Intelligent optimization system and related products for improving radar measurement accuracy

Abstract

The present invention relates to an intelligent optimization system and related products for improving radar measurement accuracy. The intelligent optimization system comprises: traversing a radar data acquisition control attribute set to obtain a radar data acquisition control attribute constraint interval set, generating a first radar data acquisition control solution set; generating an error fitness set according to an environmental factor monitoring value set; traversing a radar data acquisition control attribute set to obtain a positively correlated radar data acquisition control attribute set and a negatively correlated radar data acquisition control attribute set, evaluating the entropy fitness of the first radar data acquisition control solution set, and generating an entropy fitness set; optimizing the first radar data acquisition control solution set based on the entropy fitness set and the error fitness set, generating a recommended first radar data acquisition control solution for radar data acquisition control. The system solves the technical problem that the stability of measurement accuracy is poor due to the failure to objectively analyze the influence of environmental factors on the control parameter setting of the radar in the prior art.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an intelligent optimization system and related products for improving radar measurement accuracy. Background Art

[0002] During the radar data acquisition process, radar control parameters such as laser power, laser frequency, motor speed, camera exposure mode, ranging range, echo number parameters, and radar installation orientation will have a significant impact on the final measurement results.

[0003] The control parameter settings of traditional radars are usually set by experts and depend on the experts' professional level. In fact, the setting of radar control parameters needs to take into account actual environmental factors. For example, factors such as rainfall, fog concentration, temperature, humidity, etc. will affect the control parameter settings of the radar. Traditional subjective decision-making methods are difficult to ensure that the radar control parameters that are most suitable for environmental factors are obtained. Summary of the invention

[0004] The present invention aims to solve the technical problem that in the prior art, due to the failure to objectively analyze the impact of environmental factors on the control parameter setting of the radar, the radar control parameter scene adaptability is poor, which in turn leads to poor measurement accuracy stability. An intelligent optimization system for improving radar measurement accuracy is provided to solve the problem.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, the present invention provides an intelligent optimization system for improving radar measurement accuracy, comprising:

[0007] A value interval configuration module is used to traverse the radar data acquisition control attribute set to perform value constraints and obtain a radar data acquisition control attribute constraint interval set;

[0008] An initial solution generating module, used for generating a first radar data acquisition control solution set by uniformly and randomly distributing the radar data acquisition control attribute constraint interval set;

[0009] An error fitness evaluation module, used to traverse the first radar data acquisition control solution set to perform error fitness evaluation according to the environmental factor monitoring value set, and generate an error fitness set;

[0010] A correlation analysis module, used for traversing the radar data acquisition control attribute set to perform correlation analysis and obtain a positive correlation radar data acquisition control attribute set and a negative correlation radar data acquisition control attribute set;

[0011] An entropy fitness evaluation module is used to evaluate the entropy fitness of the first radar data acquisition control solution set based on the positive correlation radar data acquisition control attribute set and the negative correlation radar data acquisition control attribute set to generate an entropy fitness set;

[0012] A dual-constraint optimization module, used for performing dual-constraint iterative optimization on the first radar data acquisition control solution set based on the entropy value fitness set and the error fitness set, to generate a recommended first radar data acquisition control solution;

[0013] The radar data acquisition module is used to perform radar data acquisition control according to the recommended first radar data acquisition control solution.

[0014] In a second aspect, the present invention provides an electronic device, comprising:

[0015] Memory for storing computer software programs;

[0016] A processor is used to read and execute the computer software program, thereby implementing the intelligent optimization system for improving radar measurement accuracy as described in any one of the first aspects.

[0017] In a third aspect, the present invention provides a non-transitory computer-readable storage medium, in which a computer software program is stored. When the computer software program is executed by a processor, the intelligent optimization system for improving radar measurement accuracy as described in any one of the first aspects is implemented.

[0018] The beneficial effects of the present invention are as follows: configuring a value constraint interval for a preset radar data acquisition control attribute set through a value interval configuration module; further using an initial solution generation module to perform random uniform distribution based on the radar data acquisition control attribute constraint interval set to obtain a first radar data acquisition control solution set; further, using a machine learning strategy to predict the error influence of an environmental factor monitoring value set on the first radar data acquisition control solution set to obtain an error fitness set; further, through dynamic entropy value weighted analysis, obtaining an entropy fitness set of the first radar data acquisition control solution set; finally, under the guidance of the entropy fitness set and the error fitness set, performing dual-constraint iterative optimization on the first radar data acquisition control solution set to generate a recommended first radar data acquisition control solution, and utilizing a combination of machine learning technology and an optimization algorithm to realize automatic configuration of radar data acquisition control solutions combined with scenarios, thereby achieving a technical effect of ensuring the stability of radar measurement accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of a flow chart of an intelligent optimization system for improving radar measurement accuracy provided by the present invention;

[0020] Figure 2A schematic diagram of the structure of an electronic device provided by the present invention;

[0021] Figure 3 A schematic diagram of the structure of a computer-readable storage medium provided by the present invention.

