Fuzzy velocity processing method and system applied to vehicle-mounted 4D imaging millimeter wave radar

Abstract

The invention discloses a fuzzy speed processing method and system applied to a vehicle-mounted 4D imaging millimeter wave radar, and the method comprises the steps: obtaining first data corresponding to a view field region of the vehicle-mounted 4D imaging millimeter wave radar, and carrying out the filtering processing of the first data; based on the first data, creating a first target point subset and a second target point subset corresponding to the first target point subset, and generating third data corresponding to the first data in combination with vehicle speed data; matching and processing target point data in a target point subset for at least two times according to the third data in combination with historical frame data and current frame data; clustering the target points moving at the same speed, generating fourth data corresponding to the target points moving at the same speed, and generating corresponding fifth data based on the fourth data; according to the system, the platform and the storage medium corresponding to the method, the fuzzy speed is effectively solved, the computing power resource is remarkably saved, and the radar processing frame period is shortened.

Description

Technical Field

[0001] The present invention belongs to the technical field of fuzzy velocity processing of vehicle-mounted 4D imaging millimeter-wave radar, and in particular relates to a fuzzy velocity processing method and system applied to vehicle-mounted 4D imaging millimeter-wave radar. Background Art

[0002] In recent years, with the gradual implementation of a series of supporting policies and the orderly implementation of pilot projects in various places, the field of autonomous driving has developed well, and its technology research and development and application are at the forefront of the world. From a global perspective, unmanned driving / intelligent driving / autonomous driving is the most cutting-edge technology in the current automotive industry and is also one of the main application scenarios of artificial intelligence. Starting from 2021, the emergence of 4D imaging millimeter wave radar technology will not only change the field of autonomous driving, but also drive the value of the entire automotive industry and promote the construction of intelligent transportation and smart cities. In addition to the 3D measurement of distance, speed and azimuth of traditional radars, 4D imaging millimeter wave radar also has an additional 1D measurement dimension (pitch angle measurement). 4D imaging millimeter wave radar can achieve accurate perception and identification of various traffic participants such as vehicles, pedestrians, bicycles, motorcycles, etc., thereby improving traffic safety and efficiency, reducing traffic accidents and congestion, reducing energy consumption and emissions, and improving travel comfort and convenience. 4D imaging millimeter wave radar can also be combined with technologies such as Internet of Vehicles, cloud computing, and big data to achieve real-time collection, analysis and sharing of traffic information, thereby providing strong support for the planning, management and services of intelligent transportation and smart cities. In addition, 4D imaging millimeter-wave radar can also be applied to other fields, such as security, medical care, industry, agriculture, etc., showing broad market prospects and social value.

[0003] Speed ​​measurement capability is one of the 4D capabilities of 4D imaging millimeter wave radar. The ability to directly measure the Doppler speed of the target is also one of the essential characteristics of millimeter wave radar compared with other sensors such as laser radar and camera. Speed ​​measurement capability usually includes three dimensions: speed resolution, speed accuracy, and maximum unambiguous speed measurement range. This patent is aimed at improving the maximum unambiguous speed measurement range.

[0004] 4D imaging radar is required to be able to measure the relative (two vehicles relative to each other) unambiguous speed range of -110m / s to +55m / s for the target vehicle: because the upper speed limit of vehicles in most countries in the world is 55m / s (200km / h), the maximum detection speed of the vehicle equipped with radar is required to be 55m / s for vehicles traveling in the same direction, and the maximum detection speed requirement for vehicles traveling in the opposite direction is 110m / s (assuming that the radar vehicle and the target vehicle are traveling in opposite directions at a maximum speed of 55m / s respectively). Car radar defines the speed direction of the target moving away from the vehicle as positive, and the speed direction approaching the vehicle as negative. Therefore, the maximum unambiguous speed measurement range of the car radar for the target is -110m / s to +55m / s.

[0005] Speed ​​ambiguity is also called Doppler ambiguity, which means that the Doppler spectrum is aliased, causing confusion in the measured target speed, making it difficult to distinguish the target's true speed. Therefore, in order to obtain the target's true speed, speed ambiguity needs to be resolved. The point cloud of a 4D imaging radar has thousands of points per frame. At the same time, the frame period of an on-board radar is strictly limited, usually requiring 50ms to 70ms. It is necessary to perform speed ambiguity calculations on thousands of target points within this time, which places higher requirements on the speed ambiguity resolution algorithm - requiring the algorithm to be accurate while taking less time.

[0006] At present, most speed ambiguity resolution algorithms use the Chinese remainder theorem to achieve ambiguity resolution by transmitting several sets of chirp waveforms with different repetition periods. A common automotive radar solution is to transmit two sets of chirp waveforms with different repetition periods to resolve speed ambiguity, also known as AB wave speed ambiguity resolution.

[0007] Automobile radar usually adopts FMCW system, and the relationship between the maximum unambiguous speed Vmax and the chirp period T is:

[0008]

[0009] Among them, λ is the wavelength corresponding to the radar operating center frequency, which is 3.92mm at 76.5GHz.

[0010] AB wave velocity ambiguity resolution principle: If there are two sets of waveforms with different chirp repetition periods, the two sets of waveform parameters are called A wave and B wave respectively, and A wave and B wave have slightly different chirp repetition periods. Then FT1:FT2=M1:M2. Among them, M1 and M2 are mutually prime positive integers. FT1 is 1 / T1, FT2 is 1 / T2, and T1 and T2 are the chirp periods of A wave and B wave respectively. The Doppler frequency of the target is expressed in each waveform as:

[0011] f D =f d1 +m1F T1 (2)

[0012] f D =f d1 +m1F T1 (3)

[0013] According to the remainder theorem, the m1·m2 fuzzy solution can be achieved.

[0014] Assuming that the pulse period ratio of wave A and wave B is 5:7, the unambiguous ranges before deambiguation are ±35m / s and ±49m / s respectively. This set of waveform parameters can deambiguate up to 5 times. And deambiguating 3 times can reach ±147m / s, which has reached the indicator requirement of the maximum unambiguous speed range of -110m / s to +55m / s, so the final deambiguation multiple is 3. Although according to the remainder theorem, the maximum unambiguous speed that can be deambiguated by this waveform is ±245m / s (5 times ambiguity).

[0015] Assume that the frame period of wave A is T_A and the frame period of wave B is T_B. List_A and List_B store the information of the target detection points of each frame, including distance (R), speed (V), angle (θ) and signal-to-noise ratio (SNR). At the same time, according to whether the matching of wave A and wave B is successful, the points of List_A can be divided into two subsets: List_A_S (success) and List_A_U (failure). Similarly, the points of List_B are also divided into two subsets: List_B_S (success) and List_B_U (failure). In addition, when performing FFT in the speed dimension of wave A and wave B, the maximum unambiguous speed measurement values ​​are Vmax_A and Vmax_B respectively, and the speed ambiguity multiple that needs to be solved is q, and the value range of q is [-n,n], where n is the ambiguity multiple. Assume that at a certain moment, wave A is the previous frame (historical frame) and wave B is the current frame.

