store historical vehicle speed parameters and a second queue to store historical vehicle coordinate parameters. It can be understood that the historical echo signal includes historical vehicle speed parameters, historical vehicle coordinate parameters, and the return time nodes of historical vehicle speed parameters and historical vehicle coordinate parameters. The first queue stores historical vehicle speed parameters of consecutive adjacent frames, and the historical vehicle coordinate parameters of the first frame in the first queue differ from the historical vehicle coordinate parameters of the last frame by a second preset number of frames, and the historical vehicle coordinate parameters of the last frame in the first queue are the historical vehicle coordinate parameters of the last frame returned in the historical echo signal. Similarly, the second queue stores historical vehicle coordinate parameters of consecutive adjacent frames, and the historical vehicle coordinate parameters of the first frame in the second queue differ from the historical vehicle coordinate parameters of the last frame by a second preset number of frames, and the historical vehicle coordinate parameters of the last frame in the second queue are the historical vehicle coordinate parameters of the last frame returned in the historical echo signal.

[0133] It should be noted that the value of the second preset number of frames exceeds the value of the first preset number of frames. The echo signals returned by the vehicle-mounted millimeter-wave radar are received in real time and continuously as historical data, and a corresponding queue is established to store the historical data. During the real-time acquisition of the historical echo signals, the data in the queue is continuously updated.

[0134] At the same time, when acquiring the vehicle motion state of the vehicle to be detected, the first and second queues can also be used to first store the corresponding vehicle speed parameters and vehicle coordinate parameters to facilitate analysis of the vehicle's motion trend. The vehicle coordinate parameters for the first preset number of frames that were most recently acquired and are consecutively adjacent are then extracted from the second queue. For example, if the second preset number of frames is set to 150 and the first preset number of frames is set to 5, the vehicle coordinate parameters for the first preset number of frames that were most recently acquired and are consecutively adjacent are the vehicle coordinate parameters for frames 146, 147, 148, 149, and 150.

[0135] After that, the vehicle's driving scenario is determined, and a correlation is established between the first queue and the driving scenario. Then, the vehicle's historical actual speed is obtained. Based on the correlation between the first queue and the driving scenario and the historical actual speed, the correction parameters can be fitted based on the differences between the historical actual speed under different driving scenarios and the historical vehicle speed parameters returned by the millimeter-wave radar.

[0136] Specifically, when the measured object is on both sides of the millimeter wave radar or at a long distance, the electromagnetic wave is easily interfered by the environment during transmission, and the angle will cause the phase to shift, resulting in a large error in the speed measurement through the Doppler effect. In this case, increase the first correction parameter To reduce the Doppler error, for example, set the first correction parameter When the vehicle is traveling on a relatively stable road or when the relative speed between the measured target and the millimeter-wave radar is large, the first correction parameter is reduced. To obtain a more accurate speed, for example, set the first correction parameter .

[0137] Interference factors that cause the Doppler effect include multipath interference (such as buildings, bridges, and tunnels), ground reflection (such as uneven or low-lying roads), moving object interference (such as pedestrians and bicycles), and other electromagnetic interference (such as other vehicle-mounted radars and base stations). These interference factors are ranked by interference intensity and common scenarios: multipath interference is the most severe, followed by ground reflection and moving object interference, and finally electromagnetic interference, which is the least severe. Therefore, after establishing an association between the first queue and the driving scene, an image recognition algorithm can be used to extract key targets from the driving scene, such as the aforementioned multipath interference, ground reflection, moving object interference, and electromagnetic interference targets. Based on the preset interference values for each key target, the total interference coefficient for the specific driving scenario is determined. A range of interference coefficients is also set for each driving scenario (including the maximum and minimum values for the interference coefficient for that driving scenario). Subsequently, correction parameters can be derived based on the interference coefficients and the differences between the historical actual speeds and historical vehicle speed parameters transmitted by the millimeter-wave radar for different driving scenarios.

[0138] S3043. Based on the first correction parameter and the second correction parameter, the coordinate transformation speed and the vehicle speed parameter are weighted to obtain the current actual speed of the vehicle at the current time node, and the vehicle motion state is determined based on the direction and magnitude of the vehicle longitudinal speed in the current actual speed.

