Self-vehicle speed estimation method and device, electronic device and storage medium

Abstract

The invention relates to a self-vehicle speed estimation method and device, an electronic device and a storage medium, and the method comprises the steps: obtaining trace point data captured by a radar sensor, the trace point data comprising the radial speed and azimuth angle of each target trace point; determining a static target trace point from the target trace points based on the azimuth angle of each target trace point; based on the radial speed and the azimuth angle of the stationary target plot, calculating a target ground speed corresponding to each stationary target plot; dividing a speed interval for the target ground speed based on a preset step length; based on the trace point distribution feature information of the static target trace point in each target speed interval, determining a target effective interval; and calculating the average value of the radial speed of each static target trace point in the target effective interval to obtain a self-vehicle speed estimation value. According to the invention, the problem that the precision of the vehicle speed calculated through the millimeter wave radar is insufficient when the vehicle runs at a low speed is solved, and the low-speed vehicle high-precision estimation independent of external vehicle speed input is realized.

Description

Technical Field

[0001] The present application relates to the field of radar signal processing, and in particular to a vehicle speed estimation method, device, electronic device and storage medium. Background Art

[0002] With the rapid development of autonomous and assisted driving technologies, vehicles are increasingly demanding the real-time and precision of their perception systems. Millimeter-wave radar, due to its 24 / 7 operation, is widely used in areas such as autonomous driving and collision warning. In practical applications, accurate acquisition of the vehicle's speed is crucial for calculating the target's ground speed.

[0003] Traditionally, ego vehicle speed is typically determined via the CAN bus of the vehicle's domain control system (DCS). This not only increases system complexity but also places high demands on real-time communication and incurs high costs when adapting to different vehicle models and radar systems. Especially in low-speed scenarios, when demonstrating or validating radar, the accuracy of vehicle speed is less critical; demonstrating radar imaging performance, such as point cloud density and ranging and angle measurement capabilities, is more important. Therefore, if the millimeter-wave radar could accurately calculate ego vehicle speed by processing and analyzing the detected point cloud data itself in low-speed scenarios, the communication module with the vehicle's DCS could be eliminated, significantly reducing the cost of millimeter-wave radar demonstration and validation.

[0004] However, existing methods for millimeter-wave radar target detection and ego-vehicle speed calculation have several challenges. Some require a vehicle speed database and rely on information about the vehicle's surroundings and real-time speed data. Others, applied to systems for identifying abnormal vehicle behavior, also require real-time vehicle information acquisition and storage. These methods are not only complex but also require even higher speed accuracy at low speeds, making it difficult to guarantee the accuracy and reliability of the calculated results.

[0005] Currently, no effective solution has been proposed to the problem in related technologies of insufficient accuracy in calculating the vehicle's speed using millimeter-wave radar when the vehicle is traveling at low speeds. Summary of the Invention

[0006] Embodiments of the present application provide a method, apparatus, system, electronic device, and storage medium for estimating the speed of a vehicle to at least address the problem in the related art of insufficient accuracy in calculating the speed of a vehicle using millimeter-wave radar when the vehicle is traveling at a low speed.

[0007] In a first aspect, an embodiment of the present application provides a method for estimating a vehicle speed, comprising:

[0008] Acquiring point track data captured by a radar sensor, wherein the point track data includes a radial velocity and an azimuth angle of each target point track;

[0009] determining a stationary target point track from the target point tracks based on the azimuth of each of the target point tracks;

[0010] Calculating the target ground velocity corresponding to each of the stationary target points based on the radial velocity and azimuth of the stationary target point tracks;

[0011] Based on a preset step size, the target ground speed is divided into speed intervals to obtain a plurality of target speed intervals; based on the point trace distribution characteristic information of the detected stationary target point trace in each of the target speed intervals, a target effective interval is determined from the plurality of target speed intervals;

[0012] The average value of the radial velocities of the stationary target points within the target effective interval is calculated to obtain an estimated value of the vehicle speed.

[0013] In some embodiments, determining a stationary target point track from the target point tracks based on the azimuth of each target point track includes:

[0014] If the azimuth angle of the target point track is within a preset angle range, a stationary target point track is determined from the target point track.

[0015] In some embodiments, obtaining the point trace data captured by the radar sensor includes:

[0016] Acquire multiple frames of scene images captured by the radar sensor;

[0017] Based on the multiple frames of scene images, obtaining point trace data in the multiple frames of scene images;

[0018] When the number of the target points is greater than the preset number of points or the number of frames traversing the scene image exceeds the preset upper limit of the number of frames, the acquisition of the point data in the plurality of scene images is stopped.

[0019] In some embodiments, determining a target valid interval from the plurality of target speed intervals based on the detected point trace distribution characteristic information of the stationary target point trace in each target speed interval includes:

[0020] Calculating the number of the stationary target traces within each of the target speed intervals; and determining a target effective interval from the plurality of target speed intervals based on the number of traces.

