Millimeter wave radar-based multi-target detection method and device and storage medium

Abstract

The embodiment of the present specification provides a kind of multi-target detection method, device and storage medium based on millimeter wave radar, the method comprises: the coordinate of the tracked target, speed and echo intensity are collected;According to the coordinate, the speed and the echo intensity, the state vector of the tracked target is obtained;According to the state vector, the nuclear distance of the tracked target is determined;According to the nuclear distance, the trajectory point of the tracked target is determined;The association type of the trajectory point of the tracked target is determined;According to the association type, the trajectory of the tracked target is determined.The technical scheme provided in the present application is used to solve the problems that the prior art has single adaptation scene and low recognition accuracy.

Description

TECHNICAL FIELD

[0001] The present document relates to the field of artificial intelligence and computer security, and particularly relates to a multi-target detection method and device based on a millimeter wave radar and a storage medium. BACKGROUND

[0002] The millimeter wave radar has the advantages of small overall volume and all-weather working due to the working wavelength in the millimeter wave band, and its detection performance is less affected by the environment, which makes up for the shortcomings of other sensors in the application of vehicle-mounted radars, and becomes an indispensable sensor in the ADAS system and the automatic driving system. Therefore, the research on the vehicle-mounted millimeter wave radar has great potential value and practical significance.

[0003] The vehicle-mounted millimeter wave radar is mainly used for tracking multiple targets.

[0004] However, in this scenario, the existing tracking multi-target algorithm has the problems of single scene adaptation and low recognition accuracy. SUMMARY

[0005] In view of the above analysis, the present application aims to provide a multi-target detection method and device based on a millimeter wave radar and a storage medium, which can adapt to multiple scenes and improve tracking accuracy.

[0006] In a first aspect, one or more embodiments of the present specification provide a multi-target detection method based on a millimeter wave radar, comprising:

[0007] Collecting the coordinates, speed and echo intensity of the tracked target;

[0008] According to the coordinates, the speed and the echo intensity, the state vector of the tracked target is obtained;

[0009] According to the state vector, the core distance of the tracked target is determined;

[0010] According to the core distance, the trajectory point of the tracked target is determined;

[0011] Determine the association type of the trajectory point of the tracked target;

[0012] According to the association type, the trajectory of the tracked target is determined.

[0013] Further, the tracked target corresponds to multiple state vectors;

[0014] According to the state vector, the core distance of the tracked target is determined, comprising:

[0015] Normalizing each element in each state vector;

[0016] determining distances between the state vectors after normalization;

[0017] determining a kernel function;

[0018] characterizing distances between the state vectors by using the kernel function, to obtain the kernel distances.

[0019] Further, the tracked target corresponds to a plurality of state vectors, each of the state vectors being a suspected trajectory point, and the kernel distances are used to characterize distances between the suspected trajectory points.

[0020] According to the kernel distances, determining a trajectory point of the tracked target, comprising:

[0021] According to a dimension of the state vectors, determining a minimum neighborhood sample number;

[0022] Based on a K-clustering algorithm, determining a distribution trend of the kernel distances of the suspected trajectory points;

[0023] According to the distribution trend of the kernel distances and the minimum neighborhood sample number, determining a neighborhood distance;

[0024] According to the neighborhood distance and the minimum neighborhood sample number, determining the trajectory point of the tracked target.

[0025] Further, according to a correlation type of determining the trajectory point of the tracked target, comprising:

[0026] determining whether there is only one trajectory point in a correlation gate;

[0027] When there is only one trajectory point in the correlation gate, the correlation type is one-to-one;

[0028] When there are a plurality of trajectory points in the correlation gate, the correlation type is many-to-one.

[0029] Further, according to the correlation type, determining a trajectory of the tracked target, comprising:

[0030] When the correlation type is one-to-one, updating the trajectory of the tracked target according to the trajectory point;

[0031] When the correlation type is many-to-one, calculating weights of the trajectory points;

[0032] According to the weights of the trajectory points and the trajectory points, determining the trajectory of the tracked target.

[0033] In a second aspect, one or more embodiments of the present specification provide a multi-target detection device based on a millimeter wave radar, comprising:

[0034] The acquisition module, the data processing module and the trajectory determination module;

[0035] The acquisition module is configured to acquire coordinates, speed and echo intensity of the tracked target.

