Scene-adaptive multi-navigation source dynamic optimization and evaluation system and method

Abstract

The present invention discloses a scene-adaptive multi-navigation source dynamic optimization and evaluation system and method, which belongs to the field of navigation positioning and identification technology, and includes: a storage module for storing navigation source evaluation data and evaluation tasks; an exchange module for converting navigation source evaluation data and performing parameter initialization settings on evaluation tasks; a data processing module for executing evaluation tasks after parameter initialization settings, collating, storing and analyzing conversion data, and outputting designated data for optimization and evaluation; an online evaluation optimization module for processing navigation source evaluation data to obtain all system-available sensors corresponding to navigation source evaluation data; a quantitative evaluation module for selecting the best sensor from all system-available sensors; an evaluation report generation and demonstration module for displaying evaluation reports and data in a visual manner. The present invention can evaluate a single type of sensor and optimize sensors compatible with multiple scenarios.

Description

Technical Field

[0001] The present invention relates to the field of navigation positioning and identification technology, and in particular to a scene-adaptive multi-navigation source dynamic optimization and evaluation system and method. Background Art

[0002] With the development of society and the progress of science and technology, navigation source technology has made great progress. From a physical point of view, navigation source sensors cover acoustics, optics, electricity, magnetic fields, mechanics and other aspects. Since the attenuation of sound propagation underwater is weaker than other radios, the performance is better, so multi-beam depth sounding sonar is widely used underwater; in optics, cameras, lidar and other sensors use optical imaging for recording; radio can reach a propagation speed of 3×10 8 m / s, and has less attenuation in the air, so this type of equipment has developed faster and is more mature, such as GNSS receivers, Loran C receivers, etc.; the magnetic field uses the earth's magnetic force to perform geomagnetic matching; in terms of mechanics, it studies gravity matching, inertial navigation and other sensors.

[0003] There are many types of navigation source sensors, the most common ones are GNSS receivers, inertial navigation sensors, etc. GNSS receivers mainly include navigation receivers, geodetic receivers, and timing receivers; other sensors can be divided into many types according to their structure, purpose, performance and other characteristics. At the same time, the scenarios in which these sensors are used are also different. For aerospace scenes, some ground-based sensor devices are inconvenient to use; in indoor and underground scenes, the applicability of current GNSS receivers will be greatly reduced because navigation signals are easily blocked; for scenes above and below water, some land-based sensors are inconvenient to use. For underwater scenes, multi-beam depth sounding sonars, gravimeters, geomagnetic matching and other sensors are more commonly used.

[0004] It can be seen that any single type of sensor cannot be applied to all scenarios. Different sensors have their applicable scopes, but there is no evaluation method that is compatible with the above different categories of sensors. The existing evaluation system supports few types of navigation source sensors and has a limited scope of application; it does not support sensor optimization in different application scenarios and has poor flexibility; or it only supports fixed navigation source combinations to provide positioning and navigation services; there are relatively few quantitative evaluation management system solutions. Summary of the invention

[0005] In order to solve some or all of the technical problems existing in the above-mentioned prior art, the present invention provides a scene-adaptive multi-navigation source dynamic optimization and evaluation system and method, which can evaluate the usage scenarios of a single type of sensor, select sensors that are suitable and compatible with multiple scenarios, support sensor optimization in different application scenarios, and have good flexibility.

[0006] The technical solution of the present invention is as follows:

[0007] In a first aspect, the present invention provides a scene-adaptive multi-navigation source dynamic optimization and evaluation system, comprising:

[0008] A storage module, wherein the storage module is used to store navigation source assessment data and assessment tasks;

[0009] An exchange module, the exchange module is used to read the navigation source evaluation data and the evaluation task stored in the storage module, convert the navigation source evaluation data to obtain conversion data, perform parameter initialization settings on the evaluation task, and store and playback the converted navigation source evaluation data and the evaluation task with parameter initialization settings;

[0010] A data processing module, the data processing module is used to perform the evaluation task after the exchange module parameters are initialized and set, organize, store and analyze the conversion data output by the exchange module according to the evaluation task, and output the specified data for navigation source optimization and evaluation;

[0011] An online evaluation optimization module, the online evaluation optimization module is used to perform online evaluation on the navigation source evaluation data according to the conversion data output by the exchange module and the specified data output by the data processing module, and judge whether the online evaluation data has a fault in a binary hypothesis manner, output the evaluation data with a fault, generate an evaluation report, and process the evaluation data without a fault in the online evaluation to obtain all system-available sensors corresponding to the navigation source evaluation data;

[0012] A quantitative evaluation module, the quantitative evaluation module is used to receive data from all sensors available to the system, calculate whether the navigation accuracy and space occupation cost of all sensors available to the system obtained exceed preset values, and select the best sensor from all sensors available to the system that do not exceed the preset values;

[0013] An evaluation report generation and demonstration module, wherein the evaluation report generation and demonstration module is used to receive the evaluation report generated by the online evaluation optimization module, as well as the data of the navigation accuracy and space occupancy cost of all sensors available to the system calculated by the quantitative evaluation module and not exceeding the preset value, and to display the received data and the received evaluation report in a visual manner.