[0021] In some embodiments, the method further comprises:

[0022] After obtaining the estimated value of the vehicle speed, reacquiring updated point track data captured by the radar sensor, the updated point track data including the radial velocity and azimuth of each updated target point track;

[0023] determining an updated stationary target point trace from the updated target point traces based on the azimuth of each of the updated target point traces;

[0024] Calculating an updated target ground velocity corresponding to each of the updated stationary target points based on the radial velocity and azimuth of the updated stationary target point tracks;

[0025] Based on a preset step size, the updated target ground speed is divided into speed intervals to obtain a plurality of updated target speed intervals; based on the detected point trace distribution feature information of the updated stationary target point trace within each of the updated target speed intervals, an updated target valid interval is determined from the plurality of updated target speed intervals;

[0026] Calculating the average radial velocity of each of the updated stationary target points within the updated target valid interval to obtain a current vehicle speed estimate;

[0027] determining a vehicle speed fluctuation range based on a preset vehicle speed fluctuation value and the estimated vehicle speed;

[0028] The estimated vehicle speed is updated based on the current estimated vehicle speed and the vehicle speed fluctuation range.

[0029] In some embodiments, updating the estimated vehicle speed based on the current estimated vehicle speed and the vehicle speed fluctuation range includes:

[0030] If the current vehicle speed estimate is within the vehicle speed fluctuation range, updating the vehicle speed estimate according to the current vehicle speed estimate;

[0031] If the current vehicle speed estimate exceeds the vehicle speed fluctuation range, updating the own vehicle speed estimate is stopped.

[0032] In some embodiments, the method further comprises:

[0033] If the number of times the update of the estimated vehicle speed is stopped exceeds a preset number of update stoppage times, a new estimated vehicle speed is recalculated.

[0034] In a second aspect, an embodiment of the present application provides a vehicle speed estimation device, comprising:

[0035] A point track acquisition module is used to acquire point track data captured by the radar sensor, wherein the point track data includes the radial velocity and azimuth angle of each target point track;

[0036] The point track acquisition module is further configured to determine a stationary target point track from the target point tracks based on the azimuth of each target point track;

[0037] a velocity calculation module, configured to calculate the target ground velocity corresponding to each of the stationary target tracks based on the radial velocity and azimuth of the stationary target tracks;

[0038] The speed calculation module is further configured to divide the target ground speed into speed intervals based on a preset step size to obtain a plurality of target speed intervals; and determine a target effective interval from the plurality of target speed intervals based on point trace distribution feature information of the detected stationary target point trace within each of the target speed intervals;

[0039] The result generating module is used to calculate the average radial velocity of each of the stationary target points within the target effective interval to obtain an estimated value of the vehicle speed.

[0040] In a third aspect, an embodiment of the present application provides an electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the vehicle speed estimation method as described in the first aspect above is implemented.

[0041] In a fourth aspect, an embodiment of the present application provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the vehicle speed estimation method as described in the first aspect above.

[0042] Compared with related technologies, the embodiments of the present application provide a method, device, system, electronic device and storage medium for estimating the speed of a self-vehicle. By screening multi-frame stationary target points and performing speed interval statistics, they solve the problem of insufficient accuracy in calculating the speed of the self-vehicle using millimeter-wave radar when the vehicle is traveling at low speeds, thereby achieving high-precision estimation of the low-speed self-vehicle without relying on external vehicle speed input.

[0043] The details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0045] Figure 1 1 is a hardware structure block diagram of a terminal of a vehicle speed estimation method according to an embodiment of the present invention;

[0046] Figure 2 is a flow chart of a method for estimating vehicle speed according to an embodiment of the present application;

[0047] Figure 3 is a flow chart of a method for estimating vehicle speed according to a preferred embodiment of the present application;

[0048] Figure 4 2 is a speed decomposition principle diagram of a vehicle speed estimation method according to a preferred embodiment of the present application;

[0049] Figure 5 2 is a schematic diagram of an outdoor parking scene according to a method for estimating the vehicle speed of a vehicle according to a preferred embodiment of the present application;

[0050] Figure 6 This is a comparison chart of the actual vehicle speed obtained by the CAN line in the prior art and the vehicle speed estimation result of the preferred embodiment of the present application;

[0051] Figure 7 A speed difference diagram comparing the actual vehicle speed obtained by the CAN line in the prior art and the vehicle speed estimation result of the preferred embodiment of the present application;

[0052] Figure 8 is a radar 4D imaging scene diagram of the vehicle speed estimation method according to the preferred embodiment of the present application;

[0053] Figure 9 4 is a structural block diagram of a vehicle speed estimation device according to an embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for ordinary technicians in the field related to the contents disclosed in the present application, some changes such as design, manufacturing or production based on the technical contents disclosed in the present application are only conventional technical means and should not be understood as the contents disclosed in the present application being insufficient.

[0055] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments unless there is a conflict.

[0056] Unless otherwise defined, the technical or scientific terms used in this application should have the ordinary meaning understood by a person of ordinary skill in the technical field to which this application belongs. The words "one", "a", "the" and the like used in this application do not indicate a limit on quantity and may indicate the singular or plural. The terms "include", "comprise", "have" and any variations thereof used in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units that are inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The word "multiple" used in this application means greater than or equal to two. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. The terms "first", "second", "third" and the like involved in this application are merely used to distinguish similar objects and do not represent a specific ordering of the objects.