[0139] Specifically, the calculation formula for correcting the vehicle speed parameter using the coordinate transformation speed and the correction parameter is:

[0140] (16)

[0141] In formula (16), Indicates the corrected lateral speed of the vehicle; represents the first correction parameter; represents the second correction parameter; Represents the vehicle's horizontal coordinate parameters.

[0142] (17)

[0143] In formula (17), represents the corrected longitudinal speed of the vehicle; Represents the vehicle's longitudinal coordinate parameter.

[0144] and Together they constitute the corrected current actual speed of the vehicle.

[0145] See also Figure 4 , the method may further comprise the following steps:

[0146] S401: Acquire the echo signal transmitted by the vehicle-mounted millimeter-wave radar. For details, refer to step S101.

[0147] S402: Filter the echo signal according to the vehicle speed parameter, the vehicle coordinate parameter, and the radar cross-sectional area parameter to remove abnormal values in the echo signal and obtain a filtered echo signal. For details, refer to step S102.

[0148] S403: Extract a coordinate transformation signal from the filtered echo signal based on the return time node. Perform a discrete wavelet transform on the coordinate transformation signal to extract the high-frequency and low-frequency components of the vehicle coordinate parameters. Determine the coordinate transformation speed in the coordinate direction based on the high-frequency and low-frequency components. For details, refer to step S103.

[0149] S404: Determine a correction parameter based on the vehicle's driving scenario, use the coordinate transformation speed and the correction parameter to correct the vehicle's speed parameter, obtain the vehicle's current actual speed at the current time point, and determine the vehicle's motion state based on the vehicle's current actual speed. For details, refer to step S104.

[0150] S405. Different levels of warning information are generated based on the vehicle's driving scenario, the direction and magnitude of the vehicle's longitudinal speed in the current actual speed. In this embodiment, by generating different levels of warning information and issuing corresponding warning reminders, timely warnings of potential collision risks are issued, thereby improving the vehicle's driving safety.

[0151] Specifically, step S405 includes:

[0152] S4051, according to the driving scene matching to obtain the corresponding warning interval, in this process, first according to the actual driving scene from the preset warning library to obtain the corresponding vehicle speed critical threshold, then according to these vehicle speed critical threshold to obtain the corresponding warning interval, for example, the first threshold is matched and the second threshold , then we can get three warning intervals, namely less than (First warning zone), to (Second warning interval) and greater than (Third warning zone).

[0153] S4052: Determine the warning interval within which the vehicle's speed falls based on the direction and magnitude of the vehicle's longitudinal speed, and generate a warning message of a corresponding level based on the warning interval. In this embodiment, different warning intervals correspond to different levels of warning messages, and higher levels of warning messages indicate a higher risk of vehicle travel.

[0154] For example, if the vehicle longitudinal velocity is detected along the negative direction of the y-axis and the magnitude of the vehicle longitudinal velocity is less than That is, if the vehicle speed is in the first warning range, no alarm will be issued; if the vehicle longitudinal speed is detected along the negative direction of the y-axis, and the vehicle longitudinal speed is greater than or equal to But less than or equal to That is, if the vehicle speed is in the second warning range, a first-level alarm is generated and a brief response is issued; if the vehicle longitudinal speed is detected along the negative direction of the y-axis and the magnitude of the vehicle longitudinal speed is greater than That is, if the vehicle speed is in the third warning range, a second-level alarm will be generated and a response will be issued continuously.

[0155] The warning interval is determined by the following method:

[0156] Obtain accident information and the driving scenario in which each accident information occurs, and summarize the accident information of the same driving scenario. When a vehicle accident occurs, the corresponding vehicle speed information will be recorded. Then, based on the vehicle speed information in the accident information, divide the corresponding warning interval for each driving scenario. It should be noted that different driving scenarios can be divided into different numbers of warning intervals, and the warning intervals and their corresponding driving scenarios are stored in a preset warning library.

[0157] The following describes a system provided by an embodiment of the present invention. The system described below and the method described above can refer to each other.

[0158] See also Figure 5 , Figure 5 A schematic diagram of the structure of a vehicle motion state detection system based on millimeter wave radar according to an embodiment of the present invention is shown. The system may include:

[0159] The parameter acquisition module 10 is used to acquire the echo signal transmitted by the vehicle's millimeter-wave radar. In this embodiment, the parameters used are the vehicle speed, vehicle coordinates, and radar cross-sectional area parameters returned by the vehicle's millimeter-wave radar. Specifically, the echo signal includes the vehicle speed, vehicle coordinates, and radar cross-sectional area parameters, as well as the corresponding transmission time points. The radar cross-sectional area parameter, also known as a radar feature, is an indicator that measures the degree to which the millimeter-wave radar can detect an object, as well as its motion state and size.