[0036] The data processing module is configured to obtain a state vector of the tracked target according to the coordinates, the speed and the echo intensity, and determine a kernel distance of the tracked target according to the state vector.

[0037] The trajectory determination module is configured to determine a trajectory point of the tracked target according to the kernel distance, determine an association type of the trajectory point of the tracked target, and determine a trajectory of the tracked target according to the association type.

[0038] Further, the tracked target corresponds to a plurality of state vectors, the data processing module is configured to normalize each element in each of the state vectors, determine distances between the normalized state vectors, and determine a kernel function.

[0039] The kernel function is used to represent the distances between the state vectors to obtain the kernel distance.

[0040] Further, the tracked target corresponds to a plurality of state vectors, each of the state vectors is a suspected trajectory point, and the kernel distance is used to represent distances between the suspected trajectory points.

[0041] The trajectory determination module is configured to determine a minimum neighborhood sample number according to a dimension of the state vector, determine a distribution trend of the kernel distance of each suspected trajectory point based on a K clustering algorithm, determine a neighborhood distance according to the distribution trend of the kernel distance and the minimum neighborhood sample number, and determine the trajectory point of the tracked target according to the neighborhood distance and the minimum neighborhood sample number.

[0042] Further, the trajectory determination module is configured to determine whether there is only one trajectory point in an association gate wave, determine that the association type is one-to-one when there is only one trajectory point in the association gate wave, and determine that the association type is many-to-one when there are a plurality of trajectory points in the association gate wave.

[0043] In a third aspect, one or more embodiments of the present specification provide a storage medium, comprising:

[0044] Computer executable instructions for storing computer executable instructions, the computer executable instructions being executed to implement the method of the first aspect.

[0045] Compared with the prior art, the present application can at least achieve the following technical effects:

[0046] 1、The kernel function has strong locality and strong anti-interference ability to noise. Therefore, clustering the suspected trajectory points using the kernel distance can better exclude the interference of noise, thereby improving the recognition accuracy.

[0047] 2、In determining the trajectory of the tracked target, there can be multiple tracked targets, or there can be only one tracked target. The methods for determining the trajectory in the two cases are slightly different. Therefore, the present application distinguishes between the two cases based on the association type to improve the applicability of the algorithm. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the one or more embodiments of the present specification or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present specification, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0049] Figure 1 A flowchart of a multi-target detection method based on a millimeter wave radar according to one or more embodiments of the present specification. DETAILED DESCRIPTION

[0050] In order to enable those skilled in the art to better understand the technical solutions in the one or more embodiments of the present specification, the technical solutions in the one or more embodiments of the present specification will be described clearly and completely below with reference to the drawings in the one or more embodiments of the present specification. Obviously, the described embodiments are only some embodiments of the present specification, not all embodiments. Based on the one or more embodiments of the present specification, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present document.

[0051] The prior art has the following problems when implementing multi-target detection based on a millimeter wave radar:

[0052] 1、Based on the Euclidean distance for clustering suspected trajectory points, the Euclidean distance has the characteristic of single measurement information. For some scenarios, such as a car in motion, there are at least three dimensions of coordinates, speed and angle. This means that the prior art can not find parameters that can achieve the correct effect within a limited range of accuracy. In addition, the weight of the Euclidean distance usually needs to be manually adjusted according to experience and data characteristics, so using multiple features to measure based on the Euclidean distance can cause dimension disaster.

[0053] 2、Based on problem 1, the test personnel cannot well exclude non-trajectory points. Therefore, when determining the trajectory point, the test personnel usually considers that the tracked target has multiple trajectory points or only one trajectory point. In this way, the accuracy of the algorithm is improved. Ultimately, the algorithm is usually designed for one scene, while the other is directly ignored. For the scene of car driving, the tracked target often switches between having one trajectory point and having multiple trajectory points. Obviously, the existing algorithm can only assume that the tracked target has one trajectory point or multiple trajectory points, and thus is not applicable to the scene of car driving.

[0054] To solve the above technical problems, the embodiment of the present application provides a multi-target detection method based on a millimeter wave radar, comprising the following steps:

[0055] Step 1, coordinates, speed and echo intensity of a tracked target are collected.