[0014] In some optional implementations, the online evaluation optimization module includes:

[0015] A fault detection and isolation module, wherein the fault detection and isolation module is used to perform fault detection on the navigation source evaluation data, isolate the navigation source evaluation data with faults, select the navigation source evaluation data without faults, and obtain the available sensor;

[0016] A scheme design module, the scheme design module is used to sort the available sensors, and obtain a sensor subset corresponding to the navigation source after generating, merging and pruning the sensors after sorting;

[0017] An evaluation standard module is used to calculate the navigation accuracy and space occupation cost of the sensors in the sensor subset obtained by the scheme design module to obtain all sensors available to the system.

[0018] In some optional implementations, determining whether there is a fault in the online evaluation data in a binary hypothesis manner includes providing a binary hypothesis fault verification function in the online evaluation optimization module, calling the binary hypothesis fault verification function when detecting a fault, and the binary hypothesis fault verification function is expressed as:

[0019] ;

[0020] in, Indicates that the navigation source obeys the degree of freedom of distributed, express The number of independent variables in represents chi-square, Represents the fault threshold, which is determined by the false detection rate of the navigation source.

[0021] In some optional implementations, the solution design module includes:

[0022] A generation module, the generation module is used to expand the available sensors obtained by the fault detection and isolation module to obtain a candidate sensor subset;

[0023] A merging module, the merging module is used to merge the repeated candidate sensor subsets generated by the generating module;

[0024] A pruning module is used to prune the candidate sensor subset merged by the merging module according to a preset sensor quantity upper limit threshold to obtain a sensor subset corresponding to the navigation source.

[0025] In some optional implementations, the evaluation criteria module includes:

[0026] A navigation accuracy determination module, the navigation accuracy determination module is used to calculate the navigation accuracy of the sensors in the sensor subset obtained by the scheme design module;

[0027] A cost calculation module, the cost calculation module is used to calculate the space occupation cost of the sensors in the sensor subset obtained by the scheme design module;

[0028] A comprehensive analysis module is used to make a comprehensive judgment based on the navigation accuracy calculation result of the navigation accuracy judgment module and the space occupancy cost calculation result of the cost calculation module, and select all system-available sensors that meet the use conditions and use requirements.

[0029] In a second aspect, the present invention further provides a scene-adaptive multi-navigation source dynamic optimization and evaluation method, comprising:

[0030] Acquire navigation source evaluation data and evaluation tasks, convert the navigation source evaluation data to obtain conversion data, perform parameter initialization settings on the evaluation tasks, and store the converted navigation source evaluation data and the evaluation tasks with the parameter initialization settings;

[0031] Execute the evaluation task after the parameter initialization setting, and organize, store and analyze the conversion data according to the evaluation task executed, and output the specified data for navigation source optimization and evaluation;

[0032] Performing online evaluation on the navigation source evaluation data according to the conversion data and the designated data, and judging whether the online evaluation data has a fault in a binary hypothesis manner, outputting the evaluation data with a fault, generating an evaluation report, and processing the evaluation data without a fault in the online evaluation to obtain all system-available sensors corresponding to the navigation source evaluation data;

[0033] Calculate whether the navigation accuracy and space occupation cost of all the sensors available to the system exceed the preset values, and select the best sensor from all the sensors available to the system that do not exceed the preset values;

[0034] The evaluation report and the obtained data that the navigation accuracy and space occupation cost of all sensors available to the system do not exceed preset values ​​are displayed in a visual manner.

[0035] In some optional embodiments, calculating whether the accuracy and space occupancy cost of all sensors available to the system exceed preset values ​​includes real-time calculation evaluation and post-calculation evaluation. The real-time calculation evaluation includes analyzing the evaluation data, calculating the evaluation results in real time and displaying them, and writing them into a database for storage. When the evaluation results exceed the standards, an alarm is issued for the exceeded results and the evaluation is terminated. The post-calculation evaluation includes reading the evaluation data from the database and performing evaluation analysis.