[0057] The method embodiment provided in this embodiment can be executed in a terminal, a computer or a similar computing device. Taking running on a terminal as an example, Figure 1 FIG. 1 is a block diagram of the hardware structure of a terminal of the vehicle speed estimation method according to an embodiment of the present invention. Figure 1 As shown, the terminal may include one or more ( Figure 1 Only one is shown) a processor 102 (the processor 102 may include but is not limited to a microprocessor MCU or a programmable logic device FPGA and other processing devices) and a memory 104 for storing data. Optionally, the terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above terminal. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0058] Memory 104 can be used to store computer programs, such as software programs and modules of application software, such as the computer program corresponding to the vehicle speed estimation method in the embodiment of the present invention. Processor 102 executes the computer program stored in memory 104 to perform various functional applications and data processing, thereby implementing the aforementioned method. Memory 104 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, memory 104 may further include memory remotely located relative to processor 102, and such remote memory may be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0059] The transmission device 106 is used to receive or send data via a network. A specific example of the aforementioned network may include a wireless network provided by the terminal's communications provider. In one embodiment, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0060] This embodiment provides a method for estimating the vehicle speed. Figure 2 is a flow chart of a method for estimating the vehicle speed according to an embodiment of the present application. Figure 2 As shown, the process includes the following steps:

[0061] Step S201, acquiring point track data captured by a radar sensor, where the point track data includes radial velocity and azimuth of each target point track;

[0062] Among them, the radar sensor (such as millimeter wave radar) continuously scans the surrounding environment, transmits signals and receives reflected signals from the target, and performs signal processing on the received reflected signals, including mixing, filtering, amplification and other steps to extract useful target information. Through the digital signal processing algorithm, the radial velocity and azimuth of each target point are extracted from the processed signal. The radial velocity is calculated by measuring the Doppler frequency shift of the reflected signal. The radial velocity represents the component of the relative velocity between the target and the radar in the direction of the radar line of sight. The azimuth is determined by measuring the angle of arrival of the reflected signal. The azimuth represents the direction of the target relative to the radar. The extracted point data (including radial velocity and azimuth) can be stored in the radar's data processing unit for subsequent processing and analysis. Through precise radar scanning and signal processing algorithms, this embodiment can obtain accurate target point track data, including radial velocity and azimuth, and can simultaneously capture point track data of multiple targets, including vehicles, pedestrians, obstacles, etc., providing a reliable data basis for subsequent vehicle speed calculation and ensuring the accuracy of the calculation results; by capturing and processing target point track data in real time through radar, the vehicle speed calculation can respond to changes in the vehicle's motion state in a timely manner.

[0063] Step S202, determining a stationary target point track from the target point tracks based on the azimuth angles of the target point tracks;

[0064] In this embodiment, the azimuth angles of all target points can be traversed first, and the points whose azimuth angles meet the preset conditions can be selected as candidate stationary target points. The preset condition can be that the azimuth angle θ falls within the preset angle range [θ0, θ1]. For the selected candidate stationary target points, their radial velocity or other motion characteristics (such as the speed change of multiple consecutive frames) are further analyzed. If the radial velocity of the candidate point is close to zero and the speed does not change much in the consecutive frames, it is determined to be a stationary target point. In this embodiment, the radial velocity of the candidate point is close to zero and the speed does not change much in the consecutive frames according to the radial velocity or other motion characteristics (such as the speed change of multiple consecutive frames) as the candidate stationary target point; then, the points whose azimuth angle θ falls within the preset angle range [θ0, θ1] are selected from the candidate stationary target points, and these points are determined as stationary target points. In addition, the embodiment of this step can also determine the stationary target points from the target points at the same time according to the two conditions that the radial velocity is within the threshold range and the azimuth angle is within the preset angle range. This step reduces the interference of dynamic targets (such as moving vehicles and pedestrians) on the vehicle speed calculation by screening stationary target points. In a complex and changing traffic environment, the number and speed of dynamic targets vary greatly. By screening stationary target points, the system can better adapt to such environmental changes, improving the robustness and stability of the system. The radial velocity changes of stationary target points mainly reflect the motion state of the vehicle, so the vehicle speed calculated based on the stationary target points is more accurate.

[0065] Step S203, calculating the target ground velocity corresponding to each stationary target track based on the radial velocity and azimuth of the stationary target track;

[0066] Among them, the radial velocity of each point is extracted from the determined stationary target point (denoted as V r ) and azimuth (denoted as θ). Using the kinematic relationship between the radar installation position and the vehicle, the radial velocity and azimuth of the stationary target point track are converted into the target's velocity in the ground coordinate system (i.e., the target ground velocity, denoted as V k ), can be achieved through V k =V r / cosθ (assuming the vehicle is traveling in a straight line and has zero yaw rate). This step converts the radial velocity and azimuth of the stationary target point trace into the target ground velocity. This more accurately reflects the target's actual motion state in the ground coordinate system, providing a more reliable data foundation for subsequent ego vehicle speed estimation and improving the accuracy of speed estimation. The target ground velocity calculation more accurately reflects the target's actual motion state in different driving scenarios and road conditions (such as curves and slopes), enhancing the system's adaptability to different driving scenarios and road conditions, and improving its reliability and stability.