[0016] The traditional AB wave velocity fuzzy solution process is as follows (the following steps correspond to Figure 3 AB defuzzification, BA defuzzification and so on): Distance prediction module: For the points where the velocity blur is successfully defuzzified in wave A (previous frame), the distance is predicted and updated; Based on the historical frame, the first matching process: Taking the elements in the target point subset List_A_S that successfully matches the wave A as the base point, the matching points are searched in the target point set List_B of wave B (current frame); Based on the historical frame, the second matching process: Taking the elements in the target point subset List_A_U that did not successfully match the wave A as the base point, the matching points are searched in the target point subset List_B_U that did not successfully match the wave B; Based on the first matching of the current frame: Taking the target points in the target point subset List_B_U that did not successfully match the wave B as the base point, the matching points are searched in the target point subset List_A_S that successfully matches the wave A; Based on the second matching of the current frame: Taking the elements in the target point subset List_B_S that successfully matches the wave B as the base point, the matching points are searched in the target point subset List_B_U that did not successfully match the wave B.

[0017] The above traditional AB wave speed ambiguity solution has the following disadvantages: large amount of calculation, multiple multiplication operations are required to match the upper and lower frames of each target point, and multiplication operations are very time-consuming in DSP. At present, the number of point clouds of vehicle-mounted 4D imaging millimeter wave radar can reach several thousand points / frame, and the amount of calculation required to traverse and search for each point will increase the frame period by 30-50ms.

[0018] That is to say, the Chinese residual determination speed ambiguity solution of traditional automotive radar adopts the following steps:

[0019] (1) Distance prediction module: For the point where the velocity blur is successfully resolved in wave A (previous frame), the distance is predicted and updated;

[0020] (2) One-time matching based on historical frames: Taking the elements in the target point subset List_A_S that successfully matched wave A as the base point, search for matching points in the target point set List_B of wave B (current frame);

[0021] (3) Secondary matching based on historical frames: Taking the elements in the target point subset List_A_U that failed to match the A wave as the base point, search for matching points in the target point subset List_B_U that failed to match the B wave;

[0022] (4) A match based on the current frame: Using the target points in the target point subset List_B_U that failed to match the B wave as the base point, search for matching points in the target point subset List_A_S that successfully matched the A wave;

[0023] (5) Secondary matching based on the current frame: Taking the elements in the target point subset List_B_S where B wave matches successfully as the base point, search for matching points in the target point subset List_B_U where B wave matches unsuccessfully.

[0024] Disadvantages of traditional solutions: The amount of calculation is too large. Matching the upper and lower frames will use multiple multiplication operations, and multiplication operations are very time-consuming in DSP. Traditional solutions do not distinguish important point targets that affect radar functions to reduce the amount of signal processing; they fail to use the difference between static target points and moving target points to simplify the calculation, and each target point must use the above process for matching calculations; they fail to use the characteristics of the 4D imaging radar target point cloud cluster to reduce the amount of calculation. At present, the number of point clouds of vehicle-mounted 4D imaging millimeter-wave radars can reach several thousand points / frame. If each point is traversed and searched for matching, the amount of calculation will increase the frame period by 30-50ms. This processing time is unacceptable for mass production and installation applications.

[0025] Therefore, in view of the above technical problems and defects, it is urgent to design and develop a fuzzy velocity processing method and system for vehicle-mounted 4D imaging millimeter-wave radar. Summary of the invention

[0026] In order to overcome the shortcomings and difficulties of the above-mentioned prior art, the purpose of the present invention is to provide an improved Chinese residual determination speed ambiguity resolution scheme to address the shortcomings of the traditional speed ambiguity resolution scheme, which can significantly save computing power resources and shorten the radar processing frame cycle while effectively resolving the ambiguous speed.

[0027] The first purpose of the present invention is to provide a fuzzy velocity processing method applied to a vehicle-mounted 4D imaging millimeter-wave radar; the second purpose of the present invention is to provide a fuzzy velocity processing system applied to a vehicle-mounted 4D imaging millimeter-wave radar; the third purpose of the present invention is to provide a fuzzy velocity processing platform applied to a vehicle-mounted 4D imaging millimeter-wave radar; the fourth purpose of the present invention is to provide a computer-readable storage medium.

[0028] The first object of the present invention is achieved in that the method comprises the following steps:

[0029] Acquire first data corresponding to the field of view area of ​​the vehicle-mounted 4D imaging millimeter-wave radar, and filter and process the first data; wherein the first data is target point data within the field of view area;

[0030] Based on the first data, a corresponding first target point subset and a second target point subset are created respectively, and combined with the vehicle speed data, third data corresponding to the first data is generated; wherein the first target point subset is static target point subset data; the second target point subset is dynamic target point subset data; and the third data is predicted target distance data;

[0031] According to the third data, and in combination with the historical frame data and the current frame data, matching and processing the target point data in the target point subset at least twice;

[0032] Clustering is performed on target points moving at the same speed, and fourth data corresponding to the target points moving at the same speed is generated, and corresponding fifth data is generated based on the fourth data; wherein the fourth data is a point cloud set of the same moving object; and the fifth data is fuzzy speed processing data corresponding to the vehicle-mounted 4D imaging millimeter wave radar.

[0033] Further, the acquiring of first data corresponding to the field of view of the vehicle-mounted 4D imaging millimeter-wave radar and filtering and processing the first data further includes:

[0034] Based on the first data, determining a scene state corresponding to the first data;

[0035] According to the scene status of the target points in the combined field of view of the millimeter-wave radar, a third target point subset corresponding to the vehicle-mounted 4D imaging millimeter-wave radar is created; wherein the third first target point subset is a point set to be defuzzified.

[0036] Furthermore, the method of creating a corresponding first target point subset and a corresponding second target point subset based on the first data, and generating third data corresponding to the first data in combination with the vehicle speed data, further includes:

[0037] Generate and obtain sixth data corresponding to the vehicle-mounted 4D imaging millimeter-wave radar, and create corresponding first target point subsets and second target point subsets according to the sixth data; wherein the sixth data is point set data of the previous frame in which velocity blur is successfully resolved;

[0038] Combined with the vehicle speed data, the position data in each target point subset is predicted and updated.

[0039] Furthermore, the matching and processing of the target point data in the target point subset at least twice based on the third data and in combination with the historical frame data and the current frame data respectively, further includes:

[0040] A first matching condition is generated and obtained, and based on the first matching condition, a matching point is searched in the target point set of the current frame, and each target point is classified into a corresponding subset according to the matching condition.

[0041] Generate and obtain a second matching condition, and based on the second matching condition, use an element in a subset of dynamic target points of the historical frame as a base point to perform a one-time matching process on the first data;

[0042] A third matching condition is generated and obtained, and based on the third matching condition, the first data is processed by secondary matching with elements in a subset of dynamic target points of the historical frame as base points.

[0043] Furthermore, the matching and processing of the target point data in the target point subset at least twice based on the third data and in combination with the historical frame data and the current frame data respectively, further includes:

[0044] Generate and obtain a fourth matching condition, and based on the fourth matching condition, take the target point in the current frame subset as a base point, perform a one-time matching process on the first data;

[0045] A fifth matching condition is generated and obtained, and based on the fifth matching condition, the first data is processed by secondary matching with the target point in the current frame subset as a base point.

[0046] Further, the clustering process of the target points moving at the same speed and generating fourth data corresponding to the target points moving at the same speed, and generating corresponding fifth data based on the fourth data, further includes:

[0047] Generate and obtain seventh data corresponding to the moving target and having a moving speed and direction approximately equal to each other; wherein the seventh data is point cloud data of the target point;

[0048] Clustering the seventh data and generating corresponding fourth data with label data;

[0049] A sixth matching condition is generated and obtained, and based on the sixth matching condition, it is determined whether each target point in the fourth data is a point cloud cluster of the same target.