[0022] In the accompanying drawings, the components represented by the reference numerals are listed as follows:

[0023] Value interval configuration module 1000, initial solution generation module 2000, error fitness evaluation module 3000, correlation analysis module 4000, entropy fitness evaluation module 5000, dual constraint optimization module 6000, radar data acquisition module 7000, electronic device 500, memory 510, processor 520, computer program 511, computer readable storage medium 600, computer program 611. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0025] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0026] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0027] Embodiment 1:

[0028] like Figure 1As shown, based on the same inventive concept as the intelligent optimization method for improving radar measurement accuracy provided in Embodiment 1, an embodiment of the present invention also provides an intelligent optimization system for improving radar measurement accuracy, including:

[0029] The value interval configuration module 1000 is used to traverse the radar data acquisition control attribute set to perform value constraints and obtain the radar data acquisition control attribute constraint interval set;

[0030] In detail, the radar data acquisition control attribute set is a set of parameters used to configure and control the radar data acquisition process, including sampling frequency, signal gain, scanning angle, echo number parameter, transmit pulse width, pulse repetition frequency, laser power, laser frequency, etc. Value constraints are restrictions on these control attributes and define their possible value ranges. The radar data acquisition control attribute constraint interval set is a collection of these value constraints, indicating the value range of each control attribute.

[0031] In the specific operation process, first traverse the radar data acquisition control attribute set. The control attribute set contains multiple parameters for configuring the radar data acquisition process. Traverse each control attribute and constrain its value. The value constraint defines the possible value range of each control attribute. For example, the sampling frequency may be constrained between 1000Hz and 5000Hz, the signal gain may be constrained between 1 and 10, and the scanning angle may be constrained between 0 and 180 degrees. During the traversal and constraint process, the constraint interval of each attribute is recorded. Finally, the constraint intervals of all control attributes are collected to form a radar data acquisition control attribute constraint interval set.

[0032] An initial solution generating module 2000 is used to generate a first radar data acquisition control solution set by uniformly and randomly distributing the radar data acquisition control attribute constraint interval set;

[0033] In detail, the first radar data acquisition control solution set refers to a set of control parameters generated by uniform random distribution based on the radar data acquisition control attribute constraint interval set.

[0034] In the specific operation process, it is used to pre-configure the number of uniformly distributed solution set requirements, and then determine the value range of each control attribute according to the radar data acquisition control attribute constraint interval set. For example, the sampling frequency may range from 1000Hz to 5000Hz, the signal gain may range from 1 to 10, and the scanning angle may range from 0 to 180 degrees. Uniform random distribution is performed within these constraint intervals. Specifically, for each control attribute, the number of solution set requirements is randomly generated within its corresponding constraint interval. These randomly generated values ​​are uniformly distributed throughout the constraint interval, ensuring that each value has the same probability of being selected. Exemplarily, a frequency value is randomly generated within the sampling frequency interval, a gain value is randomly generated within the signal gain interval, and an angle value is randomly generated within the scanning angle interval. These randomly generated control attribute values ​​are combined to form the first radar data acquisition control solution set. Any control solution set contains a set of parameters for configuring the radar data acquisition process, which can be used for actual data acquisition tasks.

[0035] The system is able to generate a variety of configurations within the range of control parameter values. These uniformly distributed random control solution sets help optimize the radar data acquisition process and ensure the comprehensiveness and diversity of data acquisition. The randomly generated control solution sets can be used in different test scenarios and experimental conditions to improve the adaptability and reliability of radar data acquisition. This method is suitable for various application scenarios that require high-precision radar data acquisition, such as autonomous driving, environmental monitoring, and safety protection, and helps improve the overall performance and data quality of the system.

[0036] The error fitness evaluation module 3000 is used to traverse the first radar data acquisition control solution set to perform error fitness evaluation according to the environmental factor monitoring value set, and generate an error fitness set;

[0037] In detail, the environmental factor monitoring value set refers to various monitoring values ​​in the environment during the radar data collection process, such as temperature, humidity, wind speed, etc., which can be customized by the user. Error fitness evaluation refers to evaluating the performance of each control solution in the actual environment and calculating its error. The error fitness set is a collection of these error evaluation results, which is used to analyze and optimize the radar data collection process.

[0038] First, according to the environmental element monitoring value set, various monitoring data in the current environment are obtained through sensors. These monitoring data may include temperature, humidity, wind speed, etc., and their impact on the radar data collection process needs to be taken into account.

[0039] Next, traverse the first radar data acquisition control solution set. Each control solution is a set of parameters for configuring radar data acquisition. During the traversal process, each control solution is evaluated for error fitness. Specifically, by analyzing the historical errors of each control solution in the environmental element monitoring value set, the specific error vector is predicted. Error calculation may involve statistical analysis, regression analysis or other data processing methods to quantify the fitness of the control solution in the current environment. The error fitness evaluation results of each control solution are recorded to form an error fitness set. This set contains the error evaluation results of all control solutions in the current environment.