[0067] Step S204: dividing the target ground speed into speed intervals based on a preset step size to obtain a plurality of target speed intervals; and determining a target effective interval from the plurality of target speed intervals based on the point trace distribution characteristic information of the detected stationary target point traces within each target speed interval;

[0068] Among them, a preset step size (such as 0.25m / s) is set to divide the range of the target ground speed. According to the preset step size, the target ground speed is divided into multiple continuous speed intervals. For example, if the preset step size is 0.25m / s, the speed interval can be [0, 0.25), [0.25, 0.5), [0.5, 0.75), etc. Traverse all stationary target traces and calculate the target ground speed corresponding to each trace. The trace distribution feature information can be the number of stationary target traces in each speed interval, or the convergence and dispersion characteristics of the stationary target traces in each speed interval; according to the trace distribution feature information, determine the target effective interval from multiple target speed intervals. This step divides the speed range and counts the distribution characteristics of the point traces to find the value that best matches the vehicle's speed from among many speed values, effectively reducing the impact of abnormal speed values on the speed estimate and improving the accuracy of the speed estimate. In complex and changing traffic environments, the target's motion state may be affected by various factors (such as wind speed and road bumps). By determining the target's effective range, the system can better adapt to such environmental changes, improving the system's robustness and stability. In addition, there is no need to perform complex mathematical operations or model fitting on each stationary target point trace. Only simple statistics and screening based on the speed range are required, which simplifies the data processing process, reduces computational complexity, and improves the system's real-time performance and response speed.

[0069] Step S205 : Calculate the average radial velocity of each stationary target point within the target effective interval to obtain the estimated value of the vehicle speed.

[0070] The radial velocities of all stationary target traces within the previously determined target validity interval are extracted. These summed values are then divided by the number of traces to obtain the average radial velocity, which serves as the vehicle's speed estimate. By calculating the average radial velocity of the stationary target traces within the target validity interval, this method effectively reduces the impact of abnormal speed values on vehicle speed estimation. This method, based on the statistical results of a large number of stationary target traces, does not rely on a single target trace or complex mathematical models. This allows the system to more stably output accurate vehicle speed estimates in complex and changing traffic environments, enhancing system reliability.

[0071] Through the above steps, the present application directly captures the point data containing the radial velocity and azimuth of each target point from the radar sensor, directly utilizing the high-precision detection capability of the millimeter-wave radar without the need for additional equipment or complex data preprocessing; then, based on the azimuth of each target point, the stationary target point is screened out, and by assuming that the number of stationary targets in the scene is greater than that of moving targets, and selecting points with angles near the normal as data input, the interference of dynamic targets on the vehicle speed calculation is effectively reduced, and the accuracy of subsequent vehicle speed calculation is significantly improved; then, using the radial velocity and azimuth of the stationary target point, the target ground speed corresponding to each stationary target point is calculated through coordinate transformation, and the radial velocity detected by the radar is accurately deduced through the mathematical model. The speed is converted into a more practical target ground speed, providing a more reliable data basis for vehicle speed calculation; the target ground speed is divided based on a preset step size to obtain multiple speed intervals, and the target effective interval is determined by statistically analyzing the point distribution characteristics of the stationary target points in each speed interval. By using the big data statistical method, the influence of abnormal speed values on the vehicle speed calculation is effectively removed, further improving the accuracy of vehicle speed estimation; finally, the average value of the radial velocity of each stationary target point in the target effective interval is calculated to obtain the estimated value of the vehicle speed. By calculating the average value, the vehicle speed estimation result is further smoothed, reducing the influence of single measurement errors on the final result, thereby obtaining a more accurate and stable estimated value of the vehicle speed. Compared with related technologies, this application effectively solves the problem of insufficient accuracy of the vehicle speed calculated by millimeter-wave radar when the vehicle is traveling at low speeds. There is no need to rely on additional vehicle speed sensors or complex mathematical models. High-precision vehicle speed calculation can be achieved only through the point data captured by the radar sensor.

[0072] In some embodiments, determining a stationary target track from the target tracks based on the azimuth of each target track includes:

[0073] The radar detects a series of target traces, each containing information such as its azimuth angle and radial velocity. Based on the radar's installation location, detection scenario, and algorithm requirements, a suitable preset angle range [θ0, θ1] is set. This range is typically chosen near the radar normal, as stationary targets (such as roadside parking and buildings) are more likely to appear at these angles. From all detected target traces, traces with azimuth angles θ falling within the preset angle range [θ0, θ1] are screened. These traces are considered possible stationary target traces. For these screened traces, their radial velocity is further analyzed. In low-speed vehicle scenarios, the radial velocity of stationary targets should be close to zero (a small velocity threshold can be set to account for measurement errors and minor vehicle vibrations). Traces with radial velocities within the threshold range are identified as stationary target traces. By screening traces with azimuth angles within the preset angle range and further determining whether they are stationary targets, this embodiment effectively reduces the impact of abnormal traces (such as dynamic targets or error traces) on vehicle speed calculation. Since the radial velocity of a stationary target is mainly determined by the vehicle speed, using these traces to estimate the vehicle speed can improve the accuracy. It does not rely on the speed information provided by an external speed sensor (such as the CAN bus), but instead uses the data detected by the radar itself to estimate the vehicle speed.