[0050] The second object of the present invention is achieved in this way: the system is applied to the fuzzy speed processing method applied to the vehicle-mounted 4D imaging millimeter wave radar, and the system includes:

[0051] A first data processing unit is used to obtain first data corresponding to the field of view area of ​​the vehicle-mounted 4D imaging millimeter wave radar, and filter and process the first data; wherein the first data is target point data within the field of view area;

[0052] A target point subset creation unit is used to create a corresponding first target point subset and a corresponding second target point subset based on the first data, and generate third data corresponding to the first data in combination with the vehicle speed data; wherein the first target point subset is static target point subset data; the second target point subset is dynamic target point subset data; and the third data is predicted target distance data;

[0053] A second data processing unit is used to match and process the target point data in the target point subset at least twice according to the third data and in combination with the historical frame data and the current frame data respectively;

[0054] The first data generating unit is used for clustering target points moving at the same speed, and generating fourth data corresponding to the target points moving at the same speed, and generating corresponding fifth data based on the fourth data; wherein the fourth data is a point cloud set of the same moving object; and the fifth data is fuzzy speed processing data corresponding to the vehicle-mounted 4D imaging millimeter wave radar.

[0055] Furthermore, the first data processing unit further includes:

[0056] A first determination module, configured to determine a scene state corresponding to the first data based on the first data;

[0057] A first creation module is used to create a third target point subset corresponding to the vehicle-mounted 4D imaging millimeter-wave radar according to the scene state of the target point in the combined field of view of the millimeter-wave radar; wherein the third first target point subset is a point set to be resolved;

[0058] And / or, the target point subset creation unit further includes:

[0059] A first generating module is used to generate and obtain sixth data corresponding to the vehicle-mounted 4D imaging millimeter wave radar, and to create a corresponding first target point subset and a second target point subset according to the sixth data; wherein the sixth data is point set data of the previous frame in which velocity blur is successfully resolved;

[0060] The first processing module is used to predict and update the position data of each target point subset in combination with the vehicle speed data;

[0061] And / or, the second data processing unit further includes:

[0062] The first matching module is used to generate and obtain a first matching condition, and based on the first matching condition, search for matching points in the target point set of the current frame, and classify each target point into a corresponding subset according to the matching situation.

[0063] A second matching module is used to generate and obtain a second matching condition, and based on the second matching condition, take an element in a subset of dynamic target points of the historical frame as a base point, perform a one-time matching process on the first data;

[0064] A third matching module is used to generate and obtain a third matching condition, and based on the third matching condition, perform secondary matching processing on the first data with elements in the subset of dynamic target points of the historical frame as base points;

[0065] A fourth matching module, used to generate and obtain a fourth matching condition, and based on the fourth matching condition, perform a one-time matching process on the first data with the target point in the current frame subset as a base point;

[0066] a fifth matching module, configured to generate and obtain a fifth matching condition, and based on the fifth matching condition, perform secondary matching processing on the first data with the target point in the current frame subset as a base point;

[0067] And / or, the first data generating unit further includes:

[0068] A second generating module is used to generate and obtain seventh data corresponding to the moving target and having a moving speed and direction approximately equal to each other; wherein the seventh data is point cloud data of the target point;

[0069] A third generating module, used for clustering the seventh data and generating corresponding fourth data with label data;

[0070] The second determination module is used to generate and obtain a sixth matching condition, and determine whether each target point in the fourth data is a point cloud cluster of the same target based on the sixth matching condition.

[0071] The third object of the present invention is achieved as follows: it includes a processor, a memory and a fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar; wherein the fuzzy speed processing platform control program applied to the vehicle-mounted 4D imaging millimeter-wave radar is executed in the processor, the fuzzy speed processing platform control program applied to the vehicle-mounted 4D imaging millimeter-wave radar is stored in the memory, and the fuzzy speed processing platform control program applied to the vehicle-mounted 4D imaging millimeter-wave radar implements the fuzzy speed processing method applied to the vehicle-mounted 4D imaging millimeter-wave radar.

[0072] The fourth object of the present invention is achieved as follows: the computer-readable storage medium stores a fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar, and the fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar implements the fuzzy speed processing method applied to a vehicle-mounted 4D imaging millimeter-wave radar.

[0073] The present invention obtains first data corresponding to the field of view area of ​​the vehicle-mounted 4D imaging millimeter wave radar through a method, and filters and processes the first data; wherein the first data is target point data within the field of view area;

[0074] Based on the first data, a corresponding first target point subset and a second target point subset are created respectively, and combined with the vehicle speed data, third data corresponding to the first data is generated; wherein the first target point subset is static target point subset data; the second target point subset is dynamic target point subset data; and the third data is predicted target distance data;

[0075] According to the third data, and in combination with the historical frame data and the current frame data, matching and processing the target point data in the target point subset at least twice;

[0076] Clustering is performed on target points moving at the same speed, and fourth data corresponding to the target points moving at the same speed is generated, and corresponding fifth data is generated based on the fourth data; wherein the fourth data is a point cloud set of the same moving object; the fifth data is fuzzy speed processing data corresponding to the vehicle-mounted 4D imaging millimeter-wave radar, as well as a system, platform and storage medium corresponding to the method, which significantly saves computing power resources while effectively solving the fuzzy speed and shortens the radar processing frame cycle.

[0077] In other words, compared with the traditional solution, the solution of the present invention makes full use of the radar installation location and the scene area information of autonomous driving, and filters out a large number of redundant and minor target points. The static target points with the highest proportion are matched separately, and the point cloud clusters of moving targets are simplified, which greatly reduces the amount of data to be processed and shortens the algorithm running time. While effectively resolving the fuzzy speed, it saves computing power resources and shortens the radar frame cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0079] Figure 1 This is a schematic diagram of the FMCW waveform of the automotive millimeter wave radar;

[0080] Figure 2 This is a schematic diagram of a 2AB waveform, with one frame of A wave, followed by one frame of B wave, and then another frame of A wave, and so on;

[0081] Figure 3 The schematic diagram of the radar AB wave cycles back and forth, and the AB deambiguation and BA deambiguation also cycles back and forth;

[0082] Figure 4 This is a schematic diagram of the traditional AB wave defuzzification process;

[0083] Figure 5 It is a schematic diagram of an improved AB wave deambiguation process of an embodiment of a fuzzy velocity processing method applied to a vehicle-mounted 4D imaging millimeter wave radar of the present invention;

[0084] Figure 6 It is a flow chart of a fuzzy velocity processing method applied to a vehicle-mounted 4D imaging millimeter wave radar according to the present invention;

[0085] Figure 7 This is a schematic diagram of the architecture of a fuzzy velocity processing system applied to a vehicle-mounted 4D imaging millimeter-wave radar according to the present invention;

[0086] Figure 8 This is a schematic diagram of the fuzzy velocity processing platform architecture applied to vehicle-mounted 4D imaging millimeter-wave radar of the present invention;

[0087] Fig. 9 A schematic diagram of a computer-readable storage medium architecture in an embodiment of the present invention. DETAILED DESCRIPTION

[0088] In order to better understand the purpose, technical solutions and advantages of the present invention, the present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification.

[0089] The present invention may also be implemented or applied through other different specific examples, and the details in this specification may also be modified and changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention.