[0040] These error fitness sets provide data on the performance of each control solution in the current environment, which helps to analyze and optimize the radar data acquisition process. By analyzing the error fitness set, the optimal control solution can be selected and adjusted to improve the accuracy and reliability of data acquisition.

[0041] A correlation analysis module 4000 is used to traverse the radar data acquisition control attribute set to perform correlation analysis and obtain a positive correlation radar data acquisition control attribute set and a negative correlation radar data acquisition control attribute set;

[0042] In detail, the positively correlated radar data acquisition control attribute set refers to a set of attributes that have the same trend in the process of change, and the negatively correlated radar data acquisition control attribute set refers to a set of attributes that have opposite trends in the process of change.

[0043] First, traverse the radar data acquisition control attribute set and collect the data of each control attribute. These control attributes may include sampling frequency, signal gain, scanning angle, etc. During the traversal process, record the value of each attribute at different time points or under different conditions.

[0044] Next, the collected data is subjected to correlation analysis. The correlation analysis method may use the Pearson correlation coefficient, the Spearman rank correlation coefficient or other suitable data analysis methods. By calculating these correlation coefficients, the correlation between each pair of control attributes is determined. Exemplarily, if the correlation analysis is the Pearson correlation coefficient, the division is as follows: if the correlation coefficient is close to 1, it indicates that there is a strong positive correlation between the two attributes; if the correlation coefficient is close to -1, it indicates that there is a strong negative correlation between the two attributes; if the correlation coefficient is close to 0, it indicates that there is no obvious correlation between the two attributes.

[0045] According to the results of correlation analysis, attributes with strong positive correlation are classified into the positive correlation radar data acquisition control attribute set, and attributes with strong negative correlation are classified into the negative correlation radar data acquisition control attribute set. Specifically, the positive correlation attribute set contains those control attributes that have the same trend in the process of change, for example, when one attribute value increases, the other attribute value also increases. The negative correlation attribute set contains those control attributes that have opposite trends in the process of change, for example, when one attribute value increases, the other attribute value decreases. Correlation analysis provides information about the relationship between control attributes, which helps to identify and adjust those parameter combinations that have a significant impact on data acquisition results.

[0046] An entropy fitness evaluation module 5000 is used to evaluate the entropy fitness of the first radar data acquisition control solution set based on the positive correlation radar data acquisition control attribute set and the negative correlation radar data acquisition control attribute set to generate an entropy fitness set;

[0047] Furthermore, the entropy fitness evaluation module 5000 executes the following steps:

[0048] According to the positive correlation radar data acquisition control attribute set and the negative correlation radar data acquisition control attribute set, forward normalization processing is performed on the first radar data acquisition control solution set to generate a first radar data acquisition control solution eigenvalue set;

[0049] Construct entropy fitness evaluation function:

[0050] ;

[0051] in, represents the fitness of the entropy value of the i-th solution, M represents the number of control attributes of radar data collection, represents the number of solutions, Characterize the j-th attribute eigenvalue of the i-th solution;

[0052] According to the entropy fitness evaluation function, the entropy values ​​of the first radar data acquisition control solution set are traversed for analysis to generate the entropy fitness set.

[0053] In detail, entropy fitness evaluation is an evaluation method based on information entropy, which is used to evaluate the fitness of the control solution set under different conditions. The entropy fitness set is a collection of these evaluation results, which is used to analyze and optimize the radar data acquisition process. The specific process is as follows:

[0054] For the control parameters belonging to the positive correlation radar data acquisition control attribute set, it is preferably processed by the following formula: x ij = y ij - min i ∈[1,N] y ij max i ∈[1,N] y ij - min i ∈[1,N] y ij , for the control parameters belonging to the negative correlation radar data acquisition control attribute set, it is preferably processed by the following formula: x ij = min i ∈[1,N] y ij - y ij max i ∈[1,N] y ij - min i ∈[1,N] y ij , Characterize the initial value of the jth attribute of the ith solution; perform forward normalization on the first radar data acquisition control solution set to generate the first radar data acquisition control solution feature value set. Construct an entropy fitness evaluation function: , traverse the entropy values ​​of the first radar data acquisition control solution set for analysis to generate the entropy value fitness set.

[0055] The entropy fitness can weaken the parameter values ​​with less obvious deviations, that is, the parameter values ​​that carry less information, and amplify the differences between different solutions, making it easier to distinguish the advantages and disadvantages of the solutions.

