Unmanned vehicle multi-target data association method and system

Abstract

The invention discloses an unmanned vehicle multi-target data association method and system. According to the method, observation data and prediction data are subjected to intuitionistic fuzzification processing through an improved intuitionistic fuzzy C-means clustering algorithm, the weighted distance between intuitionistic fuzzy sets is calculated to obtain the membership degree of observation and a track, and association of observation and the track is achieved by searching the maximum membership degree. According to the method, the uncertainty and fuzziness in the data can be effectively processed, the accuracy of target tracking is improved, and meanwhile, the high efficiency and the real-time performance of the algorithm are kept. The method is especially suitable for multi-target tracking in a complex traffic environment, such as multi-target, strong interference, dense clutter, cross track and the like, and the driving safety and reliability of the unmanned vehicle in the complex environment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of target detection, and in particular to a method and system for associating multi-target data of an unmanned vehicle. Background Art

[0002] Autonomous vehicles need to obtain information about the surrounding static environment in the driving map, and also need to pay attention to information about moving obstacles, especially pedestrians and other vehicles. Real-time multi-target tracking of pedestrians and other vehicles around the vehicle is an important technology that autonomous vehicles must have.

[0003] With the continuous advancement of signal processing technology, the association of track and point data has become a crucial part of the target tracking system. Especially in complex traffic environments, such as multiple targets, strong interference, dense clutter and cross tracks, target data association becomes extremely difficult.

[0004] From the perspective of data association methods, the most commonly used are Bayesian data association algorithms based on the Bayesian criterion, such as the nearest neighbor algorithm, probabilistic data association algorithm, joint probabilistic data association algorithm and multiple hypothesis algorithm. However, the association accuracy of these association algorithms is usually low. When the clutter density is high, the targets overlap with each other or the number of targets is large, it is easy to have problems of false association and missed association.

[0005] In the prior art, the joint probabilistic data association (JPDA) is usually used to update the state by calculating the association probability of the target and the measurement, taking into account the influence of all targets and measurements on the association probability, so it has a good association effect in dense target and dense clutter environment. The calculation of the association probability is based on the feasible event assumption. When the number of targets and measurements increases, the number of feasible events increases sharply, causing the calculation of the association probability to grow exponentially. In practical applications, the processor speed and content capacity are limited, and the actual target tracking is more complicated. The existing JPDA algorithm cannot meet the above calculation requirements.

[0006] In addition, due to the limitations of the sensor itself, there is a certain degree of uncertainty and ambiguity in the measurement, and the tracking effect of the classic data association method is often not ideal in a strong clutter environment. Summary of the invention

[0007] Based on this, the embodiments of the present application provide a method and system for associating multi-target data of an unmanned vehicle, which solves the problem of uncertainty in sensor observation data.

[0008] In a first aspect, a method for associating multi-target data of an unmanned vehicle is provided, the method comprising:

[0009] The measurement data and prediction data obtained by the unmanned vehicle at time k are subjected to intuitive fuzzy processing; wherein the intuitive fuzzy processing includes weighting the square of the Euclidean distance between each measurement data and the cluster center by constructing a membership matrix for the verified measurement data to determine the membership of the measurement data to each target category to obtain an intuitive fuzzy set;

[0010] After completing the intuitive fuzzy processing of the data, the weighted distance between the intuitive fuzzy sets is calculated to obtain the membership degree between the observation value and the track;

[0011] The association between the measurement data and the unmanned vehicle track is achieved by searching for the maximum membership degree.

[0012] Optionally, the weighted distance between the intuitionistic fuzzy sets is calculated to obtain the membership between the observations and the tracks, including:

[0013] Define an objective function that minimizes the objective function through an iterative process under the constraints of satisfying the multi-target tracking characteristics to determine the sum of the membership of each measurement value to all cluster centers, and update the cluster centers accordingly.