[0090] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0091] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. Secondly, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0092] Preferably, the fuzzy speed processing method for vehicle-mounted 4D imaging millimeter wave radar of the present invention is applied in one or more terminals or servers. The terminal is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), digital signal processors (DSP), embedded devices, etc.

[0093] The terminal can be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal can interact with the client through a keyboard, a mouse, a remote control, a touch pad, or a voice control device.

[0094] The present invention is to realize a fuzzy velocity processing method, system, platform and storage medium applied to a vehicle-mounted 4D imaging millimeter wave radar.

[0095] like Figure 6 , which is a flow chart of a fuzzy velocity processing method applied to a vehicle-mounted 4D imaging millimeter-wave radar provided by an embodiment of the present invention.

[0096] In this embodiment, the fuzzy speed processing method applied to the vehicle-mounted 4D imaging millimeter-wave radar can be applied to a terminal with a display function or a fixed terminal. The terminal is not limited to a personal computer, a smart phone, a tablet computer, a desktop computer or an all-in-one computer equipped with a camera, etc.

[0097] The fuzzy speed processing method applied to the vehicle-mounted 4D imaging millimeter wave radar can also be applied to a hardware environment composed of a terminal and a server connected to the terminal through a network. The network includes but is not limited to: a wide area network, a metropolitan area network or a local area network. The fuzzy speed processing method applied to the vehicle-mounted 4D imaging millimeter wave radar of the embodiment of the present invention can be executed by a server, can be executed by a terminal, or can be executed by both a server and a terminal.

[0098] For example, for a terminal that needs to perform fuzzy speed processing for a vehicle-mounted 4D imaging millimeter-wave radar, the fuzzy speed processing function for a vehicle-mounted 4D imaging millimeter-wave radar provided by the method of the present invention can be directly integrated on the terminal, or a client for implementing the method of the present invention can be installed. For another example, the method provided by the present invention can also be run on a server or other device in the form of a software development kit (SDK), and an interface for the fuzzy speed processing function for a vehicle-mounted 4D imaging millimeter-wave radar is provided in the form of an SDK. The terminal or other device can implement the fuzzy speed processing function for a vehicle-mounted 4D imaging millimeter-wave radar through the provided interface. The present invention is further described below in conjunction with the accompanying drawings.

[0099] like Figure 1-Figure 6 As shown, the present invention provides a fuzzy velocity processing method applied to a vehicle-mounted 4D imaging millimeter wave radar, wherein the method comprises the following steps:

[0100] S1. Acquire first data corresponding to the field of view area of ​​the vehicle-mounted 4D imaging millimeter-wave radar, and filter and process the first data; wherein the first data is target point data within the field of view area;

[0101] S2. Based on the first data, respectively create a corresponding first target point subset and a second target point subset, and combine the vehicle speed data to generate third data corresponding to the first data; wherein the first target point subset is static target point subset data; the second target point subset is dynamic target point subset data; and the third data is predicted target distance data;

[0102] S3, matching and processing target point data in the target point subset at least twice according to the third data and in combination with the historical frame data and the current frame data respectively;

[0103] S4. Clustering the target points moving at the same speed, and generating fourth data corresponding to the target points moving at the same speed, and generating corresponding fifth data based on the fourth data; wherein the fourth data is a point cloud set of the same moving object; and the fifth data is fuzzy speed processing data corresponding to the vehicle-mounted 4D imaging millimeter wave radar.

[0104] The acquiring of first data corresponding to the field of view of the vehicle-mounted 4D imaging millimeter-wave radar and filtering and processing the first data further includes:

[0105] S11. Based on the first data, determining a scene state corresponding to the first data;

[0106] S12. Create a third target point subset corresponding to the vehicle-mounted 4D imaging millimeter-wave radar according to the scene state of the target points in the combined field of view of the millimeter-wave radar; wherein the third first target point subset is a point set to be defuzzified.

[0107] The creating a corresponding first target point subset and a corresponding second target point subset based on the first data, and combining the vehicle speed data to generate third data corresponding to the first data, further includes:

[0108] S21, generating and acquiring sixth data corresponding to the vehicle-mounted 4D imaging millimeter-wave radar, and creating corresponding first target point subsets and second target point subsets according to the sixth data; wherein the sixth data is point set data of the previous frame in which velocity blurring is successfully resolved;

[0109] S22. Combine the vehicle speed data to predict and update the position data of each target point subset.

[0110] The matching and processing of the target point data in the target point subset at least twice based on the third data and in combination with the historical frame data and the current frame data respectively, further includes:

[0111] S31, generating and acquiring a first matching condition, and based on the first matching condition, searching for matching points in the target point set of the current frame, and classifying each target point into a corresponding subset according to the matching situation.

[0112] S32, generating and acquiring a second matching condition, and based on the second matching condition, taking an element in a subset of dynamic target points of the historical frame as a base point, performing a one-time matching process on the first data;

[0113] S33, generating and acquiring a third matching condition, and based on the third matching condition, performing secondary matching processing on the first data with elements in a subset of dynamic target points in the historical frame as base points.

[0114] The matching and processing of the target point data in the target point subset at least twice based on the third data and in combination with the historical frame data and the current frame data respectively, further includes:

[0115] S34, generating and acquiring a fourth matching condition, and based on the fourth matching condition, taking the target point in the current frame subset as a base point, performing a matching process on the first data;

[0116] S35 , generating and acquiring a fifth matching condition, and based on the fifth matching condition, performing secondary matching processing on the first data with the target point in the current frame subset as a base point.

[0117] The clustering process is to process the target points moving at the same speed, and generate fourth data corresponding to the target points moving at the same speed, and generate corresponding fifth data based on the fourth data, and further includes:

[0118] S41, generating and acquiring seventh data corresponding to the moving target and having a moving speed and direction approximately equal to each other; wherein the seventh data is point cloud data of the target point;

[0119] S42, clustering the seventh data and generating corresponding fourth data with label data;

[0120] S43: Generate and obtain a sixth matching condition, and based on the sixth matching condition, determine whether each target point in the fourth data is a point cloud cluster of the same target.

[0121] Specifically, in the embodiment of the present invention, the improved Chinese residual determination speed fuzzy solution for vehicle-mounted 4D imaging radar introduced in this solution utilizes the following useful information:

[0122] A. The on-board 4D imaging radar is installed near the front logo of the vehicle, on the longitudinal center axis of the vehicle. The maximum FOV area of ​​the radar is within ±60° in front of the vehicle, and is mainly used for the automatic driving function in front of the vehicle. It mainly focuses on the targets on the lanes within the guardrails on both sides of the vehicle, or the targets within 50m to the left and right in front of the vehicle on sections without roadside guardrails. This information can be used for filtering once, significantly reducing the number of target points that need to be defuzzified.

[0123] B. At least 80% of the target points detected by the vehicle-mounted radar are static target points, such as guardrails, street lamp poles, buildings on both sides of the road, trees, etc. The main amount of computation for the remaining defuzzification is consumed in matching. Fast matching of static targets can significantly reduce the amount of algorithm computation.

[0124] C. The on-board 4D imaging radar has a high resolution, and the detected moving targets are usually in the form of point cloud clusters / a group of points. The point cloud clusters of the same target are close in distance and have almost the same speed. This feature can be used to cluster the point cloud clusters of the same moving target into one point. When the algorithm performs speed matching, the speed of one point is used instead of the point cloud cluster to reduce the amount of calculation.