[0036] In some optional implementations, determining whether online evaluation data has a fault in a binary hypothesis manner includes:

[0037] Assume that the navigation data corresponding to the navigation source device is set to two states: faulty and non-faulty. Indicates no fault. Indicates a fault, judge according to the measured value or If true, there are four cases:

[0038] False positives: True, judge is true, the probability is ;

[0039] Missed detection: True, judge is true, the probability is ;

[0040] No false detection: True, judge is true, the probability is ;

[0041] Correct detection: True, judge is true, the probability is ;

[0042] Measurement value The entire observation space is , judged by the measured value, the measurement space is divided into two non-intersecting subspaces and ;

[0043] According to the Bayesian formula, we get:

[0044] ;

[0045] According to the chi-square test, the residual of fault detection is:

[0046] ;

[0047] When there is no fault in the system, the residual mean is 0, and the calculated variance is:

[0048] ;

[0049] Therefore, the above residuals and variances are used as verification conditions to determine whether the navigation source equipment has failed. The binary hypothesis is as follows:

[0050] No fault, then ;

[0051] If there is a fault, ;

[0052] The fault detection function is:

[0053] ;

[0054] The fault judgment criteria are:

[0055] ;

[0056] in, and Respectively represent the given measurement value Under the conditions and The probability of being true, and Respectively indicate the measured value exist and The conditional probability density when is true, express The probability of being true, express The probability of being true, represents the residual, is the linear regression parameter, represents the one-step prediction value of the state, represents the variance, is the total probability density function, represents the noise covariance matrix, represents the mathematical expectation, represents the mean of the residuals, represents the fault threshold, which is determined by the false detection rate. The degrees of freedom are of distributed, express The number of independent variables in represents chi-square, superscript represents a transpose operation, and the superscript -1 represents an inverse operation.

[0057] In some optional implementations, processing the online evaluation data of the fault-free state to obtain all system-available sensors corresponding to the navigation source evaluation data includes:

[0058] Based on the evaluation data of navigation sources without faults, available sensors are determined and sorted. After sorting, a sensor subset corresponding to the navigation source is obtained through generation, merging and pruning. The navigation accuracy and space occupancy cost of the sensors in the sensor subset are calculated. A comprehensive judgment is made on the sensors after the navigation accuracy is calculated and the space occupancy cost is calculated, and all system-available sensors that meet the usage conditions and requirements are selected.

[0059] In some optional implementations, the navigation accuracy of all sensors available to the system is determined using the following formula:

[0060] ;

[0061] in, represents a comprehensive position evaluation index, which is used as a measure of navigation accuracy. represents the average error value of longitude error, The average error value of latitude error, Indicates the average error value of the height error.

[0062] The main advantages of the technical solution of the present invention are as follows:

[0063] The scenario-adaptive multi-navigation source dynamic optimization and evaluation system and method of the present invention selects all system-available sensors by evaluating the navigation sources, and selects the optimal sensor from all system-available sensors by calculating the navigation accuracy and space occupancy cost of all selected system-available sensors. The system and method can evaluate the usage scenario of a single type of sensor, optimize sensors that are applicable and compatible with multiple scenarios, support sensor optimization in different application scenarios, and have good flexibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The drawings described herein are used to provide a further understanding of the embodiments of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0065] Figure 1 It is a structural schematic diagram of a scene-adaptive multi-navigation source dynamic optimization and evaluation system according to an embodiment of the present invention;

[0066] Figure 2A schematic diagram of a flow chart of a scene-adaptive multi-navigation source dynamic optimization and evaluation method provided by an embodiment of the present invention.

[0067] Description of reference numerals:

[0068] 1. Storage module; 2. Exchange module; 3. Data processing module; 4. Online evaluation and optimization module; 5. Quantitative evaluation module; 6. Evaluation report generation and demonstration module. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. 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 ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0070] The technical solution provided by the embodiments of the present invention is described in detail below with reference to the accompanying drawings.

[0071] As attached Figure 1 As shown, an embodiment of the present invention provides a scene-adaptive multi-navigation source dynamic optimization and evaluation system, which includes a storage module 1, an exchange module 2, a data processing module 3, an online evaluation optimization module 4, a quantitative evaluation module 5 and an evaluation report generation and presentation module 6, wherein:

[0072] The storage module 1 is used to store the navigation source evaluation data and the evaluation task; the exchange module 2 is used to read the navigation source evaluation data and the evaluation task stored in the storage module 1, convert the navigation source evaluation data to obtain the conversion data, initialize the parameters of the evaluation task, and can store the converted navigation source evaluation data and the evaluation task with the parameter initialization setting, and replay and read the stored data; the data processing module 3 is used to execute the evaluation task after the parameter initialization setting of the exchange module 2, organize, store and analyze the conversion data output by the exchange module 2 according to the evaluation task, and output the most suitable PNT (Positioning, Navigation, Timing) technology and combination method for navigation source optimization and evaluation; the online evaluation optimization module 4 is used to select the navigation source based on the conversion data output by the exchange module 2 and the specified data output by the data processing module 3. The source evaluation data is evaluated online, and it is determined whether the online evaluation data has faults in a binary hypothesis manner. The evaluation data with faults is output, an evaluation report is generated, and the evaluation data without faults in the online evaluation is processed to obtain all system-available sensors corresponding to the navigation source evaluation data; the quantitative evaluation module 5 is used to receive data from all system-available sensors, and calculate whether the navigation accuracy and space occupancy cost of all system-available sensors obtained exceed preset values, and select the optimal sensor from all system-available sensors that do not exceed the preset values; the evaluation report generation and demonstration module 6 is used to receive the evaluation report generated by the online evaluation optimization module 4, and receive data that the navigation accuracy and space occupancy cost of all system-available sensors calculated by the quantitative evaluation module 5 do not exceed the preset values, and display the received data and the received evaluation report in a visual manner.

[0073] It should be noted that the specified data output by the data processing module 3 above represents the most suitable PNT technology and combination method for navigation source optimization and evaluation.

[0074] Therefore, the scenario-adaptive multi-navigation source dynamic optimization and evaluation system of the present invention selects all system-available sensors by evaluating the navigation sources, and selects the optimal sensor from all system-available sensors by calculating the navigation accuracy and space occupancy cost of all selected system-available sensors. It can evaluate the usage scenarios of a single type of sensor, optimize sensors that are applicable and compatible with multiple scenarios, support sensor optimization in different application scenarios, and has good flexibility.

[0075] In some optional implementations of this embodiment, a SQL Server database may be selected as the storage module 1, and a server corresponding to the database may be selected as the server of the storage module 1 for permanently storing the evaluation data and evaluation tasks.

[0076] Specifically, the exchange module 2 of the present invention has, in addition to the above functions, operation control and deduction functions, data recording and playback functions. The operation control and deduction functions are mainly used to perform various types of control on the operation of the entire system, as well as to initialize the relevant parameters of the test task. The data recording and playback function, as the data distribution center of the system, realizes the storage, reading, updating, and playback of the data generated by the system, and provides data exchange or conversion functions for the system or device where the navigation source is located, thereby ensuring the stability of the system or device.

[0077] In the scenario-adaptive multi-navigation source dynamic optimization and evaluation system of the present invention, the online evaluation optimization module 4 includes: a fault detection and isolation module, a solution design module and an evaluation standard module, wherein:

[0078] The fault detection and isolation module is used to perform fault detection on the navigation source evaluation data, isolate the navigation source evaluation data with faults, select the navigation source evaluation data without faults, and obtain the available sensors; the scheme design module is used to sort the available sensors, and after sorting, generate, merge and prune to obtain the sensor subset corresponding to the navigation source; the evaluation standard module is used to calculate the navigation accuracy and space occupancy cost of the sensors in the sensor subset obtained by the scheme design module, and obtain all the sensors available to the system.

[0079] In the scenario-adaptive multi-navigation source dynamic optimization and evaluation system of the present invention, judging whether the online evaluation data has a fault in a binary hypothesis manner includes providing a binary hypothesis fault verification function in the online evaluation optimization module 4, calling the binary hypothesis fault verification function when detecting a fault, and the binary hypothesis fault verification function expression is:

[0080] ;

[0081] in, Indicates that the navigation source obeys the degree of freedom of distributed, express The number of independent variables in represents chi-square, that is , Represents the fault threshold, which is determined by the false detection rate of the navigation source.

[0082] In the scene-adaptive multi-navigation source dynamic optimization and evaluation system of the present invention, the scheme design module includes: a generation module, a merging module and a pruning module, wherein:

[0083] The generation module is used to expand the available sensors obtained by the fault detection and isolation module to obtain a candidate sensor subset; the merging module is used to merge the repeated candidate sensor subsets generated by the generation module; the pruning module is used to prune the candidate sensor subsets merged by the merging module according to a preset upper limit threshold of the number of sensors to obtain a sensor subset corresponding to the navigation source.

[0084] In the scene-adaptive multi-navigation source dynamic optimization and evaluation system of the present invention, the evaluation standard module includes: a navigation accuracy judgment module, a cost calculation module and a comprehensive analysis module, wherein:

[0085] The navigation accuracy judgment module is used to calculate the navigation accuracy of the sensors in the sensor subset obtained by the scheme design module; the cost calculation module is used to calculate the space occupancy cost of the sensors in the sensor subset obtained by the scheme design module; the comprehensive analysis module is used to make a comprehensive judgment based on the navigation accuracy calculation result of the navigation accuracy judgment module and the space occupancy cost calculation result of the cost calculation module, and select all system-available sensors that meet the usage conditions and usage requirements.