[0014] Optionally, the association between the measurement data and the unmanned vehicle track is achieved by searching for a maximum degree of membership, including:

[0015] The association between the measurement data and the track is achieved by searching for the maximum membership. If the measurement value is associated with only one target, its intuitive fuzzy membership is directly used as the association probability. If the measurement value is associated with multiple targets, the association probability is set to the minimum value of the association probability between the measurement value and the target according to the reconstructed association probability matrix.

[0016] Optionally, the measurement data and prediction data acquired by the driverless vehicle at time k are subjected to intuitive fuzzy processing, including:

[0017] At time k there exists m k Verified measurements, denoted as

[0018] In order to divide the measurement set into n t categories, n t Represents the number of targets; the optimal fuzzy clustering objective function is defined as:

[0019]

[0020] μ ij ∈[0,1];i=1,2,…,n t ; 1≤j≤m k

[0021]

[0022] Among them, μ ij is the membership degree of measurement value j to category i; is the measured value z j With cluster center v i The square of the Euclidean distance between them; c∈[1,∞] is the weighted index, and the sum of the membership of the measured value to all cluster centers is equal to 1. According to the characteristics of multi-target tracking

[0023]

[0024] The objective function is defined as

[0025]

[0026] And through the iterative process of the formula

[0027]

[0028] In the iterative process, the membership matrices U and V are selected to minimize J2(U, V).

[0029] Alternatively, according to the formula:

[0030]

[0031] The IF point operator is introduced in fuzzy clustering, where R i is the size of the tracking gate for target i, and Represents the proportional factor between membership information and non-membership information, Z i is the effective measurement set of target i at time k, τ is a positive number, which is determined by the tracking gate probability P G Fixed so that P G =P(χ 2 ≤τ); the chi-square distribution has N degrees of freedom Z , Indicates that measurement j belongs to the intuitive fuzzy membership of the i-th category, then the modified cluster center is;

[0032]

[0033] Update the cluster center and membership matrix at the same time, by

[0034]

[0035] Predict and measure the cluster center v i .

[0036] Optionally, the association between the measurement data and the unmanned vehicle track is achieved by searching for a maximum degree of membership, including:

[0037] If measurement j is correlated with only one target, then the correlation probability Set as intuitionistic fuzzy membership

[0038] If measurement j falls into multiple tracking gates at the same time, then;

[0039]

[0040] Where S represents the set of tracks associated with measurement j. If the association probability of measurement j is the largest among the measurements belonging to target i, the association probability of measurement j remains unchanged; otherwise, the association probability will be set to the minimum value of the probability that measurement j is associated with the target;

[0041] Reconstruct the association probability matrix A based on the fuzzy membership matrix U:

[0042]

[0043] in

[0044] In a second aspect, a multi-target data association system for an unmanned vehicle is provided, the system comprising:

[0045] The intuitive fuzzification module is used to perform intuitive fuzzification processing on the measurement data and prediction data obtained by the unmanned vehicle at time k; wherein the intuitive fuzzification processing includes weighting the square of the Euclidean distance between each measurement data and the cluster center by constructing a membership matrix for the verified measurement data to determine the membership of the measurement data to each target category to obtain an intuitive fuzzy set;

[0046] A calculation module is used to calculate the weighted distance between the intuitionistic fuzzy sets to obtain the membership degree between the observation value and the track after completing the intuitionistic fuzzification processing of the data;

[0047] The association module is used to associate the measurement data with the unmanned vehicle track by searching for the maximum membership degree.

[0048] In a third aspect, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the unmanned vehicle multi-target data association method described in any one of the first aspects is implemented.

[0049] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the unmanned vehicle multi-target data association method described in any one of the first aspects above is implemented.

[0050] In a fifth aspect, a computer program product is provided, comprising a computer program / instruction, which, when executed by a processor, implements the unmanned vehicle multi-target data association method described in any one of the first aspects above.