[0125] The improved Chinese residual determination velocity fuzzy process applied to vehicle-mounted 4D imaging millimeter wave radar in this scheme is as follows:

[0126] (1) Preliminary point filtering: Filter the points detected in each frame that are outside the focus area. First, perform scene judgment. If the guardrail cannot be fitted using the point position information, the points outside ±50m in front of the vehicle (radar vehicle) are not included in the unambiguous point set List_A or List_B; if the guardrail is fitted, the points outside the guardrail are not included in List_A and List_B.

[0127] (2) Improved distance prediction module: The remaining A wave (previous frame) points that successfully resolve velocity ambiguity are divided into two subsets: static target point subset List_A_S_0 and dynamic target point subset List_A_S_1. Combined with the ego vehicle speed V_ego information, assuming that the points in List_A_S_0 and List_A_S_1 are stationary in the next frame, their positions are predicted and updated:

[0128] R' A =R A +V_ego·T A (4)

[0129] In the above formula, V_ego is the vehicle speed, R A is the target distance before prediction, R A ' is the predicted target distance.

[0130] (3) Matching static target points: For the points whose positions are updated in step (2), search for matching points in List_B (the target point set of this frame). If the two match, the successfully matched points in List_B are added to the List_B_S_0 subset (the static point set that is successfully matched in the B frame). The conditions for determining whether the match is successful are as follows:

[0131] THR=|R′ A -R B |;THV=|V B |;THθ=|θ A -θ B |;THS=|SNR A -SNR B |; THR<TH0_R; THV<TH0_V; THθ<TH0_A; THS<TH0_S.

[0132] THR, THV, THθ, THS are distance difference, speed difference, angle difference, SNR difference respectively; TH0_R, TH0_V, TH0_A, TH0_S are distance threshold, speed threshold, angle threshold, SNR threshold respectively. Among them, TH0_R can be set to 3 distance units, TH0_V can be set to 3 speed units, TH0_A can be set to 10°, TH1_S can be set to within 10dB, etc. These thresholds need to be further adjusted according to the measured data.

[0133] After this round of search, if some target points in the A-wave subset List_A_S_0 are not matched successfully, they will be included in List_A_U (the subset of target points in the previous frame that were not matched successfully);

[0134] (4) One-time matching based on historical frames: Using the elements in the subset List_A_S_1 of the dynamic target points of the historical frame A wave as the base point, search for matching points in the set of List_B except List_B_S_0. If the two match, the successfully matched points in List_B are included in the subset List_B_S_1 (the set of motion points successfully matched in the B frame), and the fuzzy multiple q when the match is successful is saved. B Value. The unsuccessful matching points in List_B are included in the List_B_U subset. In addition, after this round of search, if some target points in the A wave subset List_A_S_1 are not matched successfully, they are included in List_A_U (the subset of unmatched target points in the previous frame). The conditions for judging whether the match is successful are as follows:

[0135] V′ Bi =V B +q i 2Vmax_B; V′A =V A +q A 2Vmax_A; THR = |R′ A -R B |;

[0136] THV i =|V′ A -V Bi '|;THθ=|θ A -θ B |;THS=|SNR A -SNR B |;THR<TH1_R;

[0137] THV i <TH1_V; THθ<TH1_A; THS<TH1_S; q i =argmin qi (a1·THR+a2·THV i +a3·THθ+a4·THS).

[0138] q i Make the weighted sum in the brackets of the above formula take the minimum value, q i Traverse [-n,n], V' A is the target point moving speed saved in List_A_S_1. THR, THθ, THS are distance difference, angle difference, SNR difference respectively; THV i The fuzzy multiple of the B wave is q i TH1_R, TH1_V, TH1_A, and TH1_S are the distance threshold, speed threshold, angle threshold, and SNR threshold, respectively. Among them, TH1_R can be set to 3 distance units, TH1_V can be set to 3 speed units, TH1_A can be set to different values ​​according to different distances, and TH1_S can be set to within 15dB. These thresholds need to be further adjusted according to the measured data.

[0139] (5) Secondary matching based on historical frames: Take the target point in the historical frame A wave subset List_A_U as the base point and search for matching points in the List_B subset List_B_U. If the two match, the matching points in the subset List_B_U are included in the List_B_S subset, and the fuzzy multiple q when the match is successful is saved. B value;

[0140] V′ Bi =V B +q i 2Vmax_B; V′ Aj =V A +qj 2Vmax_A; R′ Aj =R A +V Aj '·T A ;

[0141] THR=|R′ Aj -R B |;THV i =|V′ Aj -V Bi '|;THθ=|θ A -θ B |;THS=|SNR A -SNR B |

[0142] THR<TH2_R;THV ij <TH2_V; THθ<TH2_A; THS<TH2_S;

[0143] (q i ,q j ) = argmin qi,qj (a1·THR+a2·THV ij +a3·THθ+a4·THS).

[0144] q i Make the weighted sum in the brackets of the above formula take the minimum value, q i Traverse [-n,n], q j Traversing [-n, n], THR, THθ, THS are distance difference, angle difference, SNR difference respectively; THV ij The fuzzy multiple of the B wave is q i and the A wave fuzziness factor is q j TH2_R, TH2_V, TH2_A, and TH2_S are the distance threshold, speed threshold, angle threshold, and SNR threshold, respectively. Among them, TH2_R can be set to 6 distance units, TH2_V can be set to 6 speed units, TH2_A can be set to different values ​​according to different distances, and TH2_S can be set to within 15dB. These thresholds need to be further adjusted according to the measured data.

[0145] (6) A match based on the current frame: After processing in step 5), the target point in the B wave subset List_B_U of the current frame is used as the base point to search for matching points in the subset List_A_S_1 of List_A. If the two match, the successfully matched points in the subset List_B_U are included in the subset List_B_S_1, and the fuzzy multiple q when the match is successful is saved. B Value. The conditions for judging whether the match is successful are as follows:

[0146] V′ Bi =V B +q i 2Vmax_B; THR = |R′ A -R B |;THV i =|V′ A -V Bi '|;THθ=|θ A -θ B |;THS=|SNR A -SNR B |;THR<TH3_R;THV i <TH3_V; THθ<TH3_A; THS<TH3_S; q i =argmin qi (a1·THR+a2·THV i +a3·THθ+a4·THS).

[0147] q i Make the weighted sum in the brackets of the above formula take the minimum value, qi traverses [-n,n], THR, THθ, THS are distance difference, angle difference, SNR difference respectively; THV i The fuzzy multiple of the B wave is q i TH3_R, TH3_V, TH3_A, and TH3_S are the distance threshold, speed threshold, angle threshold, and SNR threshold, respectively. Among them, TH3_R can be set to 6 distance units, TH1_V can be set to 6 speed units, TH3_A can be set to different values ​​according to different distances, and TH3_S can be set to within 15dB. These thresholds need to be further adjusted according to the measured data.

[0148] (7) Secondary matching based on the current frame: After processing in step 6), the target point in the B wave subset List_B_S_1 of the current frame is used as the base point to search for matching points in the subset List_B_U of List_B. If the two match, the successfully matched points in the subset List_B_U are included in the subset List_B_S_1, and the fuzzy multiple q when the match is successful is saved. B The conditions for determining whether a match is successful are as follows:

[0149] |R BS -R BU |<TH4_R;|V BS -V BU |<TH4_V;|θ BS -θ BU |<TH4_A;|SNR BS -SNRBU |<TH4_S.