[0056] In the embodiment of the present application, the coordinates, speed and echo intensity of the tracked target are collected by the millimeter wave radar.

[0057] Step 2, a state vector of the tracked target is obtained according to the coordinates, speed and echo intensity.

[0058] In the embodiment of the present application, the form of the coordinates is usually (x, y), and thus the state vector of the present application is (x, y, v, I), wherein (x, y) is the coordinates of the tracked target, v is the speed of the tracked target, and I is the echo intensity of the tracked target.

[0059] Step 3, a core distance of the tracked target is determined according to the state vector.

[0060] In the embodiment of the present application, step 3 is specifically:

[0061] Step 31, each element in each state vector is normalized.

[0062] In the embodiment of the present application, as described above, the units of the coordinates, speed and echo intensity are different, and thus normalization is needed to unify the units of the elements.

[0063] Step 32, the distance between each state vector after normalization is determined.

[0064] In the embodiment of the present application, the millimeter wave radar continuously tracks the target, and thus there are multiple state vectors of the tracked target, and there is a distance between these vectors.

[0065] Step 33, a core function is determined.

[0066] In the embodiments of the present application, the kernel function generally has types such as a linear kernel function, a polynomial kernel function, and a Gaussian kernel function. The Gaussian kernel function is used to calculate the kernel distance in the present application. The definition of the Gaussian function is as follows:

[0067]

[0068] where kgau(x i , y j ) is the kernel distance, x i and x j are two vectors, and σ is the function bandwidth. In different scenarios, different function bandwidths need to be selected.

[0069] Therefore, determining the kernel function includes determining the type of the kernel function and determining the function bandwidth.

[0070] Step 34: The kernel function is used to represent the distance between the state vectors, and the kernel distance is obtained.

[0071] In the embodiments of the present application, the vector distance formula is as follows:

[0072]

[0073] D N (x, y) is the distance between the vectors Φ(x) and Φ(y), and the definition of the Gaussian kernel function is brought into the above formula to obtain the kernel distance:

[0074]

[0075] Step 4: The trajectory point of the tracked target is determined according to the kernel distance.

[0076] In the embodiments of the present application, the tracked target corresponds to a plurality of state vectors, each state vector is a suspected trajectory point, and the kernel distance is used to represent the distance between the suspected trajectory points.

[0077] Step 4 is specifically as follows:

[0078] Step 41: The number of minimum neighborhood samples is determined according to the dimension of the state vector.

[0079] In the embodiments of the present application, the number of minimum neighborhood samples is the dimension of the state vector + 1.

[0080] Step 42: The distribution trend of the kernel distance is determined based on the K clustering algorithm.

[0081] In the embodiments of the present application, for each suspected trajectory point, the kernel distance of the K nearest points is found, which is equivalent to classifying each suspected trajectory point as a clustering center. Then, the K kernel distances are sorted in descending order, and the distribution trend of the kernel distance of each suspected trajectory point is obtained.

[0082] Step 43, determining the neighborhood distance according to the distribution trend of the core distance and the minimum neighborhood sample number.

[0083] In the embodiment of the present application, the K value is the minimum neighborhood sample number + 1. The distribution trend of the core distance of each suspected trajectory point is fitted to obtain a curve. The inflection point of the curve is the neighborhood distance.

[0084] Step 44, determining the trajectory point of the tracked target according to the neighborhood distance and the minimum neighborhood sample number.

[0085] In the embodiment of the present application, the neighborhood distance and the minimum neighborhood sample number are important parameters in the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm. The neighborhood distance and the minimum neighborhood sample number are obtained, and the trajectory point can be obtained based on the DBSCAN algorithm.

[0086] Therefore, compared with the prior art, since the core distance is introduced in the present application, the neighborhood distance and the minimum neighborhood sample number are no longer determined empirically, but can be calculated according to the actual scene, so that the tracking accuracy can be improved.

[0087] Step 5, determining the association type of the trajectory point of the tracked target.

[0088] In the embodiment of the present application, the association type is two kinds: one is one-to-one, and the other is many-to-one.

[0089] Specifically, it is determined whether there is only one trajectory point in the association gate; when there is only one trajectory point in the association gate, the association type is one-to-one; when there are multiple trajectory points in the association gate, the association type is many-to-one.