[0051] The beneficial effects of the technical solution provided by the embodiments of the present application include:

[0052] (1) The improved intuitionistic fuzzy C-means clustering (IFCM) algorithm is applied to multi-target data association of unmanned vehicles, which can effectively deal with the uncertainty and ambiguity in observation data and prediction data. By calculating the weighted distance between intuitionistic fuzzy sets to obtain the membership degree of observations and tracks, the observation data and tracks can be more accurately associated, thereby improving the accuracy of target tracking.

[0053] (2) The amount of computation only increases linearly with the clutter density and the number of targets, ensuring the high efficiency and real-time performance of the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0055] Figure 1 This is a flowchart of the steps of the multi-target data association method for unmanned vehicles in this application;

[0056] Figure 2 It is a schematic diagram of the relationship between the fuzzy intuition index and parameters;

[0057] Figure 3 A flow chart of an intuitive fuzzy joint probability data association filtering algorithm provided for an optional embodiment of the present application. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0059] In the description of the present invention, the terms "comprises", "has" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may also include other steps or units that are not explicitly listed but are inherent to these processes, methods, products or apparatuses, or steps or units added based on further optimization schemes conceived by the present invention.

[0060] Taking into account the fuzziness and uncertainty of point-track-track association judgment itself, the present invention develops an unmanned vehicle multi-target data association algorithm.

[0061] Aiming at the problem of uncertainty in sensor observation data, a new data association algorithm is proposed based on intuitionistic fuzzy clustering. The improved intuitionistic fuzzy C-means clustering (IFCM) algorithm is applied to data association. First, the observation data and prediction data are intuitively fuzzified, then the weighted distance between the intuitionistic fuzzy sets is calculated to obtain the membership of the observation and the track, and finally the maximum membership is searched in sequence to achieve the association of the observation and the track.

[0062] Please refer to Figure 1 , which shows a flowchart of a method for associating multi-target data of an unmanned vehicle provided in an embodiment of the present application. The method may include the following steps:

[0063] S1, perform intuitive fuzzy processing on the measurement data and prediction data obtained by the unmanned vehicle at time k.

[0064] Among them, the intuitive fuzzy processing includes constructing a membership matrix for the verified measurement data, weighting the square of the Euclidean distance between each measurement data and the cluster center to determine the membership of the measurement data to each target category to obtain an intuitive fuzzy set.

[0065] S2, after completing the intuitionistic fuzzy processing of the data, the weighted distance between the intuitionistic fuzzy sets is calculated to obtain the membership degree between the observation value and the track.

[0066] The objective function is defined, which is minimized through an iterative process under the constraint of satisfying the multi-target tracking characteristics to determine the sum of the membership of each measurement value to all cluster centers, and the cluster centers are updated accordingly.

[0067] The association between the measurement data and the track is achieved by searching for the maximum membership. If the measurement value is associated with only one target, its intuitive fuzzy membership is directly used as the association probability. If the measurement value is associated with multiple targets, the association probability is set to the minimum value of the association probability between the measurement value and the target according to the reconstructed association probability matrix.

[0068] Specifically, a specific implementation process of the above steps is given below:

[0069] 1. Intuitive fuzzification of observed data and predicted data

[0070] Assume that at time k there exists m k Verified measurements, denoted as

[0071] The goal of this application is to divide the measurement set into n t categories, n t represents the number of targets. For this purpose, the optimal fuzzy clustering objective function is defined as follows:

[0072]

[0073] μ ij ∈[0,1]; i=1,2,…,n t ; 1≤j≤m k

[0074]

[0075] Among them, μ ij is the degree of membership of measurement value j to category i; is the measured value z j With cluster center v i The square of the Euclidean distance between them. c∈[1,∞] is a weighted index, usually 2. Formula (2) shows that the sum of the membership of the measurement value to all cluster centers is equal to 1. According to the characteristics of multi-target tracking, Formula (2) can be modified as follows:

[0076]