[0160] When detecting vehicle motion, millimeter-wave radar determines a parameter, the radar cross-sectional area, based on the amount of electromagnetic waves reflected back along the original path after the radar wave hits the surface of the object. The radar cross-sectional area can reflect the relative size of the object. The smaller the object, the fewer electromagnetic waves are reflected back, the smaller the radar signal characteristics to the target, and the shorter the detection range. Specifically, the size of the radar cross-sectional area depends on the object reflecting a limited amount of radar RF energy back to the millimeter-wave radar. Factors affecting this include the material of the target, the absolute size of the target, the reflection angle, and the angle of incidence.

[0161] By continuously receiving the echo signals returned by the vehicle-mounted millimeter-wave radar in real time as perception data, the echo signals are obtained and provide massive data support for the subsequent detection of the vehicle's motion status.

[0162] Preferably, in this embodiment, a 77GHz FMCW radar is selected as the vehicle-mounted millimeter-wave radar and the required echo signal is transmitted through the vehicle-mounted millimeter-wave radar. The 77GHz FMCW radar emits scanning waves approximately 17 times per second, and can achieve higher accuracy in object position detection.

[0163] The target filtering module 20 is used to filter the echo signal according to the vehicle speed parameter, the vehicle coordinate parameter and the radar reflection cross-sectional area parameter, filter out abnormal values in the echo signal, and obtain a filtered echo signal.

[0164] In this embodiment, combining the radar reflection cross-sectional area parameter with the vehicle coordinate parameter and vehicle speed parameter information can effectively filter out irrelevant targets, and by performing a preliminary screening of objects based on the radar reflection cross-sectional area, not only can the impact of interference on the terminal interface be greatly reduced, but the number of objects in subsequent calculations can also be greatly reduced, thereby improving calculation efficiency.

[0165] The wavelet transform module 30 is configured to extract a coordinate transformation signal from the filtered echo signal based on the return time node, perform a discrete wavelet transform on the coordinate transformation signal, extract high-frequency and low-frequency components from the vehicle coordinate parameters, and determine the coordinate transformation speed in the coordinate direction based on the high-frequency and low-frequency components. In this embodiment, the coordinate transformation signal includes vehicle coordinate parameters of consecutive adjacent frames. The vehicle coordinate parameters of the first frame in the coordinate transformation signal differ from the vehicle coordinate parameters of the last frame by a first preset number of frames, and the vehicle coordinate parameters of the last frame in the coordinate transformation signal are the vehicle coordinate parameters of the frame closest to the current time node in the echo signal. In other words, the coordinate transformation signal is the vehicle coordinate parameters of the most recently acquired echo signal that are continuous and include the first preset number of frames. The value of the first preset number of frames is determined based on the vehicle driving scene. The position change rate of the vehicle coordinate parameters over time is obtained through discrete wavelet transform, i.e., the speed information is extracted. The extracted speeds are then averaged to obtain the real-time coordinate transformation speed.

[0166] As an effective time-frequency analysis tool, the wavelet transform is used in radar signal processing due to its advantages in processing non-stationary signals. The discrete wavelet transform can decompose the signal into time-frequency components at different scales, effectively extracting local features from the signal, such as high-frequency information such as breakpoints and edges. This, in turn, extracts the high-frequency motion characteristics of the target object, providing a more accurate velocity estimate than traditional methods. In this embodiment, by performing a continuous wavelet transform on the echo signal of a vehicle-mounted millimeter-wave radar, the target object's motion frequency can be separated and its velocity calculated. Furthermore, the discrete wavelet transform can be used for signal denoising and feature extraction, further improving the performance of subsequent vehicle motion state detection.