[0090] The association gate is a basic technology of data association, and the center of the gate is determined by the predicted value of the measurement. The trajectory points appearing in the association gate can be regarded as part of the trajectory of the tracked target. When there is only one trajectory point, it means that the trajectory of the tracked target corresponds to only one trajectory point. When there are multiple trajectory points, it means that the trajectory of the tracked target corresponds to multiple trajectory points. The trajectory determination methods corresponding to the above two cases are different.

[0091] Step 6, determining the trajectory of the tracked target according to the association type.

[0092] In the embodiment of the present application, when the association type is one-to-one, the trajectory of the tracked target is updated according to the trajectory point.

[0093] When the association type is many-to-one, the weight of each trajectory point is calculated. For example, the weight of each trajectory point is determined according to the signal-to-noise ratio of each trajectory point.

[0094] According to the weight of the trajectory point and the trajectory point, the trajectory of the tracked target is determined.

[0095] Therefore, by the above method, the application can flexibly select the trajectory determination method, thereby improving the applicability of the algorithm.

[0096] The application embodiment provides a multi-target detection device based on a millimeter wave radar, comprising: an acquisition module, a data processing module and a trajectory determination module.

[0097] The acquisition module is configured to acquire the coordinates, the speed and the echo intensity of the tracked target.

[0098] The data processing module is configured to obtain the state vector of the tracked target according to the coordinates, the speed and the echo intensity; and determine the core distance of the tracked target according to the state vector.

[0099] The trajectory determination module is configured to determine the trajectory point of the tracked target according to the core distance; determine the association type of the trajectory point of the tracked target; and determine the trajectory of the tracked target according to the association type.

[0100] In the application embodiment, the tracked target corresponds to multiple state vectors; the data processing module is configured to normalize each element in each state vector; determine the distance between each normalized state vector; determine the core function; and use the core function to represent the distance between the state vectors to obtain the core distance.

[0101] In the application embodiment, the tracked target corresponds to multiple state vectors, and each state vector is a suspected trajectory point; the core distance is used to represent the distance between each suspected trajectory point.

[0102] The trajectory determination module is configured to determine the minimum neighborhood sample number according to the dimension of the state vector; determine the distribution trend of the core distance of each suspected trajectory point based on a K clustering algorithm; determine the neighborhood distance according to the distribution trend of the core distance and the minimum neighborhood sample number; and determine the trajectory point of the tracked target according to the neighborhood distance and the minimum neighborhood sample number.

[0103] In the application embodiment, the trajectory determination module is configured to determine whether there is only one trajectory point in the association gate; when there is only one trajectory point in the association gate, the association type is one-to-one; and when there are multiple trajectory points in the association gate, the association type is many-to-one.

[0104] The embodiments of the present application provide a storage medium, comprising:

[0105] for storing computer-executable instructions which, when executed, implement the method described in the above embodiments.

[0106] The above describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different than the order in which they are recited in the embodiments and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous or necessary.

[0107] In the 1930s, it was clear to distinguish whether an improvement in a technology was in hardware (e.g., improvement in circuit structure of diodes, transistors, switches, etc.) or in software (e.g., improvement in method flow). However, as technology has evolved, many improvements in method flow today can be considered as direct improvements in hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented by a hardware entity module. For example, a programmable logic device (PLD) (e.g., a field programmable gate array (FPGA)) is an integrated circuit whose logic function is determined by user programming of the device. A digital system is "integrated" on a PLD by the designer programming it, rather than by asking a chip manufacturer to design and fabricate a custom integrated circuit chip. Moreover, instead of manually fabricating an integrated circuit chip, this programming is now mostly implemented by "logic compiler" software, which is similar to software compilers used in program development, and the original code to be compiled is written in a specific programming language, called a hardware description language (HDL), of which there are many, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc., the most commonly used being VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. It should be clear to those skilled in the art that, by simply logically programming a method flow in one of the above hardware description languages and programming it into an integrated circuit, a hardware circuit implementing the logical method flow can be easily obtained.