[0074] In some embodiments, obtaining point trace data captured by a radar sensor includes:

[0075] Acquire multiple frames of scene images captured by a radar sensor;

[0076] Based on the multiple frames of scene images, obtaining point trace data in the multiple frames of scene images;

[0077] When the number of target traces is greater than the preset number of traces or the number of frames traversing the scene image exceeds the preset upper limit of the frame number, the acquisition of trace data in the multiple scene images is stopped.

[0078] Among them, the radar sensor continuously captures multiple frames of scene images of the vehicle's surrounding environment. These images contain information such as the position and speed of the target object. From each frame of the scene image, the radar signal processing technology is used to extract the target point track data. Each point track data usually contains information such as the target's azimuth, pitch angle, radial speed, etc. Two stopping conditions are set: one is that the number of target track points is greater than the preset number of track points (N det); second, the number of frames of the traversed scene image exceeds the preset frame limit (the default is 6 frames, but it can be adjusted according to the actual scene). In the process of acquiring point trace data, the number of target point traces in the current frame is counted in real time. If the number of target point traces reaches or exceeds the preset number of point traces, or the number of traversed frames reaches the preset frame limit, the acquisition of new scene images and point trace data is stopped. By setting the stop acquisition condition, this embodiment can effectively reduce unnecessary calculations while ensuring the accuracy of the algorithm. When enough point trace data is obtained or a sufficient number of frames are traversed, the data acquisition process is stopped in time, thereby improving the overall efficiency of the algorithm. In actual applications, the data processing capacity and storage resources of the radar sensor are limited. By reasonably setting the stop acquisition condition, the waste of precious computing resources and storage space can be avoided, so that the system can run more efficiently. In addition, different traffic scenarios and driving speeds have different requirements for radar data. By flexibly adjusting parameters such as the preset number of point traces and the frame limit, the algorithm can be better adapted to the needs of different scenarios, thereby improving the applicability and flexibility of the algorithm.

[0079] In some embodiments, determining a target valid interval from a plurality of target speed intervals based on point distribution characteristic information of detected stationary target points within each target speed interval includes:

[0080] The number of stationary target traces within each target speed interval is calculated; and based on the number of traces, a target valid interval is determined from the multiple target speed intervals.

[0081] Among them, according to the radial velocity and azimuth of the target point track, the velocity V of each point track in the ground coordinate system is calculated k (Through the formula V k =V r / cosθ is obtained, where V r is the radial velocity, θ is the azimuth angle); V kDivide into multiple target speed intervals (for example, divided at intervals of 0.25m / s); count the number of stationary target traces in each target speed interval. Among all target speed intervals, find the interval with the largest number of traces, that is, the target effective interval, which is the interval that best represents the current speed of the vehicle. This embodiment can more accurately estimate the current speed of the vehicle by counting the number of stationary target traces in each target speed interval and determining the interval with the largest number of traces as the target effective interval, because the number distribution of stationary target traces is closely related to the speed of the vehicle, and the interval with the largest number of traces often corresponds to the most likely speed of the vehicle. By counting the number distribution of multiple traces to determine the vehicle speed, it does not rely on the precise measurement value of a single trace, so that the algorithm has stronger robustness to the measurement errors and outliers of a single trace, and only focuses on the interval with the largest number of traces (that is, the target effective interval), which can reduce the amount of data for subsequent processing, thereby improving the overall efficiency of the algorithm.

[0082] In some embodiments, the method further comprises:

[0083] After obtaining the estimated value of the vehicle speed, reacquire the updated point track data captured by the radar sensor, the updated point track data including the radial speed and azimuth of each updated target point track;

[0084] determining an updated stationary target point trace from the updated target point traces based on the azimuths of the updated target point traces;

[0085] Calculating the updated target ground velocity corresponding to each updated stationary target point track based on the radial velocity and azimuth of the updated stationary target point track;

[0086] Based on a preset step size, the updated target ground speed is divided into speed intervals to obtain a plurality of updated target speed intervals; based on the point trace distribution characteristic information of the detected updated stationary target point trace in each updated target speed interval, an updated target valid interval is determined from the plurality of updated target speed intervals;

[0087] Calculate the average radial velocity of each updated stationary target point within the updated target valid interval to obtain the current vehicle speed estimate;

[0088] Determining a vehicle speed fluctuation range based on a preset vehicle speed fluctuation value and an estimated vehicle speed;

[0089] Update the estimated vehicle speed based on the current estimated vehicle speed and the speed fluctuation range.

[0090] After obtaining the initial or last vehicle speed estimate, the radar sensor continues to capture new scene images and extract update point data from them. The update point data includes the radial velocity and azimuth of each updated target point. Based on the azimuth of each updated target point, the same method as the initial screening is used (i.e., determining whether the azimuth is within the preset angle range) to determine the updated stationary target point from the updated target point. For each updated stationary target point, the formula V is used. k =V r / cosθ(where V k is the speed in the ground coordinate system, that is, the updated target ground speed; V r is the radial velocity of the updated target point trace, θ is the azimuth angle), and the corresponding updated target ground speed is calculated. Based on the preset step size (such as 0.25m / s), the updated target ground speed is divided to obtain multiple updated target speed intervals. According to the point distribution characteristic information (such as the number of points) of the detected updated stationary target point traces in each updated target speed interval, the updated target effective interval is determined from the multiple updated target speed intervals, which can be determined by finding the interval with the largest number of points. The average value of the radial velocity of each updated stationary target point trace in the updated target effective interval is calculated as the current vehicle speed estimate. Based on the preset vehicle speed fluctuation value (such as ±0.5m / s), the vehicle speed fluctuation interval is determined. The vehicle speed estimate is updated according to the relationship between the current vehicle speed estimate and the vehicle speed fluctuation interval. This embodiment can track the changes in the vehicle's driving state in real time by continuously updating the point data and recalculating the vehicle speed estimate, thereby improving the real-time and accuracy of the vehicle speed estimation; estimating the vehicle speed by multi-frame accumulation and multiple iterations can effectively reduce the impact of a single measurement error on the result and enhance the robustness of the algorithm.