[0056] A dual-constraint optimization module 6000 is used to perform dual-constraint iterative optimization on the first radar data acquisition control solution set based on the entropy fitness set and the error fitness set to generate a recommended first radar data acquisition control solution;

[0057] Furthermore, the dual-constraint optimization module 6000 executes the following steps:

[0058] Construct a dual-constrained fitting function:

[0059] ;

[0060] in, Characterizes the fitting fitness of the i-th solution, Characterizes the fitness of the entropy value of the i-th solution, The error vector prediction value representing the ith solution, and Characterize the weight adjustment parameters, and The sum is equal to 1;

[0061] According to the double-constraint fitting function, based on the entropy value fitness set and the error fitness set, the first radar data acquisition control solution set is double-constrained to generate a first solution set fitting fitness set;

[0062] extracting a first quantity optimal solution from the first radar data acquisition control solution set according to the first solution set fitting fitness set;

[0063] extracting a second quantity inferior solution from the first radar data acquisition control solution set according to the first solution set fitting fitness set;

[0064] Taking the first number of optimal solutions as expansion targets, adjusting the second number of inferior solutions to generate a second radar data acquisition control solution set;

[0065] Iterative optimization is performed based on the second radar data acquisition control solution set, and when a preset number of times is met, an optimal solution is output and set as the recommended first radar data acquisition control solution.

[0066] Furthermore, the dual-constraint optimization module 6000 executes the following steps:

[0067] Randomly selecting the first number of optimal solutions to obtain a first selected optimal solution;

[0068] Randomly selecting the second number of inferior solutions to obtain a first selected inferior solution;

[0069] Constructing a first high-dimensional coordinate according to the first selected optimal solution;

[0070] Constructing a second high-dimensional coordinate according to the first selected inferior solution;

[0071] Taking the first high-dimensional coordinate as the target point and the second high-dimensional coordinate as the starting point, the first selected inferior solution is adjusted several times based on a preset adjustment step set and added into the second radar data acquisition control solution set.

[0072] In detail, since in the iterative optimization process, the initial solution needs to be expanded in the hope of obtaining a solution closer to the global optimal solution. In this process, the means adopted by the embodiment of the present application is to use the optimal solution as the target to adjust the inferior solution. Whether this process can obtain a solution that is widely distributed globally, it focuses on the selection of superior solutions and inferior solutions, and requires that the parameters of the superior solutions and inferior solutions have large differences. If only based on the analysis of error fitness, the separated solutions may not actually have a large degree of deviation, or the amount of information of the solutions with differences is insufficient, which makes it difficult to expand the solutions to a larger range. The entropy fitness is considered. Since the entropy fitness is generated based on the parameters of the two solutions themselves, the deviation of the two solutions is mainly due to the evaluation of the control attributes with large differences. Therefore, the solution combined with the entropy fitness is easy to obtain superior solutions and inferior solutions with large information carrying capacity and large deviations between solutions, thereby improving the probability of expanding to obtain the global optimal solution.

[0073] Based on the above principles, the detailed process of dual-constraint iterative optimization is as follows:

[0074] Construct a dual-constrained fitting function: ; According to the dual-constraint fitting function, based on the entropy fitness set and the error fitness set, the first radar data acquisition control solution set is evaluated one by one, and a first solution set fitting fitness set is output; a first number of optimal solutions are extracted from the first radar data acquisition control solution set according to the first solution set fitting fitness set; a second number of inferior solutions are extracted from the first radar data acquisition control solution set according to the first solution set fitting fitness set. The first number and the second number are set by the user. Preferably, in order to ensure a certain randomness, the first number is at least 2 and the second number is at least 5. Taking the first number of optimal solutions as the expansion target, the second number of inferior solutions is adjusted to generate a second radar data acquisition control solution set, that is, taking the optimal solution as the target, the inferior solution is adjusted to obtain an expanded solution, and when the expanded number is met, the second radar data acquisition control solution set is output. Iterative optimization is performed based on the second radar data acquisition control solution set, and when the preset number is met, the optimal solution is output and set as the recommended first radar data acquisition control solution. The so-called iterative optimization is to perform the double-constrained iterative optimization again based on the first radar data acquisition control solution set and the second radar data acquisition control solution set. Each time it is executed, the number of iterations increases by one. When it is greater than or equal to the preset number of times, the optimal solution is output and set as the recommended first radar data acquisition control solution.

[0075] The first number of optimal solutions is used as the expansion target, the second number of inferior solutions is adjusted, and any expansion of the second radar data acquisition control solution set is generated as follows:

[0076] The first number of optimal solutions is randomly selected to obtain a first selected optimal solution; the second number of inferior solutions is randomly selected to obtain a first selected inferior solution; according to the various control parameters of the first selected optimal solution, a first high-dimensional coordinate is constructed, and the parameter corresponding to any control attribute in the first high-dimensional coordinate is a coordinate element, and the parameters corresponding to multiple control attributes constitute the first high-dimensional coordinate; in the same way, the second high-dimensional coordinate is constructed according to the first selected inferior solution. Taking the first high-dimensional coordinate as the target point and the second high-dimensional coordinate as the starting point, the first selected inferior solution is adjusted several times based on the preset adjustment step set, and added to the second radar data acquisition control solution set. The adjustment is the process of adjusting the radar data acquisition control parameters. The preset adjustment step set refers to the constraint step of each adjustment preset by the user and corresponding to the control parameter attribute. When the generated solution meets the number of constraints of the expanded solution, the second radar data acquisition control solution set is output.