[0077] The new objective function can be defined as

[0078]

[0079] Under the constraints of formula (3), U (membership matrix) and V are appropriately selected to minimize J2(U, V) through the iterative process of the following formula:

[0080]

[0081] Based on equations (5) and (6), in order to integrate the intuitive fuzzy characteristics into the traditional fuzzy C-means clustering algorithm, the IF (intuitive fuzzy) point operator is introduced in the fuzzy clustering. According to equation (5), a new intuitive fuzzy membership calculation method is proposed as follows:

[0082]

[0083] Among them, R i is the size of the tracking gate for target i, and Represents the proportional factor between membership information and non-membership information, Z i is the effective measurement set of target i at time k, τ is a positive number, which is determined by the tracking gate probability P G Fixed so that P G =P(χ 2 ≤τ). The chi-square distribution has N degrees of freedom. Z (the dimension of the measurement vector), represents the intuitionistic fuzzy membership of measurement j to the i-th category.

[0084] In formula (6), μ ij Substituting into formula (7), the modified cluster center is:

[0085]

[0086] Using equation (8), update the cluster center and the membership matrix at the same time. In fuzzy clustering, in order to avoid falling into local extreme values, the initialization of the cluster center is very important.

[0087] According to the characteristics of target tracking, the cluster center v can be provided by the following prediction measurement i (i=1 to n t ),Right now:

[0088]

[0089] 2. Definition of intuitive indicators

[0090] If the element z j For intuitionistic fuzzy set Z i The membership degree of is equal to 1, which means that the element z j Totally Z i , there is no uncertainty, so the intuitionistic fuzzy index π ij (z j ) is equal to zero. If the element z j For the intuitionistic fuzzy set Z i The non-membership degree v ij (z j ) is equal to 1, then the element z j Not Z i , so the intuitionistic fuzzy index π ij (z j ) is also equal to zero.

[0091] According to the characteristics of multi-target tracking, the intuitive index π ij (z j )satisfy:

[0092]

[0093] This occurs if and only if the measured value j is at the same position as the predicted position of the measurement, and the membership degree of the measured value j belonging to the target i is equal to 1.

[0094]

[0095] where g is the fixed threshold located at the gate probability P G on it.

[0096] This occurs if and only if the measured value j is outside the tracking gate, which means that the non - membership degree of the measured value j belonging to the target i should be equal to 1.

[0097] When 0 < d < g, π ij (z j ) is a continuous increasing - decreasing function.

[0098] When d changes from 0 to infinity, π ij (z j ) increases from 0, then decreases back to 0 at the gate boundary. This means that the intuition index reaches its maximum value at a certain point of the target tracking gate.

[0099] According to these properties of the intuition index π ij (z j ) in multi - target tracking, the fuzzy intuition index is defined as:

[0100]

[0101] where N(x) is the Yager intuitionistic fuzzy complement, x = d / g, x ∈ [0, 1]. ξ is a positive number. Figure 2 The relationship between the fuzzy intuition index π ij (z j ) and the parameter x (ξ = 0.85) is given.

[0102] 3. Association Probability Matrix A

[0103] Based on the intuitionistic fuzzy membership matrix U, a new reconstruction rule for the association probability matrix A is proposed. If the measurement j is only mutually associated with one target, the association probability is set as the intuitionistic fuzzy membership

[0104] If the measurement j falls within multiple tracking gates simultaneously, then:

[0105]

[0106] Where S represents the set of tracks associated with measurement j. In (11), the main idea of ​​the rule is that each target i can only produce one true measurement. Therefore, if measurement j has the largest association probability among the measurements belonging to target i, the association probability of measurement j remains unchanged. Otherwise, the association probability will be set to the minimum of the probabilities that measurement j is associated with the target.