[0086] Specifically, in the navigation accuracy judgment module, it includes detecting and judging the precision of the sensor measurement range, such as variance and repeatability, etc.; detecting and judging the accuracy of the sensor relative to the true value, such as mean and centrality, etc. When detecting and judging the above-mentioned accuracy, a relative method, an absolute method or a statistical method can be used. Among them, in the present invention, the relative method represents the percentage of the maximum error to the true value; the absolute method represents the absolute value of the maximum error; the statistical method represents the method of using probability to characterize the error distribution to judge and calculate the accuracy of the sensor. In the cost calculation module, since the use environment of the selected sensor is a specific unconventional environment such as air and space, indoors and outdoors, or on and under water, there are strict requirements on the space occupied by the sensor during use and the weight of the sensor itself. Therefore, it is necessary to calculate the weight of the sensor and the space occupied, and analyze the cost and weight of the space resources occupied by the selected sensor, so that the space resources and weight costs of the selected sensor are selected when they are within the preset range.

[0087] Exemplarily, assuming that each sensor occupies several resource units and each resource unit has the same fixed number of computational costs, the resource units occupied by the candidate sensor subset should be less than the maximum number of resource units that the system can provide.

[0088] Specifically, in some optional implementations of this embodiment, the comprehensive analysis module is used to perform a comprehensive analysis on the navigation accuracy and cost of the above-selected sensors, and select all system-available sensors whose navigation accuracy and cost meet preset requirements.

[0089] In order to make the evaluation of the preferred sensor more accurate, the evaluation report generation and demonstration module 6 evaluates the navigation test data of the navigation source, such as inertial navigation, hydroacoustic navigation, radio navigation, visual navigation, speed / distance data, geomagnetic navigation and celestial navigation, according to the different environments in which the navigation source is used, and generates corresponding navigation test data reports, such as inertial navigation test data report, hydroacoustic navigation test data report, radio navigation test data report, visual navigation test data report, speed / distance sensor test data report, geomagnetic navigation test data report, and celestial navigation test data report.

[0090] Specifically, as an example, in indoor and underground scenarios, the above-mentioned navigation source detection can be applied to drones, unmanned vehicles, unmanned ships and unmanned submarines to generate drone evaluation reports, unmanned vehicle evaluation reports, unmanned ship evaluation reports and unmanned submarine evaluation reports.

[0091] In some optional implementations of this embodiment, the visual display methods include but are not limited to numerical data, curves, text, statistical graphics and radar charts.

[0092] At the same time, in order to be suitable for occasions where it is necessary to pay attention to the size of parameters, the visualization results can also be presented in the form of a list to directly display the data; in order to be suitable for occasions where it is necessary to pay attention to signal parameters that change with a fixed rule over time or several parameters with correlation, it is beneficial for commanders or analysts to be able to intuitively see the fluctuation range and basic change trend of the parameter values, and the visualization results can also be presented in the form of a curve; in order to be suitable for the relationship between subsystem parameters and total system parameters, the visualization results can be presented in the form of a radar chart; in order to be suitable for comparing the advantages and disadvantages of different parameters, the visualization results can also be presented in the form of statistical graphics, etc.

[0093] Second, as Figure 2 As shown, the present invention also provides a scene-adaptive multi-navigation source dynamic optimization and evaluation method, comprising:

[0094] Obtain navigation source evaluation data and evaluation tasks, convert the navigation source evaluation data to obtain conversion data, initialize parameters for the evaluation task, and store the converted navigation source evaluation data and the evaluation task with the parameter initialization setting; execute the evaluation task after the parameter initialization setting, and organize, store and analyze the conversion data according to the executed evaluation task, and output the specified data for navigation source optimization and evaluation; conduct online evaluation of the navigation source evaluation data according to the conversion data and the specified data, and determine whether the online evaluation data has faults in a binary hypothesis manner, output the evaluation data with faults, generate an evaluation report, and process the evaluation data without faults in the online evaluation to obtain all system-available sensors corresponding to the navigation source evaluation data; calculate whether the navigation accuracy and space occupancy cost of all system-available sensors obtained exceed preset values, and select the optimal sensor from all system-available sensors that do not exceed the preset values; and display the evaluation report and the data of the navigation accuracy and space occupancy cost of all system-available sensors obtained that do not exceed the preset values ​​in a visual manner.

[0095] It should be noted that the above-mentioned specified data represent the most suitable PNT technology and combination for navigation source optimization and evaluation.

[0096] Specifically, in the present invention, whether the navigation accuracy and space occupancy cost of all system-available sensors obtained by calculating exceed preset values ​​includes real-time calculation evaluation and post-calculation evaluation. The real-time calculation evaluation includes analyzing the evaluation data, calculating the evaluation results in real time and displaying them, and writing them into a database for storage. When the evaluation results exceed the standards, an alarm is issued for the exceeded results and the evaluation is terminated. The post-calculation evaluation includes reading the evaluation data from the database and performing evaluation analysis.