[0077] The radar data acquisition module 7000 is used to perform radar data acquisition control according to the recommended first radar data acquisition control solution.

[0078] In detail, starting the radar system to perform radar data acquisition control according to the recommended first radar data acquisition control solution can ensure the accuracy and stability of the measurement results. If the same scenario occurs later, the radar control parameters can be directly called without further analysis.

[0079] Furthermore, the error fitness evaluation module 3000 performs the following steps:

[0080] The initial environmental element set is sorted by relevance to generate a sensitive environmental element set;

[0081] Taking the radar model as a constraint, the sensitive environmental element set, the radar data acquisition control attribute set and the radar measurement error vector attribute are assigned values ​​through networking to generate a sensitive environmental element assignment data set, a radar data acquisition control factor assignment data set and a radar measurement error vector assignment data set;

[0082] Training an error vector evaluation network according to the sensitive environmental element value assignment data set, the radar data acquisition control factor value assignment data set, and the radar measurement error vector value assignment data set;

[0083] According to the error vector evaluation network, based on the environmental factor monitoring value set, the first radar data acquisition control solution set is traversed for fitting to generate the error fitness set.

[0084] In detail, the sensitive environmental element set refers to the environmental attributes that have a greater impact on the radar measurement accuracy, and the environmental element attributes of the subsequent environmental element monitoring value set are the sensitive environmental element set. Taking the radar model as a constraint, the sensitive environmental element set, the radar data acquisition control attribute set and the radar measurement error vector attribute are assigned to the network to obtain a one-to-one sensitive environmental element assignment data set, a radar data acquisition control factor assignment data set, and a radar measurement error vector assignment data set; then, the radar measurement error vector assignment data set is used as the output supervision data, and the sensitive environmental element assignment data set and the radar data acquisition control factor assignment data set are used as input data to train models such as neural networks, decision trees, and support vector machines, so as to obtain an application model that can predict the radar measurement error vector, which is set as an error vector evaluation network. Furthermore, according to the error vector evaluation network, based on the environmental element monitoring value set, the first radar data acquisition control solution set is traversed, each solution is combined with the environmental element monitoring value set, and each solution is input into the error vector evaluation network for processing to generate the error fitness set.

[0085] Furthermore, the error fitness evaluation module 3000 performs the following steps:

[0086] Taking the radar model and the radar data acquisition control attribute set as quantitative, the radar measurement error vector as dependent variable, and the initial environmental element set as independent variable to perform data acquisition, and obtain an initial environmental element record data set and a radar measurement error modulus record data set;

[0087] Normalizing the radar measurement error modulus record data set to construct a first benchmark sequence;

[0088] Normalizing the initial environmental factor record data set to construct a plurality of first comparison sequences;

[0089] A grey relational analysis is performed on the first reference sequence and the plurality of first comparison sequences to generate a grey relational set, wherein the sensitive environmental element set is an environmental element whose grey relational degree is greater than or equal to a relational threshold.

[0090] In detail, the initial environmental element record data set refers to the environmental element historical record data collected with the radar model and the radar data acquisition control attribute set as the quantitative, the radar measurement error vector as the dependent variable, and the initial environmental element set as the independent variable; the radar measurement error vector record data set refers to the radar measurement error vector historical record data that corresponds to the initial environmental element record data set one by one. The radar measurement error modulus record data set is normalized to construct a first reference sequence. The initial environmental element record data set is normalized to construct a number of first comparison sequences, and any first comparison sequence corresponds to an initial environmental element. Then, the first reference sequence is used as the first column data of the matrix, and each sequence of the several first comparison sequences is set as the other column data of the matrix to generate a gray correlation analysis matrix, and then a conventional gray correlation analysis is performed to obtain the correlation value of each initial environmental element and the radar measurement error modulus. Furthermore, the environmental elements with a gray correlation greater than or equal to the correlation threshold are screened and set as a sensitive environmental element set. Through the gray correlation analysis, the environmental elements with a large correlation with the radar measurement error are determined, and the elements with a small correlation are deleted to avoid the interference of redundant data.

[0091] Furthermore, the error fitness evaluation module 3000 executes the following steps:

[0092] Dividing the sensitive environmental element value assignment data set, the radar data acquisition control factor value assignment data set, and the radar measurement error vector value assignment data set into k training data sets;

[0093] Training k pre-fitting networks based on the k training data sets as supervision, wherein the k pre-fitting networks have different network topological structures;

[0094] When the k pre-fitting networks converge, obtaining the k pre-fitting network output accuracy rates;

[0095] Performing output weighting on the k front-end fitting networks according to the output accuracy of the k front-end fitting networks to generate k weight distribution results;

[0096] Taking the output values ​​of the k front-end fitting networks and the k weight distribution results as input, and taking the radar measurement error vector assignment data set as the supervision truth value, training the back-end fitting network;

[0097] The output layers of the k front-end fitting networks and the input layer of the rear-end fitting network are merged to generate the error vector evaluation network.