[0091] In some embodiments, updating the vehicle speed estimate based on the current vehicle speed estimate and the vehicle speed fluctuation range includes:

[0092] If the current estimated speed is within the speed fluctuation range, the estimated speed of the vehicle is updated based on the current estimated speed.

[0093] If the current estimated vehicle speed exceeds the speed fluctuation range, the update of the estimated vehicle speed will be stopped.

[0094] In the updated target effective interval, the average radial velocity of each updated stationary target point is calculated as the current vehicle speed estimate (denoted as V c The speed fluctuation range is based on the estimated value of the vehicle speed (denoted as V o ) is the center, and the fluctuation range is determined by the preset vehicle speed fluctuation value. For example, the vehicle speed fluctuation range is [V o -0.5, V o +0.5]. The current vehicle speed estimate Vc Compared with the speed fluctuation range, if the current speed estimate V c If the vehicle speed is within the speed fluctuation range, the current estimated vehicle speed is considered acceptable (the difference between the two real-time calculated vehicle speeds is not large). c Update the estimated speed of the vehicle (V o =V c If the current estimated vehicle speed V c If the speed fluctuation range is exceeded, it is considered that the current speed estimate may have a large error or abnormality, so the update of the vehicle speed estimate is stopped and the previous speed estimate is continued to be used (i.e., V o This embodiment sets the speed fluctuation range and verifies the current speed estimate, which can avoid large fluctuations in the speed estimate caused by single measurement errors or abnormal values, thereby improving the stability of the speed estimate. c When it is within the normal fluctuation range, the value is directly used to update the estimated speed of the vehicle, which reduces unnecessary calculation and update processes and improves the efficiency of the algorithm.

[0095] In some embodiments, the method further comprises:

[0096] If the number of times the update of the estimated vehicle speed is suspended exceeds a preset update suspension number, a new estimated vehicle speed is recalculated.

[0097] Among them, during the operation of the algorithm, the number of consecutive stop updates of the vehicle speed estimate is recorded. When the current vehicle speed estimate exceeds the speed fluctuation range, the vehicle speed estimate is stopped and updated, and the number of stop updates is increased by 1. A preset update stop number is set (this value can be adjusted according to the actual application scenario and algorithm performance requirements). After each stop update, it is determined whether the number of stop updates reaches or exceeds the preset value. If the number of stop updates exceeds the preset value, it is considered that the current vehicle speed estimate may have a large deviation or the algorithm state is abnormal. At this time, the complete process of vehicle speed estimation needs to be re-executed, including re-acquiring multiple frames of scene images, extracting point data, filtering stationary target points, calculating the number of points in the target speed range, determining the target effective range, calculating the current vehicle speed estimate, and other steps to obtain a new vehicle speed estimate. This embodiment sets a preset number of update pauses and recalculates the estimated vehicle speed when this number is exceeded. This effectively avoids problems with excessive deviations in the estimated vehicle speed due to abnormal algorithm states or prolonged pauses in updating, thereby enhancing the reliability of the algorithm. When the algorithm pauses in updating the estimated vehicle speed for a long time, it may mean that the current estimated vehicle speed has deviated significantly from the actual value. Therefore, by recalculating the estimated vehicle speed, the deviation can be corrected in a timely manner, thereby improving the accuracy of the estimated vehicle speed.

[0098] The embodiments of the present application are described and illustrated below through preferred embodiments.

[0099] Figure 3 is a flow chart of a method for estimating the vehicle speed according to a preferred embodiment of the present application. Figure 3 As shown, this preferred embodiment mainly consists of two parts: initial vehicle speed estimation and driving speed calculation.

[0100] The initial vehicle speed estimation has the following steps:

[0101] Step S1: Count the relative radial velocity V of all radar point targets in the current scene within n frames r and azimuth angle θ.

[0102] Step S2: Count the number of point targets within the set angle [θ0, θ1] in n frames until the number of point traces is greater than N det Or n reaches the set frame limit. The set frame limit is a variable value, the default is 6 frames.

[0103] Step S3: Figure 4 : is a speed decomposition principle diagram of the vehicle speed estimation method according to the preferred embodiment of the present application, such as Figure 4 As shown, the velocity of the point target in the ground coordinate system within n frames can be obtained by the following formula: V k =V r / cosθ.

[0104] Step S4: From V k Starting from the minimum speed, divide V by 0.25m / s k , and find V k The largest number of intervals S k Statistics S k V corresponding to all elements in r and included in the set S r .