[0167] In this embodiment, the first preset frame number is determined based on the vehicle's driving scenario. The specific frame number of the first preset frame number needs to be determined based on the vehicle's actual driving scenario while ensuring the timeliness and accuracy of speed calculation. Since the first preset frame number is set to a large number, such as 10 frames, the time difference between the first frame and the tenth frame is large, but a more comprehensive consideration will be given in the subsequent calculation of the vehicle's current actual speed. Therefore, when the frame number is set to a large number, it is suitable for driving scenarios with smooth road conditions. Since the first preset frame number is set to a small number, such as 5 frames, the time difference between the first frame and the fifth frame is small, and the time error generated will also be smaller. Therefore, when the frame number is set to a large number, it is suitable for driving scenarios with more complex road conditions. The above-mentioned driving scenarios can be determined using various sensors on the vehicle, which will not be elaborated in more detail here.

[0168] The state detection module 40 is used to determine the correction parameters according to the driving scene of the vehicle, use the coordinate transformation speed and the correction parameters to correct the vehicle speed parameters, obtain the current actual speed of the vehicle at the current time node, and determine the vehicle motion state according to the current actual speed of the vehicle.

[0169] In this embodiment, the coordinate transformation velocity and vehicle speed parameters are also fused to correct the vehicle speed parameters returned by the millimeter-wave radar. During this correction process, corresponding correction parameters are assigned to the coordinate transformation velocity and vehicle speed parameters, and these correction parameters are determined based on the vehicle's driving scenario. Since the vehicle speed parameters returned by the millimeter-wave radar are calculated using the Doppler effect, and different driving scenarios will cause different interference with the Doppler effect, matching different correction parameters to different driving scenarios can optimize the factors that interfere with the Doppler effect, resulting in a more effective correction of the speed during the subsequent fusion process, thereby obtaining a more accurate current actual vehicle speed.

[0170] The vehicle motion state detection device based on millimeter-wave radar of the present invention realizes adaptive threshold and filtering of the target through radar reflection cross-sectional area, which can effectively filter out the presence of interference caused by trees or multipath interference, reduce the amount of calculation for subsequent calculations, improve the response speed of the system, and reduce terminal information redundancy.

[0171] The vehicle coordinate parameters of the first preset number of frames that have been recently acquired and are continuously adjacent are used for discrete wavelet transform, and the high-frequency and low-frequency components in the vehicle coordinate parameters are extracted. The coordinate transformation speed in the coordinate direction is determined based on the high-frequency and low-frequency components. By performing discrete wavelet transform on the echo signal returned to the vehicle by the millimeter-wave radar, the motion frequency of the target object can be separated, and then its speed can be calculated. The discrete wavelet transform can also perform signal denoising and feature extraction to further improve the performance of vehicle motion state detection. The correction parameters are then determined based on the vehicle's driving scenario. For different driving scenarios, the correction parameters can be adaptively adjusted to fit the changes in the environment and can respond to environmental changes in real time, such as weather conditions. Air quality, road conditions, etc., ensuring safety and stability under various conditions. The coordinate transformation speed and correction parameters are then used to correct the vehicle speed parameters, which can eliminate noise and other interference factors, obtain the current actual speed of the vehicle, and determine the vehicle motion state based on the current actual speed of the vehicle, realize the correction of the measured speed, and obtain more accurate speed parameters and vehicle motion state. This method has the advantages of high precision, small computational complexity, and high robustness. It solves the problem of certain errors in the measurement of vehicle motion state by millimeter-wave radar. At the same time, according to the real-time determination of the vehicle motion state, early warning information can be generated in time during the vehicle driving process and early warning can be issued, ensuring the safe driving of the vehicle.

[0172] Figure 6 An example of a physical structure diagram of an electronic device is shown below. Figure 6 As shown, the electronic device may include: a processor 610 (processor), a communication interface 620 (Communications Interface), a memory 630 (memory) and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other via the communication bus 640. The processor 610 may call the logic commands in the memory 630 to execute a vehicle motion state detection method based on millimeter wave radar, which includes:

[0173] Acquire an echo signal transmitted back by a vehicle-mounted millimeter-wave radar; the echo signal includes a vehicle speed parameter, a vehicle coordinate parameter, and a time node at which the vehicle speed parameter and the vehicle coordinate parameter are transmitted back;

[0174] The echo signal is filtered according to the vehicle speed parameter, the vehicle coordinate parameter and the radar reflection cross-sectional area parameter to filter out abnormal values in the echo signal and obtain a filtered echo signal;