[0098] In detail, the detailed construction process of the error vector evaluation network is as follows:

[0099] The sensitive environmental factor assignment data set, the radar data acquisition control factor assignment data set, and the radar measurement error vector assignment data set are equally divided into k training data sets; according to the k training data sets as supervision, k pre-fitting networks are trained, wherein the network topological structures of the k pre-fitting networks are different. For example, if the k value is 3, the k pre-fitting networks can be BP neural network, random forest, support vector machine, etc. After the k pre-fitting networks converge, the k pre-fitting networks are tested respectively to obtain the output accuracy of the k pre-fitting networks.

[0100] Furthermore, the output accuracies of the k front-end fitting networks are added, and then the ratios of the output accuracies of the k front-end fitting networks to the summed results are respectively calculated, and set as the k weight distribution results of the k front-end fitting networks. Furthermore, the output values ​​of the k front-end fitting networks and the k weight distribution results are used as input to train the rear-end fitting network. Preferably, the network topology of the rear-end fitting network is the network topology structure with the maximum output accuracy of the k front-end fitting networks. The output layers of the k front-end fitting networks and the input layers of the rear-end fitting network are merged to generate the error vector evaluation network. Through the integration concept, the integrated model is trained to improve the accuracy of the output results.

[0101] Furthermore, the correlation analysis module 4000 executes the following steps:

[0102] randomly extracting a first radar data collection control attribute from the radar data collection control attribute set;

[0103] Taking the sensitive environmental factor set and the unselected radar data collection control attributes as quantitative, taking the radar data measurement error modulus as dependent variable, and taking the first radar data collection control attribute as independent variable to perform data collection, and obtain a first radar data collection control attribute data set and a radar data measurement error modulus collection data set;

[0104] Serializing and adjusting the first radar data collection control attribute data set from small to large to generate a first radar data collection control attribute data sequence;

[0105] According to the first radar data acquisition control attribute data sequence, the radar data measurement error modulus acquisition data set is serialized and followed to generate a radar data measurement error modulus acquisition data sequence;

[0106] Performing a Pearson correlation coefficient analysis on the first radar data acquisition control attribute data sequence and the radar data measurement error modulus acquisition data sequence to generate a Pearson correlation coefficient evaluation value;

[0107] The first radar data acquisition control attribute is added into the positive correlation radar data acquisition control attribute set or the negative correlation radar data acquisition control attribute set according to the Pearson correlation coefficient evaluation value.

[0108] In detail, the correlation analysis optimization steps are as follows:

[0109] The sensitive environmental element set and the unselected radar data acquisition control attribute are used as quantitative, the radar data measurement error modulus is used as the dependent variable, and the first radar data acquisition control attribute is used as the independent variable to perform data acquisition, and obtain the first radar data acquisition control attribute data set and the radar data measurement error modulus acquisition data set; the first radar data acquisition control attribute data set is serialized and adjusted from small to large to generate the first radar data acquisition control attribute data sequence; the radar data measurement error modulus acquisition data set is serialized and followed according to the first radar data acquisition control attribute data sequence to generate the radar data measurement error modulus acquisition data sequence, that is, since the radar data measurement error modulus acquisition data set and the first radar data acquisition control attribute data set are one-to-one corresponding data, the radar data measurement error modulus acquisition data set is sequentially adjusted one-to-one according to the first radar data acquisition control attribute data sequence. The Pearson correlation coefficient analysis is performed on the first radar data acquisition control attribute data sequence and the radar data measurement error modulus acquisition data sequence to generate the Pearson correlation coefficient evaluation value. According to the Pearson correlation coefficient evaluation value, the first radar data acquisition control attribute is added to the positive correlation radar data acquisition control attribute set or the negative correlation radar data acquisition control attribute set. Exemplarily, if the Pearson correlation coefficient evaluation value is greater than 0, it is added to the positive correlation radar data acquisition control attribute set; if the Pearson correlation coefficient evaluation value is less than 0, it is added to the negative correlation radar data acquisition control attribute set; the control attribute equal to 0 does not participate in the entropy weight fitness evaluation.

[0110] The intelligent optimization system for improving radar measurement accuracy provided by the embodiment of the present invention has at least the following technical effects:

[0111] 1. Configure a value constraint interval for a preset radar data acquisition control attribute set through a value interval configuration module; further use an initial solution generation module to perform random uniform distribution based on the radar data acquisition control attribute constraint interval set to obtain a first radar data acquisition control solution set; further, use a machine learning strategy to predict the error impact of the environmental factor monitoring value set on the first radar data acquisition control solution set to obtain an error fitness set; further, through dynamic entropy value weighted analysis, obtain an entropy fitness set of the first radar data acquisition control solution set; finally, under the guidance of the entropy fitness set and the error fitness set, perform dual-constraint iterative optimization on the first radar data acquisition control solution set to generate a recommended first radar data acquisition control solution, and use a combination of machine learning technology and optimization algorithm to achieve automatic configuration of radar data acquisition control solutions combined with scenarios, thereby achieving a technical effect of ensuring the stability of radar measurement accuracy.