[0107] Finally, the association probability matrix A is reconstructed according to the fuzzy membership matrix U:

[0108]

[0109] in,

[0110] The specific steps of the above method intuitive fuzzy joint probability data association filter algorithm are as follows Figure 3 As shown, specifically including:

[0111] Step 1. Assume that at time k-1, the previous estimate of the target state is and P i (k-1|k-1),i=1,2,…,n t Then the target prediction state can be calculated by formula (13).

[0112]

[0113] Step 2. Calculate the membership matrix U using (7).

[0114] Step 3. Calculate the cluster center V using (7) and (8).

[0115] Step 4. Use intuitionistic fuzzy membership Instead of association probability

[0116]

[0117] Step 5. Reconstruct the association probability matrix A using (12).

[0118] Step 6. Update the target state and covariance estimate using (14) and (15).

[0119]

[0120] Step 7. Repeat steps 1-6 next time.

[0121] In summary, it can be seen that the key point of this application is that according to the multi-target data association method provided by the present invention, the uncertainty association between observation points and track points is taken into account, so the features extracted from tracks and observations are also uncertain. In order to describe these features, intuitive fuzzy sets are used. The features of tracks and observation points are described by three aspects: membership, non-membership and intuitive index, so as to increase the authenticity and reliability of tracking results.

[0122] Moreover, the amount of calculation will only increase linearly with the clutter density and the number of targets, thus avoiding the "combinatorial explosion", thus ensuring high efficiency and meeting real-time requirements. This effectively solves the real-time problem of the algorithm.

[0123] This method is particularly suitable for multi-target tracking in complex traffic environments, such as multiple targets, strong interference, dense clutter and cross-tracks, improving the driving safety and reliability of unmanned vehicles in complex environments.

[0124] In one embodiment of the present application, a multi-target data association system for an unmanned vehicle is also provided, the system comprising:

[0125] The acquisition module obtains cluster points, unmanned vehicle multi-target data association numbers and geometric parameters from point cloud data;

[0126] A conversion module puts the geometric parameters into a buffer, filters the buffer data, and converts the filtered data into feature points;

[0127] The determination module determines the distance image of the object using the feature point and cluster point information.

[0128] For the specific definition of the unmanned vehicle multi-target data association system, please refer to the definition of the unmanned vehicle multi-target data association method above, which will not be repeated here. Each module in the above-mentioned unmanned vehicle multi-target data association system can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0129] In one embodiment, an electronic device is provided, which may be a computer. The electronic device includes a processor, a memory and a network interface connected via a system bus. The processor of the device is used to provide computing and control capabilities. The memory of the device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used for unmanned vehicle multi-target data association data. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for associating multi-target data of an unmanned vehicle is implemented.

[0130] In one embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned unmanned vehicle multi-target data association method are implemented.

[0131] In one embodiment of the present application, a computer program product is provided, including a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned unmanned vehicle multi-target data association method.

[0132] The computer-readable storage medium and computer program product provided in this embodiment have similar implementation principles and technical effects to those of the above method embodiments, and are not described in detail here.

[0133] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in M ​​forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SyMchliMk) DRAM (SLDRAM), memory bus (RaMbus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0134] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0135] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the patent application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent application shall be subject to the attached claims.

Claims

1. A multi-target data association method for an unmanned vehicle, characterized in that: The method comprises: The measurement data and prediction data obtained by the unmanned vehicle at time k are subjected to intuitive fuzzy processing; wherein the intuitive fuzzy processing includes weighting the square of the Euclidean distance between each measurement data and the cluster center by constructing a membership matrix for the verified measurement data to determine the membership of the measurement data to each target category to obtain an intuitive fuzzy set; After completing the intuitive fuzzy processing of the data, the weighted distance between the intuitive fuzzy sets is calculated to obtain the membership degree between the observation value and the track; The association between the measurement data and the unmanned vehicle track is achieved by searching for the maximum membership degree.