[0105] Step S5: Calculate S r The arithmetic mean of the elements in V o , as the vehicle speed of the frame. Repeat steps 1 to 3 until the output V o Stable within 0.5m / s. Output stable V o As the initial speed of the vehicle, enter the driving speed calculation.

[0106] The vehicle speed estimation has the following steps:

[0107] Step S6: Count the relative radial velocity V of all radar point targets in the current scene within n frames r and azimuth angle θ.

[0108] Step S7: Count the number of point targets within the set angle [θ0, θ1] in n frames until the number of point traces is greater than N det Or n reaches the upper limit of the set frame number. The upper limit of the frame number is a variable value, the default is 6 frames. Calculate the velocity V of all point targets in the ground coordinate system k =V r / cosθ.

[0109] Step S8: From V k Starting from the minimum speed, divide V by 0.25m / s k , and find V k The largest number of intervals S k .

[0110] Step S9: Count S k V corresponding to all elements in r , according to the vehicle speed V in the previous frame o Filter out those that meet V o -0.5 <V r <V o +0.5 set Sr. V o -0.5 and V o +0.5 is a general setting value that can be modified according to the actual scenario. The upper limit of the interval affects the estimated acceleration speed of the car, while the lower limit of the interval affects the estimated deceleration speed of the car.

[0111] Step S10: Determine S r Is it an empty set?

[0112] If S r If the set is empty, it means that the vehicle speed has changed significantly compared with the previous n frames, so V is not performed. o Change, continue to use the previous frame to calculate the vehicle speed, V r =V o .

[0113] If S r is not an empty set, then let V r For S r Arithmetic mean. The initial speed is set to: V o =V r .

[0114] Step S11: Output the current frame speed V o According to V o Recalculate the ground speed of all point targets. Update V o , enter the speed calculation of the next n frames and repeat steps 5 to 10. o If multiple frames are not updated, the initial vehicle speed estimation process is restarted.

[0115] Steps S1 to S5 have a high degree of overlap with steps S6 to S10, and these modules can be reused to reduce algorithm complexity and code storage memory.

[0116] Taking the corner radar as an example, the millimeter wave radar is installed on the side and rear of the car. First, verify that this embodiment Figure 5 Performance in an outdoor parking scenario is shown.

[0117] Figure 6 This is a comparison chart of the actual vehicle speed obtained by the prior art CAN line and the vehicle speed estimation result of the preferred embodiment of the present application, as shown in FIG. Figure 6 As shown, the vehicle speed output by the CAN line (blue line) and the calculated vehicle speed (red line) are compared. Figure 7 This is a speed difference diagram comparing the actual vehicle speed obtained by the prior art CAN line and the vehicle speed estimation result of the preferred embodiment of the present application. The calculated difference is as follows: Figure 7 As shown in the figure, the maximum calculated vehicle speed difference is within ±0.5m / s, and the average vehicle speed difference per frame is 0.082m / s. This error is in the vehicle-borne millimeter-wave 4D radar imaging.

[0118] According to the vehicle speed estimated in the preferred embodiment of the present application, the vehicle speed is plotted as follows: Figure 8 The radar 4D imaging scene diagram is shown. It can be seen that the vehicle speed estimation using the preferred embodiment of this application can relatively accurately create a radar point cloud map and identify empty parking spaces and roadside parking. It has been verified that this application can accurately calculate the own-vehicle speed when the own-vehicle speed is low by screening small azimuth angle points and accumulating multiple frames. The calculated own-vehicle speed can replace the CAN line output speed, supporting 4D millimeter-wave radar imaging demonstrations.

[0119] This embodiment also provides a vehicle speed estimation device for implementing the above-described embodiments and preferred embodiments. Details already described will not be repeated. As used below, terms such as "module," "unit," and "subunit" may refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0120] Figure 9 is a structural block diagram of a vehicle speed estimation device according to an embodiment of the present application, such as Figure 9 As shown, the device includes:

[0121] The point track acquisition module 10 is used to acquire the point track data captured by the radar sensor, the point track data including the radial velocity and azimuth of each target point track;

[0122] The point track acquisition module 10 is further used to determine a stationary target point track from the target point tracks based on the azimuth of each target point track;

[0123] The velocity calculation module 20 is used to calculate the target ground velocity corresponding to each stationary target track based on the radial velocity and azimuth of the stationary target track;

[0124] The speed calculation module 20 is further configured to divide the target ground speed into speed intervals based on a preset step size to obtain a plurality of target speed intervals; and determine a target effective interval from the plurality of target speed intervals based on the point trace distribution characteristic information of the detected stationary target point traces within each target speed interval;

[0125] The result generation module 30 is used to calculate the average radial velocity of each stationary target point within the target effective interval to obtain the estimated value of the vehicle speed.

[0126] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0127] This embodiment further provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0128] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0129] Optionally, in this embodiment, the processor may be configured to execute the following steps through a computer program:

[0130] S101, acquiring point track data captured by a radar sensor, where the point track data includes radial velocity and azimuth of each target point track;

[0131] S102, determining a stationary target point track from the target point tracks based on the azimuth angles of the target point tracks;

[0132] S103, calculating the target ground velocity corresponding to each stationary target point track based on the radial velocity and azimuth angle of the stationary target point track;

[0133] S104: Dividing the target ground speed into speed intervals based on a preset step size to obtain a plurality of target speed intervals; and determining a target effective interval from the plurality of target speed intervals based on point trace distribution characteristic information of the detected stationary target point traces within each target speed interval.