[0112] 2. Since the initial solution needs to be expanded in the iterative optimization process in the hope of obtaining a solution closer to the global optimal solution. In this process, the method adopted by the embodiment of the present application is to use the optimal solution as the target to adjust the inferior solution. Whether this process can obtain a solution with a wider global distribution, it focuses on the selection of superior solutions and inferior solutions, requiring that the parameters of the superior solutions and inferior solutions have large differences. If only based on the analysis of error fitness, the separated solutions may not actually have a large degree of deviation, or the amount of information of the solutions with differences is insufficient, which makes it difficult to expand the solutions to a larger range. The entropy fitness is considered. Since the entropy fitness is generated based on the parameters of the two solutions themselves, the deviation of the two solutions is mainly due to the evaluation of the control attributes with large differences. Therefore, the solution combined with the entropy fitness is easy to obtain superior solutions and inferior solutions with large information carrying capacity and large deviation between solutions, thereby improving the probability of expanding to obtain the global optimal solution.

[0113] Embodiment 2:

[0114] See also Figure 2 , Figure 2 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 2 As shown, an embodiment of the present invention provides an electronic device 500, including a memory 510, a processor 520, and a computer program 511 stored in the memory 510 and executable on the processor 520. When the processor 520 executes the computer program 511, the following steps are implemented: traversing the radar data acquisition control attribute set to obtain the radar data acquisition control attribute constraint interval set, and generating a first radar data acquisition control solution set; generating an error fitness set according to the environmental factor monitoring value set; traversing the radar data acquisition control attribute set to obtain a positively correlated radar data acquisition control attribute set and a negatively correlated radar data acquisition control attribute set, evaluating the entropy fitness of the first radar data acquisition control solution set, and generating an entropy fitness set; optimizing the first radar data acquisition control solution set based on the entropy fitness set and the error fitness set, and generating a recommended first radar data acquisition control solution for radar data acquisition control.

[0115] Embodiment three:

[0116] See also Figure 3 , Figure 3 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 3As shown, this embodiment provides a computer-readable storage medium 600, on which a computer program 611 is stored. When the computer program 611 is executed by a processor, the following steps are implemented: traversing the radar data acquisition control attribute set to obtain the radar data acquisition control attribute constraint interval set, and generating a first radar data acquisition control solution set; generating an error fitness set according to the environmental factor monitoring value set; traversing the radar data acquisition control attribute set to obtain a positively correlated radar data acquisition control attribute set and a negatively correlated radar data acquisition control attribute set, evaluating the entropy fitness of the first radar data acquisition control solution set, and generating an entropy fitness set; optimizing the first radar data acquisition control solution set based on the entropy fitness set and the error fitness set, and generating a recommended first radar data acquisition control solution for radar data acquisition control.

[0117] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0118] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0119] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0120] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0121] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0122] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0123] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. An intelligent optimization system for improving radar measurement accuracy, characterized in that: include: A value interval configuration module is used to traverse the radar data acquisition control attribute set to perform value constraints and obtain a radar data acquisition control attribute constraint interval set; An initial solution generating module, used for generating a first radar data acquisition control solution set by uniformly and randomly distributing the radar data acquisition control attribute constraint interval set; An error fitness evaluation module, used to traverse the first radar data acquisition control solution set to perform error fitness evaluation according to the environmental factor monitoring value set, and generate an error fitness set; A correlation analysis module, used for traversing the radar data acquisition control attribute set to perform correlation analysis and obtain a positive correlation radar data acquisition control attribute set and a negative correlation radar data acquisition control attribute set; An entropy fitness evaluation module is used to evaluate the entropy fitness of the first radar data acquisition control solution set based on the positive correlation radar data acquisition control attribute set and the negative correlation radar data acquisition control attribute set to generate an entropy fitness set; A dual-constraint optimization module, used for performing dual-constraint iterative optimization on the first radar data acquisition control solution set based on the entropy value fitness set and the error fitness set, to generate a recommended first radar data acquisition control solution; A radar data acquisition module, used for performing radar data acquisition control according to the recommended first radar data acquisition control solution; The error fitness evaluation module executes the following steps: The initial environmental element set is sorted by relevance to generate a sensitive environmental element set; Taking the radar model as a constraint, the sensitive environmental element set, the radar data acquisition control attribute set and the radar measurement error vector attribute are assigned values ​​through networking to generate a sensitive environmental element assignment data set, a radar data acquisition control factor assignment data set and a radar measurement error vector assignment data set; Training an error vector evaluation network according to the sensitive environmental element value assignment data set, the radar data acquisition control factor value assignment data set, and the radar measurement error vector value assignment data set; According to the error vector evaluation network, based on the environmental factor monitoring value set, the first radar data acquisition control solution set is traversed for fitting to generate the error fitness set.