[0097] In the embodiment of the present invention, a binary hypothesis fault checking program is set to realize the judgment of whether the online evaluation data has a fault in a binary hypothesis manner. The specific principle is as follows:

[0098] Specifically, for the online evaluation of different navigation source evaluation data, when detecting faults, the following fault detection procedure is executed, and the evaluated navigation source can be divided into two hypotheses: faulty and non-faulty. According to the current measurement value, it is determined which hypothesis it belongs to. Assume that there are two hypotheses for the fault state of the autonomous navigation system and ,in, Indicates no fault. Indicates a fault, judge according to the measured value or True, specifically, there are four situations according to the judgment combination of the measurement value:

[0099] False positives: True, judge is true, and the probability of its judgment is recorded as: ,Right now ;

[0100] Missed detection: True, judge is true, and the probability of its judgment is recorded as: ,Right now ;

[0101] No false detection: True, judge is true, and the probability of its judgment is recorded as: ;

[0102] Correct detection: True, judge is true, and the probability of its judgment is recorded as: ;

[0103] Assumed measurement value The entire observation space is , to judge by the measured value, the measurement space needs to be divided into two non-intersecting subspaces and , but:

[0104] (1)

[0105] If the judgment is based on the measured value, the judgment criteria can be designed: When is true; when When is true. exist and The conditional probability density when is true is denoted by and , then the above probability can be written as:

[0106] (2)

[0107] (3)

[0108] In the present invention, fault detection consists in making the detection probability As large as possible and the probability of false detection As small as possible, but since the above-mentioned as large as possible and as small as possible are relative concepts and are for comparison, in the implementation process, since the above-mentioned single judgment criterion has errors for the overall judgment result, it is necessary to select a suitable judgment criterion so that the two probabilities reach moderate values. In the autonomous navigation system, the measurement is a known quantity, and it is necessary to compare the probabilities of the two hypotheses under the condition of the measurement value. Under the conditions and The probability of being true is recorded as the posterior probability and So, when The probability is greater than , the system is judged to have a fault. is true; when The probability is greater than , it is determined that there is no fault in the system. is true, the above judgment result can be recorded as:

[0109] (4)

[0110] According to the Bayesian formula, we get:

[0111] (5)

[0112] in, express The corresponding total probability density function is, express The probability of being true, express The probability of being true, and

[0113] (6)

[0114] Substituting formula (5) into formula (6), we get:

[0115] (7)

[0116] Right now

[0117] (8)

[0118] is the statistical detection parameter.

[0119] Based on the above statistical detection theoretical analysis, the theoretical analysis and formula derivation of the traditional fault detection method are given through the chi-square verification method.

[0120] Then the residual error of fault detection is:

[0121] (9)

[0122] in, is the fault detection residual, is the measured value The entire observation space constitutes is the linear regression parameter, One-step predicted value for the state.

[0123] In the formula, the state one-step prediction value for:

[0124] (10)

[0125] in, is the state transfer matrix, is the state prediction value.

[0126] When there is no fault, the residual If the mean is 0, the calculated variance is:

[0127] (11)

[0128] in, is the variance, is the total probability density function, represents the noise covariance matrix, express The transpose of .

[0129] In the case of a fault, the residual The mean of is not 0, so the above variance and residual can be used as verification conditions to determine whether the system has a fault, that is, the residual The binary hypothesis is as follows:

[0130] No fault, then ;

[0131] If there is a fault, ;

[0132] in, represents mathematical expectation, superscript represents the transpose operation, represents the mean of the residuals;

[0133] The fault detection function is:

[0134] (12)

[0135] In the formula, The degrees of freedom are of distributed, express The number of independent variables in represents chi-square, that is , the superscript -1 indicates the inverse operation.

[0136] The fault judgment criteria are:

[0137] (13)

[0138] in, is the fault threshold, and the false detection rate Sure.

[0139] Further, in the scene-adaptive multi-navigation source dynamic optimization and evaluation method of the present invention, the evaluation data of online evaluation without faults is processed to obtain all system-available sensors corresponding to the navigation source evaluation data, including:

[0140] Based on the evaluation data of navigation sources without faults, available sensors are determined and sorted. After sorting, a sensor subset corresponding to the navigation source is obtained through generation, merging and pruning. The navigation accuracy and space occupancy cost of the sensors in the sensor subset are calculated. A comprehensive judgment is made on the sensors after the navigation accuracy is calculated and the space occupancy cost is calculated, and all system-available sensors that meet the usage conditions and requirements are selected.