[0150] In the above formula, TH4_R, TH4_V, TH4_A, and TH4_S are the distance threshold, speed threshold, angle threshold, and SNR threshold, respectively. Among them, TH4_R can be set to 10 distance units, TH4_V can be set to 3 speed units, and can be set to different values ​​according to different distances, and TH4_S can be set to within 15dB, etc. These threshold values ​​are for reference only and need to be adjusted according to actual measured data. BS is the target distance of the B wave subset List_B_S_1, R BU is the target distance of the B wave subset List_B_U, V BS is the target velocity of the B-wave subset List_B_S_1, V BU is the target velocity of the B-wave subset List_B_U, θ BS is the target angle of the B-wave subset List_B_S_1, θ BU It is the target angle of the B-wave subset List_B_U, where the target angle includes the azimuth angle and the elevation angle.

[0151] (8) Clustering of points moving at the same speed: Moving targets such as cars and trucks at medium and short distances will produce a cluster of point clouds with close distances and approximately equal speeds and directions. This feature is used to cluster the points in List_B_S_1, filter out the point cloud set belonging to the same moving object, and label it. The points in the point cloud set share the same speed (including the same fuzzy multiple q B ). The conditions for judging whether point i and point j are point cloud clusters of the same target are as follows:

[0152] THR=|R Bi -R Bj |;THV ij =|V' Bi -V Bj '|;THθ=|θ Bi -θ Bj |;THS=|SNR Bi -SNR Bj |;THR<TH5_R;THV ij <TH5_V; THθ<TH5_A; THS<TH5_S.

[0153] In the above formula, TH5_R, TH5_V, TH5_A, and TH5_S are the distance threshold, speed threshold, angle threshold, and SNR threshold, respectively. Among them, TH5_R can be set to 5 distance units, TH5_V can be set to 3 speed units, TH5_A can be set to 3 angle units, and TH5_S can be set to within 10dB. These threshold values ​​are only for reference and need to be adjusted according to the actual measured data. If they are satisfied, the speed information of i and j is merged.

[0154] That is to say, the Chinese residual definition solution speed fuzzy solution of the present invention is:

[0155] (1) Preliminary point filtering: Filter the points detected in each frame that are outside the focus area. First, perform scene judgment. If the guardrail cannot be fitted using the point position information, the points outside ±50m in front of the vehicle (radar vehicle) are not included in the unambiguous point set List_A or List_B; if the guardrail is fitted, the points outside the guardrail are not included in List_A and List_B.

[0156] (2) Improved distance prediction module: The remaining A-wave (previous frame) point set that successfully resolves velocity ambiguity is divided into two subsets: the static target point subset List_A_S_0 and the dynamic target point subset List_A_S_1. Combined with the ego vehicle speed information V_ego, it is assumed that the points in List_A_S_0 and List_A_S_1 are stationary in the next frame, and their positions are predicted and updated.

[0157] (3) For the point whose position is updated in step (2), search for matching points in List_B (the target point set of this frame). If the two match, the successfully matched points in List_B are added to the List_B_S_0 subset (the static point set that successfully matches in frame B).

[0158] (4) One-time matching based on historical frames: Using the elements in the subset List_A_S_1 of the dynamic target points of the historical frame A wave as the base point, search for matching points in the set of List_B except List_B_S_0. If the two match, the successfully matched points in List_B are included in the subset List_B_S_1 (the set of motion points successfully matched in the B frame), and the fuzzy multiple q when the match is successful is saved. B Value. The unsuccessful matching points in List_B are included in the List_B_U subset. In addition, after this round of search, if some target points in the A wave subset List_A_S_1 are not matched successfully, they are included in List_A_U (the subset of unmatched target points in the previous frame).

[0159] (5) Secondary matching based on historical frames: Take the target point in the historical frame A wave subset List_A_U as the base point and search for matching points in the List_B subset List_B_U. If the two match, the matching points in the subset List_B_U are included in the List_B_S subset, and the fuzzy multiple q when the match is successful is saved. B value.

[0160] (6) A match based on the current frame: After processing in step 5), the target point in the B wave subset List_B_U of the current frame is used as the base point to search for matching points in the subset List_A_S_1 of List_A. If the two match, the successfully matched points in the subset List_B_U are included in the subset List_B_S_1, and the fuzzy multiple q when the match is successful is saved. B value.

[0161] (7) Secondary matching based on the current frame: After processing in step 6), the target point in the B wave subset List_B_S_1 of the current frame is used as the base point to search for matching points in the subset List_B_U of List_B. If the two match, the successfully matched points in the subset List_B_U are included in the subset List_B_S_1, and the fuzzy multiple q when the match is successful is saved. B .

[0162] (8) Clustering of points moving at the same speed: Moving targets such as cars and trucks at medium and short distances will produce a cluster of point clouds with close distances and approximately equal speeds and directions. This feature is used to cluster the points in List_B_S_1, filter out the point cloud set belonging to the same moving object, and label it. The points in the point cloud set share the same speed (including the same fuzzy multiple q B ).

[0163] To achieve the above object, the present invention also provides a fuzzy speed processing system for vehicle-mounted 4D imaging millimeter wave radar, such as Figure 7 As shown, the system is applied to the fuzzy speed processing method applied to the vehicle-mounted 4D imaging millimeter wave radar, and the system specifically includes:

[0164] A first data processing unit is used to obtain first data corresponding to the field of view area of ​​the vehicle-mounted 4D imaging millimeter wave radar, and filter and process the first data; wherein the first data is target point data within the field of view area;

[0165] A target point subset creation unit is used to create a corresponding first target point subset and a corresponding second target point subset based on the first data, and generate third data corresponding to the first data in combination with the vehicle speed data; wherein the first target point subset is static target point subset data; the second target point subset is dynamic target point subset data; and the third data is predicted target distance data;

[0166] A second data processing unit is used to match and process the target point data in the target point subset at least twice according to the third data and in combination with the historical frame data and the current frame data respectively;

[0167] The first data generating unit is used for clustering target points moving at the same speed, and generating fourth data corresponding to the target points moving at the same speed, and generating corresponding fifth data based on the fourth data; wherein the fourth data is a point cloud set of the same moving object; and the fifth data is fuzzy speed processing data corresponding to the vehicle-mounted 4D imaging millimeter wave radar.

[0168] The first data processing unit further includes:

[0169] A first determination module, configured to determine a scene state corresponding to the first data based on the first data;

[0170] A first creation module is used to create a third target point subset corresponding to the vehicle-mounted 4D imaging millimeter-wave radar according to the scene state of the target point in the combined field of view of the millimeter-wave radar; wherein the third first target point subset is a point set to be resolved;

[0171] And / or, the target point subset creation unit further includes:

[0172] A first generating module is used to generate and obtain sixth data corresponding to the vehicle-mounted 4D imaging millimeter wave radar, and to create a corresponding first target point subset and a second target point subset according to the sixth data; wherein the sixth data is point set data of the previous frame in which velocity blur is successfully resolved;

[0173] The first processing module is used to predict and update the position data of each target point subset in combination with the vehicle speed data;

[0174] And / or, the second data processing unit further includes:

[0175] The first matching module is used to generate and obtain a first matching condition, and based on the first matching condition, search for matching points in the target point set of the current frame, and classify each target point into a corresponding subset according to the matching situation.