[0108] The controller can be implemented in any suitable way, for example, the controller can take the form of a microprocessor or processor and a computer readable medium storing computer readable program code, such as software or firmware, executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller and an embedded microcontroller, examples of which include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320, the memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also know that, in addition to implementing the controller purely in terms of computer readable program code, it is possible to implement the controller to perform the same functions using logic gates, switches, application specific integrated circuits, programmable logic controllers and embedded microcontrollers, by logically programming the method steps. The controller can therefore be considered a hardware component, and the means for performing the various functions comprised therein can be considered structures within the hardware component. Alternatively, or even additionally, the means for performing the various functions can be considered both software modules which implement the method and structures within the hardware component.

[0109] The systems, apparatuses, modules or units illustrated by the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0110] For the sake of description, the above apparatuses are described in various units by functions respectively. Of course, the functions of each unit can be implemented in the same or multiple software and / or hardware in the implementation of the embodiments of the present specification.

[0111] Those skilled in the art will appreciate that one or more embodiments of the specification can be provided as a method, system or computer program product. Therefore, one or more embodiments of the specification can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the specification can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0112] The specification is presented with reference to flow diagrams and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the specification. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing element or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Although the flow diagrams and / or block diagrams can present a method, apparatus or computer program product in a particular, it is understood that the method, apparatus and computer program product can include one or more additional steps, operations, or functions, and the method, apparatus and computer program product can include one or more other steps, operations, functions or combinations of steps, operations, or functions in Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Although the flow diagrams and / or block diagrams can present a method, apparatus or computer program product in a particular, it is understood that the method, apparatus and computer program product can include one or more additional steps, operations, or functions, and the method, apparatus and computer program product can include one or more other steps, operations, functions or combinations of steps, operations, or functions in The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Although the flow diagrams and / or block diagrams can present a method, apparatus or computer program product in a particular, it is understood that the method, apparatus and computer program product can include one or more additional steps, operations, or functions, and the method, apparatus and computer program product can include one or more other steps, operations, functions or combinations of steps, operations, or functions in

[0113] The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Although the flow diagrams and / or block diagrams can present a method, apparatus or computer program product in a particular, it is understood that the method, apparatus and computer program product can include one or more additional steps, operations, or functions, and the method, apparatus and computer program product can include one or more other steps, operations, functions or combinations of steps, operations, or functions in Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Although the flow diagrams and / or block diagrams can present a method, apparatus or computer program product in a particular, it is understood that the method, apparatus and computer program product can include one or more additional steps, operations, or functions, and the method, apparatus and computer program product can include one or more other steps, operations, functions or combinations of steps, operations, or functions in Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Although the flow diagrams and / or block diagrams can present a method, apparatus or computer program product in a particular, it is understood that the method, apparatus and computer program product can include one or more additional steps, operations, or functions, and the method, apparatus and computer program product can include one or more other steps, operations, functions or combinations of steps, operations, or functions in The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Although the flow diagrams and / or block diagrams can present a method, apparatus or computer program product in a particular, it is understood that the method, apparatus and computer program product can include one or more additional steps, operations, or functions, and the method, apparatus and computer program product can include one or more other steps, operations, functions or combinations of steps, operations, or functions in

[0114] The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Although the flow diagrams and / or block diagrams can present a method, apparatus or computer program product in a particular, it is understood that the method, apparatus and computer program product can include one or more additional steps, operations, or functions, and the method, apparatus and computer program product can include one or more other steps, operations, functions or combinations of steps, operations, or functions in Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Although the flow diagrams and / or block diagrams can present a method, apparatus or computer program product in a particular, it is understood that the method, apparatus and computer program product can include one or more additional steps, operations, or functions, and the method, apparatus and computer program product can include one or more other steps, operations, functions or combinations of steps, operations, or functions in Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Although the flow diagrams and / or block diagrams can present a method, apparatus or computer program product in a particular, it is understood that the method, apparatus and computer program product can include one or more additional steps, operations, or functions, and the method, apparatus and computer program product can include one or more other steps, operations, functions or combinations of steps, operations, or functions in The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Although the flow diagrams and / or block diagrams can present a method, apparatus or computer program product in a particular, it is understood that the method, apparatus and computer program product can include one or more additional steps, operations, or functions, and the method, apparatus and computer program product can include one or more other steps, operations, functions or combinations of steps, operations, or functions in

[0115] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0116] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory, non-volatile memory, etc. in the form of computer-readable media. The memory is an example of computer-readable media.