[0141] In some optional implementations of this embodiment, for the convenience of comparison, the navigation accuracy of all sensors available to the system can be calculated by using the position error of the navigation system as a measure of navigation accuracy, wherein the average error value of the longitude error is expressed as Indicated in m, the average error value of latitude error is expressed as Indicated in m, the average error value of height error is expressed as In order to integrate the position error into an indicator, the square sum of the longitude, latitude and altitude errors is averaged and then squared to obtain the comprehensive position evaluation index. , in m, the comprehensive location evaluation index As a measure of navigation accuracy.

[0142] Specifically, based on the above settings, the calculation expression of the navigation accuracy of the sensor is:

[0143] .

[0144] Specifically, the calculation of the above-mentioned computational time cost is represented by the simulation running time consumption, the unit of which is s.

[0145] In summary, the scene-adaptive multi-navigation source dynamic optimization and evaluation system and method of the present invention evaluates the navigation sources, selects all system-available sensors, calculates the navigation accuracy and space occupancy cost of all selected system-available sensors, selects the optimal sensor from all system-available sensors, and can evaluate the usage scenarios of a single type of sensor, optimize sensors that are applicable and compatible with multiple scenarios, support sensor optimization in different application scenarios such as aerospace scenes, on and under water, or indoors and outdoors, and has good flexibility.

[0146] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In addition, "front", "back", "left", "right", "upper" and "lower" in this article are all referenced to the placement state shown in the accompanying drawings.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A scene-adaptive multi-navigation source dynamic optimization and evaluation system, characterized in that: include: A storage module, wherein the storage module is used to store navigation source assessment data and assessment tasks; An exchange module, the exchange module is used to read the navigation source evaluation data and the evaluation task stored in the storage module, convert the navigation source evaluation data to obtain conversion data, perform parameter initialization settings on the evaluation task, and store and playback the converted navigation source evaluation data and the evaluation task with parameter initialization settings; A data processing module, the data processing module is used to perform the evaluation task after the exchange module parameters are initialized and set, organize, store and analyze the conversion data output by the exchange module according to the evaluation task, and output the specified data for navigation source optimization and evaluation; An online evaluation optimization module, the online evaluation optimization module is used to perform online evaluation on the navigation source evaluation data according to the conversion data output by the exchange module and the specified data output by the data processing module, and judge whether the online evaluation data has a fault in a binary hypothesis manner, output the evaluation data with a fault, generate an evaluation report, and process the evaluation data without a fault in the online evaluation to obtain all system-available sensors corresponding to the navigation source evaluation data; A quantitative evaluation module, the quantitative evaluation module is used to receive data from all sensors available to the system, calculate whether the navigation accuracy and space occupation cost of all sensors available to the system obtained exceed preset values, and select the best sensor from all sensors available to the system that do not exceed the preset values; An evaluation report generation and demonstration module, the evaluation report generation and demonstration module is used to receive the evaluation report generated by the online evaluation optimization module, as well as the data that the navigation accuracy and space occupation cost of all system-available sensors calculated by the quantitative evaluation module do not exceed preset values, and display the received data and the received evaluation report in a visual manner; The online evaluation optimization module includes: A fault detection and isolation module, wherein the fault detection and isolation module is used to perform fault detection on the navigation source evaluation data, isolate the navigation source evaluation data with faults, select the navigation source evaluation data without faults, and obtain the available sensor; A scheme design module, the scheme design module is used to sort the available sensors, and obtain a sensor subset corresponding to the navigation source after generating, merging and pruning the sensors after sorting; An evaluation standard module is used to calculate the navigation accuracy and space occupation cost of the sensors in the sensor subset obtained by the scheme design module to obtain all sensors available to the system.

2. The scene-adaptive multi-navigation source dynamic optimization and evaluation system according to claim 1, characterized in that: Determining whether there is a fault in the online evaluation data in a binary hypothesis manner includes setting a binary hypothesis fault verification function in the online evaluation optimization module, calling the binary hypothesis fault verification function when detecting the fault, and the binary hypothesis fault verification function expression is: ； in, Indicates that the navigation source obeys the degree of freedom of distributed, express The number of independent variables in represents chi-square, Represents the fault threshold, which is determined by the false detection rate of the navigation source.

3. The scene-adaptive multi-navigation source dynamic optimization and evaluation system according to claim 1, characterized in that: The scheme design module includes: A generation module, the generation module is used to expand the available sensors obtained by the fault detection and isolation module to obtain a candidate sensor subset; A merging module, the merging module is used to merge the repeated candidate sensor subsets generated by the generating module; A pruning module is used to prune the candidate sensor subset merged by the merging module according to a preset sensor quantity upper limit threshold to obtain a sensor subset corresponding to the navigation source.