[0176] A second matching module is used to generate and obtain a second matching condition, and based on the second matching condition, take an element in a subset of dynamic target points of the historical frame as a base point, perform a one-time matching process on the first data;

[0177] A third matching module is used to generate and obtain a third matching condition, and based on the third matching condition, perform secondary matching processing on the first data with elements in the subset of dynamic target points of the historical frame as base points;

[0178] A fourth matching module, used to generate and obtain a fourth matching condition, and based on the fourth matching condition, perform a one-time matching process on the first data with the target point in the current frame subset as a base point;

[0179] a fifth matching module, configured to generate and obtain a fifth matching condition, and based on the fifth matching condition, perform secondary matching processing on the first data with the target point in the current frame subset as a base point;

[0180] And / or, the first data generating unit further includes:

[0181] A second generating module is used to generate and obtain seventh data corresponding to the moving target and having a moving speed and direction approximately equal to each other; wherein the seventh data is point cloud data of the target point;

[0182] A third generating module, used for clustering the seventh data and generating corresponding fourth data with label data;

[0183] The second determination module is used to generate and obtain a sixth matching condition, and determine whether each target point in the fourth data is a point cloud cluster of the same target based on the sixth matching condition.

[0184] In the system solution embodiment of the present invention, the method steps involved in the fuzzy velocity processing applied to the vehicle-mounted 4D imaging millimeter-wave radar have been described above in detail, that is, the functional modules in the system are used to implement the steps or sub-steps in the above method embodiment, which will not be repeated here.

[0185] To achieve the above object, the present invention also provides a fuzzy velocity processing platform for vehicle-mounted 4D imaging millimeter wave radar, such as Figure 8 As shown, it includes a processor, a memory, and a fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar; wherein the fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar is executed by the processor, the fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar is stored in the memory, and the fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar implements the fuzzy speed processing method steps applied to a vehicle-mounted 4D imaging millimeter-wave radar. For example:

[0186] S1. Acquire first data corresponding to the field of view area of ​​the vehicle-mounted 4D imaging millimeter-wave radar, and filter and process the first data; wherein the first data is target point data within the field of view area;

[0187] S2. Based on the first data, respectively create a corresponding first target point subset and a second target point subset, and combine the vehicle speed data to generate third data corresponding to the first data; wherein the first target point subset is static target point subset data; the second target point subset is dynamic target point subset data; and the third data is predicted target distance data;

[0188] S3, matching and processing target point data in the target point subset at least twice according to the third data and in combination with the historical frame data and the current frame data respectively;

[0189] S4. Clustering the target points moving at the same speed, and generating fourth data corresponding to the target points moving at the same speed, and generating corresponding fifth data based on the fourth data; wherein the fourth data is a point cloud set of the same moving object; and the fifth data is fuzzy speed processing data corresponding to the vehicle-mounted 4D imaging millimeter wave radar.

[0190] The specific details of the steps have been explained above and will not be repeated here.

[0191] In the embodiment of the present invention, the built-in processor of the fuzzy speed processing platform applied to the vehicle-mounted 4D imaging millimeter wave radar can be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of integrated circuits with the same or different functions, including one or more central processing units (CPU), microprocessors, digital processing chips, graphics processors, and a combination of various control chips. The processor uses various interfaces and lines to connect various components, and executes or executes programs or units stored in the memory, and calls data stored in the memory to perform various functions and process data of the fuzzy speed processing applied to the vehicle-mounted 4D imaging millimeter wave radar;

[0192] The memory is used to store program codes and various data. It is installed in the fuzzy speed processing platform used for vehicle-mounted 4D imaging millimeter-wave radar, and realizes high-speed and automatic access to programs or data during operation.

[0193] The memory includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electronically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0194] To achieve the above object, the present invention also provides a computer readable storage medium, such as Fig. 9 As shown, the computer-readable storage medium stores a fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar, and the fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar implements the fuzzy speed processing method steps applied to a vehicle-mounted 4D imaging millimeter-wave radar, for example:

[0195] S1. Acquire first data corresponding to the field of view area of ​​the vehicle-mounted 4D imaging millimeter-wave radar, and filter and process the first data; wherein the first data is target point data within the field of view area;

[0196] S2. Based on the first data, respectively create a corresponding first target point subset and a second target point subset, and combine the vehicle speed data to generate third data corresponding to the first data; wherein the first target point subset is static target point subset data; the second target point subset is dynamic target point subset data; and the third data is predicted target distance data;

[0197] S3, matching and processing target point data in the target point subset at least twice according to the third data and in combination with the historical frame data and the current frame data respectively;

[0198] S4. Clustering the target points moving at the same speed, and generating fourth data corresponding to the target points moving at the same speed, and generating corresponding fifth data based on the fourth data; wherein the fourth data is a point cloud set of the same moving object; and the fifth data is fuzzy speed processing data corresponding to the vehicle-mounted 4D imaging millimeter wave radar.

[0199] The specific details of the steps have been explained above and will not be repeated here.

[0200] In the description of the embodiments of the present invention, it should be noted that any process or method description in the flowchart or otherwise described herein may be understood as representing a module, fragment or portion of a code comprising one or more executable instructions for implementing steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations, in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present invention belong.

[0201] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processing module, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM).

[0202] In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0203] The present invention obtains first data corresponding to the field of view area of ​​the vehicle-mounted 4D imaging millimeter wave radar through a method, and filters and processes the first data; wherein the first data is target point data within the field of view area;

[0204] Based on the first data, a corresponding first target point subset and a second target point subset are created respectively, and combined with the vehicle speed data, third data corresponding to the first data is generated; wherein the first target point subset is static target point subset data; the second target point subset is dynamic target point subset data; and the third data is predicted target distance data;

[0205] According to the third data, and in combination with the historical frame data and the current frame data, matching and processing the target point data in the target point subset at least twice;

[0206] Clustering is performed on target points moving at the same speed, and fourth data corresponding to the target points moving at the same speed is generated, and corresponding fifth data is generated based on the fourth data; wherein the fourth data is a point cloud set of the same moving object; the fifth data is fuzzy speed processing data corresponding to the vehicle-mounted 4D imaging millimeter-wave radar, as well as a system, platform and storage medium corresponding to the method, which significantly saves computing power resources while effectively solving the fuzzy speed and shortens the radar processing frame cycle.

[0207] In other words, compared with the traditional solution, the solution of the present invention makes full use of the radar installation location and the scene area information of autonomous driving, and filters out a large number of redundant and minor target points. The static target points with the highest proportion are matched separately, and the point cloud clusters of moving targets are simplified, which greatly reduces the amount of data to be processed and shortens the algorithm running time. While effectively resolving the fuzzy speed, it saves computing power resources and shortens the radar frame cycle.

[0208] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A fuzzy velocity processing method applied to a vehicle-mounted 4D imaging millimeter wave radar, characterized in that: The method comprises: Acquire first data corresponding to the field of view area of ​​the vehicle-mounted 4D imaging millimeter-wave radar, and filter and process the first data; wherein the first data is target point data within the field of view area; Based on the first data, a corresponding first target point subset and a second target point subset are created respectively, and combined with the vehicle speed data, third data corresponding to the first data is generated; wherein the first target point subset is static target point subset data; the second target point subset is dynamic target point subset data; and the third data is predicted target distance data; According to the third data, and in combination with the historical frame data and the current frame data, matching and processing the target point data in the target point subset at least twice; Clustering is performed on target points moving at the same speed, and fourth data corresponding to the target points moving at the same speed is generated, and corresponding fifth data is generated based on the fourth data; wherein the fourth data is a point cloud set of the same moving object; and the fifth data is fuzzy speed processing data corresponding to the vehicle-mounted 4D imaging millimeter wave radar.