[0175] extracting a coordinate transformation signal from the echo signal according to a return time node, performing a discrete wavelet transform on the coordinate transformation signal, extracting high-frequency and low-frequency components from the vehicle coordinate parameters, and determining a coordinate transformation speed in a coordinate direction based on the high-frequency and low-frequency components; the coordinate transformation signal includes the vehicle coordinate parameters of consecutive adjacent frames, the vehicle coordinate parameters of a preceding frame in the coordinate transformation signal differ from the vehicle coordinate parameters of a last frame by a first preset number of frames, and the vehicle coordinate parameters of the last frame in the coordinate transformation signal are the vehicle coordinate parameters of a frame in the echo signal that is closest to the current time node, and the value of the first preset number of frames is determined based on a driving scenario of the vehicle;

[0176] The correction parameters are determined according to the driving scenario of the vehicle, and the vehicle speed parameters are corrected using the coordinate transformation speed and the correction parameters to obtain the current actual speed of the vehicle at the current time node, and the vehicle motion state is determined according to the current actual speed.

[0177] In addition, the logic instructions in the aforementioned memory 530 can be implemented in the form of a software functional unit and, when sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0178] On the other hand, the present invention further provides a computer program product, comprising a computer program stored on a non-transitory computer-readable storage medium, wherein the computer program includes program instructions. When the program instructions are executed by a computer, the computer can perform the vehicle motion state detection method based on millimeter-wave radar provided by the above methods, which method includes:

[0179] Acquire an echo signal transmitted back by a vehicle-mounted millimeter-wave radar; the echo signal includes a vehicle speed parameter, a vehicle coordinate parameter, and a time node at which the vehicle speed parameter and the vehicle coordinate parameter are transmitted back;

[0180] The echo signal is filtered according to the vehicle speed parameter, the vehicle coordinate parameter and the radar reflection cross-sectional area parameter to filter out abnormal values in the echo signal and obtain a filtered echo signal;

[0181] extracting a coordinate transformation signal from the echo signal according to a return time node, performing a discrete wavelet transform on the coordinate transformation signal, extracting high-frequency and low-frequency components from the vehicle coordinate parameters, and determining a coordinate transformation speed in a coordinate direction based on the high-frequency and low-frequency components; the coordinate transformation signal includes the vehicle coordinate parameters of consecutive adjacent frames, the vehicle coordinate parameters of a preceding frame in the coordinate transformation signal differ from the vehicle coordinate parameters of a last frame by a first preset number of frames, and the vehicle coordinate parameters of the last frame in the coordinate transformation signal are the vehicle coordinate parameters of a frame in the echo signal that is closest to the current time node, and the value of the first preset number of frames is determined based on a driving scenario of the vehicle;

[0182] The correction parameters are determined according to the driving scenario of the vehicle, and the vehicle speed parameters are corrected using the coordinate transformation speed and the correction parameters to obtain the current actual speed of the vehicle at the current time node, and the vehicle motion state is determined according to the current actual speed.

[0183] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the computer program is implemented to perform the above-mentioned method for detecting a vehicle motion state based on a millimeter-wave radar, the method comprising:

[0184] Acquire an echo signal transmitted back by a vehicle-mounted millimeter-wave radar; the echo signal includes a vehicle speed parameter, a vehicle coordinate parameter, and a time node at which the vehicle speed parameter and the vehicle coordinate parameter are transmitted back;

[0185] The echo signal is filtered according to the vehicle speed parameter, the vehicle coordinate parameter and the radar reflection cross-sectional area parameter to filter out abnormal values in the echo signal and obtain a filtered echo signal;

[0186] extracting a coordinate transformation signal from the echo signal according to a return time node, performing a discrete wavelet transform on the coordinate transformation signal, extracting high-frequency and low-frequency components from the vehicle coordinate parameters, and determining a coordinate transformation speed in a coordinate direction based on the high-frequency and low-frequency components; the coordinate transformation signal includes the vehicle coordinate parameters of consecutive adjacent frames, the vehicle coordinate parameters of a preceding frame in the coordinate transformation signal differ from the vehicle coordinate parameters of a last frame by a first preset number of frames, and the vehicle coordinate parameters of the last frame in the coordinate transformation signal are the vehicle coordinate parameters of a frame in the echo signal that is closest to the current time node, and the value of the first preset number of frames is determined based on a driving scenario of the vehicle;

[0187] The correction parameters are determined according to the driving scenario of the vehicle, and the vehicle speed parameters are corrected using the coordinate transformation speed and the correction parameters to obtain the current actual speed of the vehicle at the current time node, and the vehicle motion state is determined according to the current actual speed.