4. The scene-adaptive multi-navigation source dynamic optimization and evaluation system according to claim 1, characterized in that: The evaluation criteria modules include: A navigation accuracy determination module, the navigation accuracy determination module is used to calculate the navigation accuracy of the sensors in the sensor subset obtained by the scheme design module; A cost calculation module, the cost calculation module is used to calculate the space occupation cost of the sensors in the sensor subset obtained by the scheme design module; A comprehensive analysis module is used to make a comprehensive judgment based on the navigation accuracy calculation result of the navigation accuracy judgment module and the space occupancy cost calculation result of the cost calculation module, and select all system-available sensors that meet the use conditions and use requirements.

5. A scene-adaptive multi-navigation source dynamic optimization and evaluation method, characterized in that: include: Acquire navigation source evaluation data and evaluation tasks, convert the navigation source evaluation data to obtain conversion data, perform parameter initialization settings on the evaluation tasks, and store the converted navigation source evaluation data and the evaluation tasks with the parameter initialization settings; Executing the evaluation task after the parameter initialization setting, and collating, storing and analyzing the conversion data according to the evaluation task executed, and outputting the specified data for navigation source optimization and evaluation; Performing online evaluation on the navigation source evaluation data according to the conversion data and the designated data, and judging whether the online evaluation data has a fault in a binary hypothesis manner, outputting the evaluation data with a fault, generating an evaluation report, and processing the evaluation data without a fault in the online evaluation to obtain all system-available sensors corresponding to the navigation source evaluation data; Calculate whether the navigation accuracy and space occupation cost of all the sensors available to the system exceed the preset values, and select the best sensor from all the sensors available to the system that do not exceed the preset values; The evaluation report and the obtained data that the navigation accuracy and space occupation cost of all sensors available for the system do not exceed preset values ​​are displayed in a visual manner; Calculating whether the accuracy and space occupancy cost of all sensors available in the system exceed the preset values ​​includes real-time calculation evaluation and post-calculation evaluation. The real-time calculation evaluation includes analyzing the evaluation data, calculating the evaluation results in real time and displaying them, and writing them into the database for storage. When the evaluation results exceed the standard, an alarm is issued for the exceeded results and the evaluation is terminated. The post-calculation evaluation includes reading the evaluation data from the database and performing evaluation analysis. Process the evaluation data of the online evaluation without faults to obtain the navigation source evaluation data corresponding to all the sensors available in the system, including: Based on the evaluation data of navigation sources without faults, available sensors are determined and sorted. After sorting, a sensor subset corresponding to the navigation source is obtained through generation, merging and pruning. The navigation accuracy and space occupancy cost of the sensors in the sensor subset are calculated. A comprehensive judgment is made on the sensors after the navigation accuracy is calculated and the space occupancy cost is calculated, and all system-available sensors that meet the usage conditions and requirements are selected.

6. The scene-adaptive multi-navigation source dynamic optimization and evaluation method according to claim 5, characterized in that: Determining whether online evaluation data is faulty by binary hypothesis includes: Assume that the navigation data corresponding to the navigation source device is set to two states: faulty and non-faulty. Indicates no fault. Indicates a fault, judge according to the measured value or If true, there are four cases: False positives: True, judge is true, the probability is ; Missed detection: True, judge is true, the probability is ; No false detection: True, judge is true, the probability is ; Correct detection: True, judge is true, the probability is ; Measurement value The entire observation space is , judged by the measured value, the measurement space is divided into two non-intersecting subspaces and ; According to the Bayesian formula, we get: ； According to the chi-square test, the residual of fault detection is: ； When there is no fault in the system, the residual mean is 0, and the calculated variance is: ； Therefore, the above residuals and variances are used as verification conditions to determine whether the navigation source equipment has failed. The binary hypothesis is as follows: No fault, then ; If there is a fault, ; The fault detection function is: ； The fault judgment criteria are: ； in, and Respectively represent the given measurement value Under the conditions and The probability of being true, and Respectively indicate the measured value exist and The conditional probability density when is true, express The probability of being true, express The probability of being true, represents the residual, is the linear regression parameter, represents the one-step prediction value of the state, represents the variance, is the total probability density function, represents the noise covariance matrix, represents the mathematical expectation, represents the mean of the residuals, represents the fault threshold, which is determined by the false detection rate. The degrees of freedom are of distributed, express The number of independent variables in represents chi-square, superscript represents a transpose operation, and the superscript -1 represents an inverse operation.

7. The scene-adaptive multi-navigation source dynamic optimization and evaluation method according to claim 5, characterized in that: The navigation accuracy obtained for all sensors available to the system is determined using the following formula: ； in, represents a comprehensive position evaluation index, which is used as a measure of navigation accuracy. represents the average error value of longitude error, The average error value of latitude error, Indicates the average error value of the height error.

Images