2. According to claim 1, a fuzzy velocity processing method applied to a vehicle-mounted 4D imaging millimeter wave radar is characterized in that: The acquiring of first data corresponding to the field of view of the vehicle-mounted 4D imaging millimeter-wave radar and filtering and processing the first data further includes: Based on the first data, determining a scene state corresponding to the first data; According to the scene status of the target points in the combined field of view of the millimeter-wave radar, a third target point subset corresponding to the vehicle-mounted 4D imaging millimeter-wave radar is created; wherein the third first target point subset is a point set to be defuzzified.

3. According to claim 1, a fuzzy velocity processing method applied to a vehicle-mounted 4D imaging millimeter wave radar is characterized in that: The creating a corresponding first target point subset and a corresponding second target point subset based on the first data, and combining the vehicle speed data to generate third data corresponding to the first data, further includes: Generate and obtain sixth data corresponding to the vehicle-mounted 4D imaging millimeter-wave radar, and create corresponding first target point subsets and second target point subsets according to the sixth data; wherein the sixth data is point set data of the previous frame in which velocity blur is successfully resolved; Combined with the vehicle speed data, the position data in each target point subset is predicted and updated.

4. The fuzzy velocity processing method for vehicle-mounted 4D imaging millimeter wave radar according to claim 1 is characterized in that: The matching and processing of the target point data in the target point subset at least twice based on the third data and in combination with the historical frame data and the current frame data respectively, further includes: Generate and obtain a first matching condition, and based on the first matching condition, search for matching points in the target point set of the current frame, and classify each target point into a corresponding subset according to the matching condition; Generate and obtain a second matching condition, and based on the second matching condition, use an element in a subset of dynamic target points of the historical frame as a base point to perform a one-time matching process on the first data; A third matching condition is generated and obtained, and based on the third matching condition, the first data is processed by secondary matching with elements in a subset of dynamic target points of the historical frame as base points.

5. A fuzzy velocity processing method for vehicle-mounted 4D imaging millimeter wave radar according to claim 1 or 4, characterized in that: The matching and processing of the target point data in the target point subset at least twice based on the third data and in combination with the historical frame data and the current frame data respectively, further includes: Generate and obtain a fourth matching condition, and based on the fourth matching condition, take the target point in the current frame subset as a base point, perform a one-time matching process on the first data; A fifth matching condition is generated and obtained, and based on the fifth matching condition, the first data is processed by secondary matching with the target point in the current frame subset as a base point.

6. The fuzzy velocity processing method for vehicle-mounted 4D imaging millimeter wave radar according to claim 1, characterized in that: The clustering process is to process the target points moving at the same speed, and generate fourth data corresponding to the target points moving at the same speed, and generate corresponding fifth data based on the fourth data, and further includes: Generate and obtain seventh data corresponding to the moving target and having a moving speed and direction approximately equal to each other; wherein the seventh data is point cloud data of the target point; Clustering the seventh data and generating corresponding fourth data with label data; A sixth matching condition is generated and obtained, and based on the sixth matching condition, it is determined whether each target point in the fourth data is a point cloud cluster of the same target.

7. A fuzzy velocity processing system applied to vehicle-mounted 4D imaging millimeter wave radar, characterized in that: The system is applied to the fuzzy speed processing method applied to the vehicle-mounted 4D imaging millimeter wave radar as claimed in any one of claims 1 to 6, and the system includes: A first data processing unit is used to obtain first data corresponding to the field of view area of ​​the vehicle-mounted 4D imaging millimeter wave radar, and filter and process the first data; wherein the first data is target point data within the field of view area; A target point subset creation unit is used to create a corresponding first target point subset and a corresponding second target point subset based on the first data, and generate third data corresponding to the first data in combination with the vehicle speed data; wherein the first target point subset is static target point subset data; the second target point subset is dynamic target point subset data; and the third data is predicted target distance data; A second data processing unit is used to match and process the target point data in the target point subset at least twice according to the third data and in combination with the historical frame data and the current frame data respectively; The first data generating unit is used for clustering target points moving at the same speed, and generating fourth data corresponding to the target points moving at the same speed, and generating corresponding fifth data based on the fourth data; wherein the fourth data is a point cloud set of the same moving object; and the fifth data is fuzzy speed processing data corresponding to the vehicle-mounted 4D imaging millimeter wave radar.

8. The fuzzy velocity processing system for vehicle-mounted 4D imaging millimeter wave radar according to claim 7, characterized in that: The first data processing unit further includes: A first determination module, configured to determine a scene state corresponding to the first data based on the first data; A first creation module is used to create a third target point subset corresponding to the vehicle-mounted 4D imaging millimeter-wave radar according to the scene state of the target point in the combined field of view of the millimeter-wave radar; wherein the third first target point subset is a point set to be resolved; And / or, the target point subset creation unit further includes: A first generating module is used to generate and obtain sixth data corresponding to the vehicle-mounted 4D imaging millimeter wave radar, and to create a corresponding first target point subset and a second target point subset according to the sixth data; wherein the sixth data is point set data of the previous frame in which velocity blur is successfully resolved; The first processing module is used to predict and update the position data of each target point subset in combination with the vehicle speed data; And / or, the second data processing unit further includes: A first matching module, used to generate and obtain a first matching condition, and based on the first matching condition, search for matching points in the target point set of the current frame, and classify each target point into a corresponding subset according to the matching condition; A second matching module is used to generate and obtain a second matching condition, and based on the second matching condition, take an element in a subset of dynamic target points of the historical frame as a base point, perform a one-time matching process on the first data; A third matching module is used to generate and obtain a third matching condition, and based on the third matching condition, perform secondary matching processing on the first data with elements in the subset of dynamic target points of the historical frame as base points; A fourth matching module, used to generate and obtain a fourth matching condition, and based on the fourth matching condition, perform a one-time matching process on the first data with the target point in the current frame subset as a base point; a fifth matching module, configured to generate and obtain a fifth matching condition, and based on the fifth matching condition, perform secondary matching processing on the first data with the target point in the current frame subset as a base point; And / or, the first data generating unit further includes: A second generating module is used to generate and obtain seventh data corresponding to the moving target and having a moving speed and direction approximately equal to each other; wherein the seventh data is point cloud data of the target point; A third generating module, used for clustering the seventh data and generating corresponding fourth data with label data; The second determination module is used to generate and obtain a sixth matching condition, and determine whether each target point in the fourth data is a point cloud cluster of the same target based on the sixth matching condition.

9. A fuzzy velocity processing platform applied to vehicle-mounted 4D imaging millimeter wave radar, characterized in that: It includes a processor, a memory, and a fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar; wherein the fuzzy speed processing platform control program applied to the vehicle-mounted 4D imaging millimeter-wave radar is executed in the processor, the fuzzy speed processing platform control program applied to the vehicle-mounted 4D imaging millimeter-wave radar is stored in the memory, and the fuzzy speed processing platform control program applied to the vehicle-mounted 4D imaging millimeter-wave radar implements the fuzzy speed processing method applied to the vehicle-mounted 4D imaging millimeter-wave radar as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar. The fuzzy speed processing platform control program applied to a vehicle-mounted 4D imaging millimeter-wave radar implements the fuzzy speed processing method applied to a vehicle-mounted 4D imaging millimeter-wave radar as described in any one of claims 1 to 6.

Images