2. The unmanned vehicle multi-target data association method according to claim 1, characterized in that: Calculate the weighted distance between intuitionistic fuzzy sets to obtain the membership between observations and tracks, including: Define an objective function that minimizes the objective function through an iterative process under the constraints of satisfying the multi-target tracking characteristics to determine the sum of the membership of each measurement value to all cluster centers, and update the cluster centers accordingly.

3. The unmanned vehicle multi-target data association method according to claim 1, characterized in that: The association between the measurement data and the unmanned vehicle track is achieved by searching for the maximum membership, including: The association between the measurement data and the track is achieved by searching for the maximum membership. If the measurement value is associated with only one target, its intuitive fuzzy membership is directly used as the association probability. If the measurement value is associated with multiple targets, the association probability is set to the minimum value of the association probability between the measurement value and the target according to the reconstructed association probability matrix.

4. The unmanned vehicle multi-target data association method according to claim 1, characterized in that: The measurement data and prediction data obtained by the driverless vehicle at time k are intuitively fuzzy processed, including: At time k there exists m k Verified measurements, denoted as In order to divide the measurement set into n t categories, n t Represents the number of targets; the optimal fuzzy clustering objective function is defined as: m ij ∈[0,1];i=1,2,…,n t ;1≤j≤m k Among them, μ ij is the membership degree of measurement value j to category i; is the measured value z j With cluster center v i The square of the Euclidean distance between them; c∈[1,∞] is the weighted index, and the sum of the membership of the measured value to all cluster centers is equal to 1. According to the characteristics of multi-target tracking The objective function is defined as And through the iterative process of the formula In the iterative process, the membership matrices U and V are selected to minimize J2(U,V).

5. The unmanned vehicle multi-target data association method according to claim 4, characterized in that: According to the formula: The IF point operator is introduced in fuzzy clustering, where R i is the size of the tracking gate for target i, and Represents the proportional factor between membership information and non-membership information, Z i is the effective measurement set of target i at time k, τ is a positive number, which is determined by the tracking gate probability P G Fixed so that P G =P(χ 2 ≤τ); the chi-square distribution has N degrees of freedom Z , Indicates that measurement j belongs to the intuitive fuzzy membership of the i-th category, then the modified cluster center is; Update the cluster center and the membership matrix at the same time, by Predict and measure the cluster center v i .

6. The unmanned vehicle multi-target data association method according to claim 5, characterized in that: The association between the measurement data and the unmanned vehicle track is achieved by searching for the maximum membership, including: If measurement j is correlated with only one target, then the correlation probability Set as intuitionistic fuzzy membership If measurement j falls into multiple tracking gates at the same time, then; Where S represents the set of tracks associated with measurement j. If the association probability of measurement j is the largest among the measurements belonging to target i, the association probability of measurement j remains unchanged; otherwise, the association probability will be set to the minimum value of the probability that measurement j is associated with the target; Reconstruct the association probability matrix A based on the fuzzy membership matrix U: in i=1,2,…,n t ; 1≤j≤m k .

7. An unmanned vehicle multi-target data association system, characterized in that: The system comprises: The intuitive fuzzification module is used to perform intuitive fuzzification processing on the measurement data and prediction data obtained by the unmanned vehicle at time k; wherein the intuitive fuzzification processing includes weighting the square of the Euclidean distance between each measurement data and the cluster center by constructing a membership matrix for the verified measurement data to determine the membership of the measurement data to each target category to obtain an intuitive fuzzy set; A calculation module is used to calculate the weighted distance between the intuitionistic fuzzy sets to obtain the membership degree between the observation value and the track after completing the intuitionistic fuzzification processing of the data; The association module is used to associate the measurement data with the unmanned vehicle track by searching for the maximum membership degree.

8. An electronic device, characterized in that: It comprises a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the unmanned vehicle multi-target data association method as described in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the unmanned vehicle multi-target data association method as described in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the unmanned vehicle multi-target data association method as described in any one of claims 1 to 6 is implemented.

Images