2. The system according to claim 1, characterized in that The error fitness evaluation module execution steps include: Taking the radar model and the radar data acquisition control attribute set as quantitative, the radar measurement error vector as dependent variable, and the initial environmental element set as independent variable to perform data acquisition, and obtain an initial environmental element record data set and a radar measurement error modulus record data set; Normalizing the radar measurement error modulus record data set to construct a first benchmark sequence; Normalizing the initial environmental factor record data set to construct a plurality of first comparison sequences; A grey relational analysis is performed on the first reference sequence and the plurality of first comparison sequences to generate a grey relational set, wherein the sensitive environmental element set is an environmental element whose grey relational degree is greater than or equal to a relational threshold.

3. The system according to claim 1, characterized in that The error fitness evaluation module execution step also includes: Dividing the sensitive environmental element value assignment data set, the radar data acquisition control factor value assignment data set, and the radar measurement error vector value assignment data set into k training data sets; Training k pre-fitting networks based on the k training data sets as supervision, wherein the k pre-fitting networks have different network topological structures; When the k pre-fitting networks converge, obtaining the k pre-fitting network output accuracy rates; Performing output weighting on the k front-end fitting networks according to the output accuracy of the k front-end fitting networks to generate k weight distribution results; Taking the output values ​​of the k front-end fitting networks and the k weight distribution results as input, and taking the radar measurement error vector assignment data set as the supervision truth value, training the back-end fitting network; The output layers of the k front-end fitting networks and the input layer of the rear-end fitting network are merged to generate the error vector evaluation network.

4. The system according to claim 1, characterized in that The correlation analysis module performs the following steps: randomly extracting a first radar data collection control attribute from the radar data collection control attribute set; Taking the sensitive environmental factor set and the unselected radar data collection control attributes as quantitative, taking the radar data measurement error modulus as dependent variable, and taking the first radar data collection control attribute as independent variable to perform data collection, and obtain a first radar data collection control attribute data set and a radar data measurement error modulus collection data set; Serializing and adjusting the first radar data collection control attribute data set from small to large to generate a first radar data collection control attribute data sequence; According to the first radar data acquisition control attribute data sequence, the radar data measurement error modulus acquisition data set is serialized and followed to generate a radar data measurement error modulus acquisition data sequence; Performing a Pearson correlation coefficient analysis on the first radar data acquisition control attribute data sequence and the radar data measurement error modulus acquisition data sequence to generate a Pearson correlation coefficient evaluation value; The first radar data acquisition control attribute is added into the positive correlation radar data acquisition control attribute set or the negative correlation radar data acquisition control attribute set according to the Pearson correlation coefficient evaluation value.

5. The system according to claim 1, wherein: The entropy fitness evaluation module performs the following steps: According to the positive correlation radar data acquisition control attribute set and the negative correlation radar data acquisition control attribute set, forward normalization processing is performed on the first radar data acquisition control solution set to generate a first radar data acquisition control solution eigenvalue set; Construct entropy fitness evaluation function: ； in, represents the fitness of the entropy value of the i-th solution, M represents the number of control attributes of radar data collection, represents the number of solutions, Characterize the j-th attribute eigenvalue of the i-th solution; According to the entropy fitness evaluation function, the entropy values ​​of the first radar data acquisition control solution set are traversed for analysis to generate the entropy fitness set.

6. The system according to claim 1, wherein: The dual-constraint optimization module execution steps include: Construct a dual-constrained fitting function: ； in, Characterizes the fitting fitness of the i-th solution, Characterizes the fitness of the entropy value of the i-th solution, The error vector prediction value representing the ith solution, and Characterize the weight adjustment parameters, and The sum is equal to 1; According to the double-constraint fitting function, based on the entropy value fitness set and the error fitness set, the first radar data acquisition control solution set is double-constrained to generate a first solution set fitting fitness set; extracting a first quantity optimal solution from the first radar data acquisition control solution set according to the first solution set fitting fitness set; extracting a second quantity inferior solution from the first radar data acquisition control solution set according to the first solution set fitting fitness set; Taking the first number of optimal solutions as expansion targets, adjusting the second number of inferior solutions to generate a second radar data acquisition control solution set; Iterative optimization is performed based on the second radar data acquisition control solution set, and when a preset number of times is met, an optimal solution is output and set as the recommended first radar data acquisition control solution.

7. The system according to claim 6, characterized in that The dual-constraint optimization module execution steps include: Randomly selecting the first number of optimal solutions to obtain a first selected optimal solution; Randomly selecting the second number of inferior solutions to obtain a first selected inferior solution; Constructing a first high-dimensional coordinate according to the first selected optimal solution; Constructing a second high-dimensional coordinate according to the first selected inferior solution; Taking the first high-dimensional coordinate as the target point and the second high-dimensional coordinate as the starting point, the first selected inferior solution is adjusted several times based on a preset adjustment step set and added into the second radar data acquisition control solution set.

8. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing the intelligent optimization system for improving radar measurement accuracy as described in any one of claims 1 to 7.

9. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores a computer software program, and when the computer software program is executed by the processor, the intelligent optimization system for improving radar measurement accuracy as described in any one of claims 1 to 7 is implemented.

Images