[0134] S105 , calculating the average radial velocity of each stationary target point within the target effective interval to obtain an estimated value of the vehicle speed.

[0135] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be repeated here.

[0136] In addition, in conjunction with the vehicle speed estimation method in the above embodiments, the present application can provide a storage medium for implementation. The storage medium stores a computer program; when the computer program is executed by a processor, any of the vehicle speed estimation methods in the above embodiments is implemented.

[0137] Those skilled in the art should understand that the various technical features of the above-described embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the various technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0138] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for estimating the vehicle speed, characterized in that: include: Acquiring point track data captured by a radar sensor, wherein the point track data includes a radial velocity and an azimuth angle of each target point track; determining a stationary target point track from the target point tracks based on the azimuth of each of the target point tracks; Calculating the target ground velocity corresponding to each of the stationary target points based on the radial velocity and azimuth of the stationary target point tracks; Dividing the target ground speed into speed intervals based on a preset step size to obtain a plurality of target speed intervals; Determining a target effective interval from the plurality of target speed intervals based on point trace distribution characteristic information of the detected stationary target point trace in each of the target speed intervals; The average value of the radial velocities of the stationary target points within the target effective interval is calculated to obtain an estimated value of the vehicle speed.

2. The vehicle speed estimation method according to claim 1, wherein: The step of determining a stationary target point track from the target point tracks based on the azimuth of each target point track comprises: If the azimuth angle of the target point track is within a preset angle range, a stationary target point track is determined from the target point track.

3. The vehicle speed estimation method according to claim 1, wherein: The step of acquiring the point trace data captured by the radar sensor includes: Acquire multiple frames of scene images captured by the radar sensor; Based on the multiple frames of scene images, obtaining point trace data in the multiple frames of scene images; When the number of the target traces is greater than a preset number of traces or the number of frames traversing the scene image exceeds a preset upper limit of the number of frames, the acquisition of trace data in the plurality of scene images is stopped.

4. The vehicle speed estimation method according to claim 1, wherein: The determining of a target effective interval from the plurality of target speed intervals based on the detected point trace distribution characteristic information of the stationary target point trace in each of the target speed intervals includes: Calculating the number of the stationary target traces within each of the target speed intervals; and determining a target effective interval from the plurality of target speed intervals based on the number of traces.

5. The vehicle speed estimation method according to claim 1, wherein: The method further comprises: After obtaining the estimated value of the vehicle speed, reacquiring updated point track data captured by the radar sensor, the updated point track data including the radial velocity and azimuth of each updated target point track; determining an updated stationary target point trace from the updated target point traces based on the azimuth of each of the updated target point traces; Calculating an updated target ground velocity corresponding to each of the updated stationary target points based on the radial velocity and azimuth of the updated stationary target point tracks; Based on a preset step size, the updated target ground speed is divided into speed intervals to obtain a plurality of updated target speed intervals; based on the detected point trace distribution feature information of the updated stationary target point trace within each of the updated target speed intervals, an updated target valid interval is determined from the plurality of updated target speed intervals; Calculating the average radial velocity of each of the updated stationary target points within the updated target valid interval to obtain a current vehicle speed estimate; determining a vehicle speed fluctuation range based on a preset vehicle speed fluctuation value and the estimated vehicle speed; The estimated vehicle speed is updated based on the current estimated vehicle speed and the vehicle speed fluctuation range.

6. The vehicle speed estimation method according to claim 5, characterized in that: The updating of the vehicle speed estimate based on the current vehicle speed estimate and the vehicle speed fluctuation range includes: If the current vehicle speed estimate is within the vehicle speed fluctuation range, updating the vehicle speed estimate according to the current vehicle speed estimate; If the current vehicle speed estimate exceeds the vehicle speed fluctuation range, updating the own vehicle speed estimate is stopped.

7. The vehicle speed estimation method according to claim 6, characterized in that: The method further comprises: If the number of times the update of the estimated vehicle speed is stopped exceeds a preset number of update stoppage times, a new estimated vehicle speed is recalculated.

8. A vehicle speed estimation device, characterized in that: include: A point track acquisition module is used to acquire point track data captured by the radar sensor, wherein the point track data includes the radial velocity and azimuth angle of each target point track; The point track acquisition module is further configured to determine a stationary target point track from the target point tracks based on the azimuth of each target point track; a velocity calculation module, configured to calculate the target ground velocity corresponding to each of the stationary target tracks based on the radial velocity and azimuth of the stationary target tracks; The speed calculation module is further configured to divide the target ground speed into speed intervals based on a preset step size to obtain a plurality of target speed intervals; Determining a target effective interval from the plurality of target speed intervals based on point trace distribution characteristic information of the detected stationary target point trace in each of the target speed intervals; The result generating module is used to calculate the average radial velocity of each of the stationary target points within the target effective interval to obtain an estimated value of the vehicle speed.

9. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to perform the vehicle speed estimation method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program is configured to execute the vehicle speed estimation method according to any one of claims 1 to 7 when executed.

Images