[0117] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0118] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover non-exclusive inclusion, such that processes, methods, articles or devices that include a series of elements not only include those elements, but also include other elements not explicitly listed, or inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or device that includes the element.

[0119] One or more embodiments of the specification can be described in the general context of computer-executable instructions being executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. One or more embodiments of the specification can also be practiced in a distributed computing environment, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0120] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment focuses on the difference from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment.

[0121] The above merely provides the example of the present document and is not intended to limit the present document. For those skilled in the art, the present document can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present document shall be included in the scope of claims of the present document.

Claims

1. A method for multi-target detection based on millimeter wave radar, characterized in that, The method comprises: collecting coordinates, velocities and echo intensities of a tracked target; obtaining state vectors of the tracked target according to the coordinates, the velocities and the echo intensities; determining kernel distances of the tracked target according to the state vectors; determining trajectory points of the tracked target according to the kernel distances; determining an association type of the trajectory points of the tracked target; determining a trajectory of the tracked target according to the association type; the tracked target corresponds to multiple state vectors; the determining of the kernel distances of the tracked target according to the state vectors comprises: normalizing each element in each of the state vectors; determining distances between the normalized state vectors; determining a kernel function; characterizing the distances between the state vectors by using the kernel function to obtain the kernel distances.

2. The method according to claim 1, wherein: the tracked target corresponds to multiple state vectors, each of the state vectors being a suspected trajectory point, and the kernel distances are used to represent distances between the suspected trajectory points; the determining of the trajectory points of the tracked target according to the kernel distances comprises: determining a minimum neighborhood sample number according to a dimension of the state vectors; determining a distribution trend of the kernel distances of the suspected trajectory points based on a K clustering algorithm; determining a neighborhood distance according to the distribution trend of the kernel distances and the minimum neighborhood sample number; determining the trajectory points of the tracked target according to the neighborhood distance and the minimum neighborhood sample number.

3. The method according to claim 1, wherein: the determining of the association type of the trajectory points of the tracked target comprises: determining whether there is only one trajectory point in an association gate; when there is only one trajectory point in the association gate, the association type is one-to-one; when there are multiple trajectory points in the association gate, the association type is many-to-one.

4. The method according to claim 3, wherein: the determining of the trajectory of the tracked target according to the association type comprises: when the association type is one-to-one, updating the trajectory of the tracked target according to the trajectory point; when the association type is many-to-one, calculating weights of the trajectory points; determining the trajectory of the tracked target according to the weights of the trajectory points and the trajectory points.

5. A multi-target detection apparatus based on millimeter wave radar, characterized by comprising: The method comprises: a collecting module, a data processing module and a trajectory determining module; the collecting module is used to collect coordinates, velocities and echo intensities of a tracked target; the data processing module is used to obtain state vectors of the tracked target according to the coordinates, the velocities and the echo intensities, and determine kernel distances of the tracked target according to the state vectors; the trajectory determining module is used to determine trajectory points of the tracked target according to the kernel distances; determine an association type of the trajectory points of the tracked target; determine a trajectory of the tracked target according to the association type; the tracked target corresponds to multiple state vectors; and the data processing module is used to normalize each element in each of the state vectors; determine distances between the normalized state vectors; determine a kernel function; and represent distances between the state vectors by using the kernel function to obtain the kernel distances.

6. The apparatus of claim 5, wherein, the tracked target corresponds to a plurality of state vectors, each of the state vectors being a suspected trajectory point, and the kernel distances represent distances between the suspected trajectory points; the trajectory determination module is configured to determine a minimum neighborhood sample number according to a dimension of the state vectors, determine a distribution trend of the kernel distances of the suspected trajectory points based on a K clustering algorithm, determine a neighborhood distance according to the distribution trend of the kernel distances and the minimum neighborhood sample number, and determine a trajectory point of the tracked target according to the neighborhood distance and the minimum neighborhood sample number.

7. The apparatus of claim 5, wherein, the trajectory determination module is configured to determine whether there is only one suspected trajectory point in an associated gate; when there is only one suspected trajectory point in the associated gate, the association type is one-to-one; and when there are multiple suspected trajectory points in the associated gate, the association type is many-to-one.

8. A storage medium, characterized by including: computer executable instructions for storing, the computer executable instructions being executed to implement the method of any one of claims 1-4.

Images