[0188] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0189] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or perform equivalent conversions on some of the technical features therein. However, these modifications or conversions do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for detecting vehicle motion status based on millimeter-wave radar, characterized in that: The method comprises: Acquire an echo signal transmitted back by a vehicle-mounted millimeter-wave radar; the echo signal includes a vehicle speed parameter, a vehicle coordinate parameter, a radar cross-sectional area parameter, and a transmission time node of the vehicle speed parameter, vehicle coordinate parameter, and radar cross-sectional area parameter; The echo signal is filtered according to the vehicle speed parameter, the vehicle coordinate parameter and the radar reflection cross-sectional area parameter to filter out abnormal values in the echo signal and obtain a filtered echo signal; extracting a coordinate transformation signal from the echo signal according to a return time node, performing a discrete wavelet transform on the coordinate transformation signal, extracting high-frequency and low-frequency components from the vehicle coordinate parameters, and determining a coordinate transformation speed in a coordinate direction based on the high-frequency and low-frequency components; the coordinate transformation signal includes the vehicle coordinate parameters of consecutive adjacent frames, the vehicle coordinate parameters of a preceding frame in the coordinate transformation signal differ from the vehicle coordinate parameters of a last frame by a first preset number of frames, and the vehicle coordinate parameters of the last frame in the coordinate transformation signal are the vehicle coordinate parameters of a frame in the echo signal that is closest to the current time node, and the value of the first preset number of frames is determined based on a driving scenario of the vehicle; The correction parameters are determined according to the driving scenario of the vehicle, and the vehicle speed parameters are corrected using the coordinate transformation speed and the correction parameters to obtain the current actual speed of the vehicle at the current time node, and the vehicle motion state is determined according to the current actual speed.

2. The vehicle motion state detection method based on millimeter wave radar according to claim 1, characterized in that: The correction parameter is determined according to the driving scenario of the vehicle; Accordingly, the method of determining the correction parameter according to the driving scene of the vehicle, correcting the vehicle speed parameter using the coordinate transformation speed and the correction parameter to obtain the current actual speed of the vehicle at the current time node, and determining the vehicle motion state according to the current actual speed specifically includes: Determine the vehicle's driving scenario; According to the driving scenario, a first correction parameter and a second correction parameter are matched to the coordinate transformation speed and the vehicle speed parameter, respectively; a mapping relationship is formed between the first correction parameter of the coordinate transformation speed, the second correction parameter of the vehicle speed parameter, and the driving scenario; The coordinate transformation speed and the vehicle speed parameter are weighted according to the first correction parameter and the second correction parameter to obtain the current actual speed, and the vehicle motion state is determined according to the direction and magnitude of the vehicle longitudinal speed in the current actual speed.

3. The vehicle motion state detection method based on millimeter wave radar according to claim 2, characterized in that: The correction parameters are determined in the following way: Acquire a historical echo signal transmitted back by a vehicle-mounted millimeter-wave radar, and establish a first queue to store historical vehicle speed parameters; the historical echo signal includes a historical vehicle speed parameter, a historical vehicle coordinate parameter, and a time node at which the historical vehicle speed parameter and the historical vehicle coordinate parameter were transmitted back; the first queue stores the historical vehicle speed parameters of consecutive adjacent frames; the historical vehicle speed parameter of the first frame in the first queue differs from the historical vehicle speed parameter of the last frame by a second preset number of frames, and the historical vehicle speed parameter of the last frame in the first queue is the historical vehicle speed parameter of the last frame transmitted back in the historical echo signal; determining a driving scenario of the vehicle, and establishing an association between the first queue and the driving scenario; The historical actual speed of the vehicle is obtained, and the correction parameter is obtained by fitting according to the association between the first queue and the driving scene and the historical actual speed.

4. The method for detecting vehicle motion status based on millimeter-wave radar according to claim 1, wherein: The echo signal is filtered according to the vehicle speed parameter, vehicle coordinate parameter and radar reflection cross-sectional area parameter to filter out abnormal values in the echo signal and obtain a filtered echo signal, specifically including: Determine the angle and position of the vehicle relative to the onboard millimeter-wave radar based on the vehicle speed and coordinate parameters; Determine the adaptive threshold according to the radar reflection cross-sectional area, angle information and position information; Adaptive threshold is used to filter out abnormal values in the echo signal to obtain a filtered echo signal.

5. The vehicle motion state detection method based on millimeter wave radar according to claim 1, characterized in that: The method further comprises the following steps: Obtain the corresponding warning interval based on driving scenario matching; Based on the direction and magnitude of the vehicle's longitudinal speed, the warning interval of the vehicle speed is determined, and warning information of corresponding levels is generated according to the warning interval; different warning intervals correspond to different levels of warning information, and the higher the level of warning information, the more dangerous the vehicle driving is.

6. The vehicle motion state detection method based on millimeter wave radar according to claim 5, characterized in that: The warning interval is determined by the following method: Obtain accident information and the driving scenario in which each accident occurred, and summarize accident information for the same driving scenario; According to the vehicle speed information in the accident information, a corresponding warning interval is divided for each driving scenario.

7. The vehicle motion state detection method based on millimeter wave radar according to claim 1, characterized in that: The discrete wavelet transform adopts Dobesi wavelet to perform discrete wavelet transform. Correspondingly, the calculation formula for discrete wavelet transform adopting Dobesi wavelet is: ； ； in, Indicates the low-frequency part; Represents the high frequency part, represents the first coefficient, which is the coefficient corresponding to the scaling function; represents the second coefficient, which is the coefficient corresponding to the Dobesi wavelet function; Indicates the input signal.

8. The vehicle motion state detection method based on millimeter wave radar according to claim 7, characterized in that: The step of extracting a coordinate transformation signal from the echo signal according to the return time node, performing discrete wavelet transform on the coordinate transformation signal, extracting high-frequency components and low-frequency components from the vehicle coordinate parameters, and determining a coordinate transformation speed in a coordinate direction according to the high-frequency components and the low-frequency components specifically includes: Extracting a coordinate transformation signal from the echo signal according to a return time node; Using the coordinate transformation signal and Dobesi wavelet to perform discrete wavelet transformation, extracting high-frequency components and low-frequency components in the vehicle coordinate parameters; Accumulating the high-frequency component and the low-frequency component to obtain a high-frequency total component and a low-frequency total component respectively; The coordinate transformation acceleration is obtained by dividing the total high-frequency component by the total low-frequency component, and the coordinate transformation speed is obtained according to the coordinate transformation speed.

9. A vehicle motion state detection device based on millimeter wave radar, characterized in that: The device comprises: A parameter acquisition module is used to acquire the echo signal returned by the vehicle-mounted millimeter-wave radar; the echo signal includes vehicle speed parameters, vehicle coordinate parameters, radar cross-sectional area parameters, and the return time nodes of the vehicle speed parameters, vehicle coordinate parameters, and radar cross-sectional area parameters; The target filtering module is used to filter the echo signal according to the vehicle speed parameter, the vehicle coordinate parameter and the radar reflection cross-sectional area parameter, filter out the abnormal values in the echo signal, and obtain the filtered echo signal; a wavelet transform module for extracting a coordinate transformation signal from the echo signal according to a return time node, performing a discrete wavelet transform using the coordinate transformation signal, extracting high-frequency and low-frequency components from the vehicle coordinate parameters, and determining a coordinate transformation speed in a coordinate direction based on the high-frequency and low-frequency components; the coordinate transformation signal comprises the vehicle coordinate parameters of consecutive adjacent frames, the vehicle coordinate parameters of the first frame in the coordinate transformation signal differ from the vehicle coordinate parameters of the last frame by a first preset number of frames, and the vehicle coordinate parameters of the last frame in the coordinate transformation signal are the vehicle coordinate parameters of a frame in the echo signal that is closest to the current time node, and the value of the first preset number of frames is determined based on a driving scenario of the vehicle; The state detection module is used to determine the correction parameters according to the vehicle's driving scenario, use the coordinate transformation speed and the correction parameters to correct the vehicle speed parameters, obtain the current actual speed of the vehicle at the current time node, and determine the vehicle's motion state based on the current actual speed.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the vehicle motion state detection method based on millimeter wave radar as described in any one of claims 1 to 8 are implemented.

Images