A data fusion method and system based on aircraft-side multi-sensor

Abstract

The present invention relates to the field of data fusion, and in particular to a data fusion method and system based on multiple sensors on an aircraft, the method comprising: calculating the correlation between data collected by any two sensors, constructing a correlation matrix based on the correlation, each sensor corresponding to a row vector; calculating the independent correlation of the data collected by each sensor; constructing a statistical feature vector for each sensor; constructing a difference vector for the sensor, the elements in the difference vector being the difference between the modulus of the statistical feature vector of the corresponding sensor and the modulus of the statistical feature vectors of other sensors, and taking the modulus of the difference vector as the statistical feature consistency of the data collected by the sensor; calculating the weight of the data collected by the sensor, and performing weighted summation on the collected data according to the weight proportion of each sensor to obtain the data fusion result of multiple sensors. The present invention can improve the accuracy of the data fusion result.

Description

Technical Field

[0001] The present invention relates to the field of data fusion, and in particular to a data fusion method and system based on multiple sensors on an aircraft. Background Art

[0002] When an aircraft flies under complex weather and terrain conditions, a single sensor cannot provide comprehensive information. It is necessary to collect data through multiple sensors to improve environmental perception capabilities and improve the overall performance of the system by integrating, correcting and interpreting data from different sensors. By fusing data from different sensors, the accuracy of the aircraft's positioning, speed and attitude estimation can be significantly improved to ensure the accuracy of the flight path. It also provides the flight control system with more comprehensive environmental data to help it make quick responses and decisions in a dynamic environment. Data fusion is the basis for realizing autonomous driving and intelligent flight.

[0003] The patent document with announcement number CN115659284B discloses a big data fusion platform, which obtains a sensor data sequence and each sensor data therein through a data acquisition module, obtains the distribution trend value, period distribution value, residual and period corresponding to the sensor data sequence according to the period acquisition module, obtains the period distribution difference and distribution trend difference through the distribution trend value, period distribution value and period according to the distribution difference feature acquisition module, and finally obtains the fusion weight value of each sensor data in the sensor data sequence at each time point through the period distribution difference, distribution trend difference and residual according to the fusion weight value acquisition module, completes data fusion according to the fusion weight value and stores it in the big data fusion platform for storage and management.

[0004] In the process of multi-sensor data fusion, the above scheme uses a data fusion method to calculate a fixed fusion weight of the sensor and perform subsequent multi-sensor data fusion based on the weight. However, in the process of multi-sensor data fusion on the aircraft side, due to the dynamic complexity of the high-altitude environment, the data collected by each sensor will be distributed unevenly in a large range. The existing data fusion method will cause the sensor with a large weight to dominate the subsequent fusion process. For example, if a sensor with a high weight suddenly fails in the air, the error information will dominate the data fusion process, thereby affecting the accuracy of the data fusion result and the decision-making performance of the entire system. Summary of the invention

[0005] In order to solve the problem that the fusion weights of sensors are fixed during the data fusion process and affect the accuracy of the data fusion results, the present invention provides a data fusion method and system based on multiple sensors on the aircraft side.

[0006] In a first aspect, the present invention provides a data fusion method based on multiple sensors on an aircraft, which adopts the following technical solution:

[0007] Get the data collected by each sensor, calculate the correlation between the data collected by any two sensors, and build a correlation matrix based on the correlation, with each sensor corresponding to a row vector;

[0008] Calculate the cumulative sum of the squares of the differences between the elements in the row vector corresponding to the sensor and the element mean, calculate the independent correlation of the data collected by each sensor, and the independent correlation is positively correlated with the cumulative sum of the squares of the differences; construct the statistical feature vector of each sensor;

[0009] Construct the difference vector of the sensor. The elements in the difference vector are the difference between the modulus of the statistical feature vector of the corresponding sensor and the modulus of the statistical feature vectors of other sensors. The modulus of the difference vector is used as the statistical feature consistency of the sensor collected data. Calculate the weight of the sensor collected data. The weight is positively correlated with the sum of the statistical feature consistency and independent correlation. According to the weight ratio of each sensor, the collected data is weighted and summed to obtain the data fusion result of multiple sensors.

[0010] Among them, the expression of the statistical eigenvector is: , Represents the statistical feature vector of the data collected by the jth sensor at the tth time, represents the rate of change of the data collected by the jth sensor at the tth time, It represents the relative prominence of the data collected by the jth sensor at the tth time, and the relative prominence represents the difference from other data.

[0011] The effect is that in the data fusion process of multiple sensors on the aircraft side, weights are assigned to the data collected by each sensor, and the multi-sensor data at the same moment are fused according to the data collected by each sensor at the same moment and its corresponding weights. Compared with the traditional data fusion based on the fixed fusion weights of the sensors, it reduces the dependence on the sensors themselves and puts the focus of the fusion weights on the data itself. At the same time, the fusion weights can be adaptively adjusted according to the collected data, thereby improving the accuracy of the data fusion results.

[0012] Preferably, the expression of independent correlation is:

[0013]

[0014] In the formula, The independent correlation of the data collected by the jth sensor at the tth time, is the jth sensor and the The correlation of the sensors, is the mean of the elements in the row vector corresponding to the jth sensor in the correlation matrix, and m is the number of sensors.

[0015] The effect is that by calculating the independent correlation of the collected data, the importance of the data collected by the corresponding sensor can be intuitively understood.

[0016] Preferably, the expression of the correlation matrix is:

[0017]

[0018] In the formula, is the correlation matrix, is the correlation between the data collected by the first sensor and the first sensor at the tth moment, is the correlation between the data collected by the 1st sensor and the mth sensor at the tth time, is the correlation between the data collected by the mth sensor and the first sensor at the tth time, It is the correlation between the data collected by the mth sensor and the mth sensor at the tth time.

[0019] Preferably, the expression of relative prominence is: ;

[0020] In the formula, is the relative prominence of the data collected by the jth sensor at the tth time among all the data collected within the preset time period, is the data collected by the jth sensor at the tth time, is the median of the data collected by the jth sensor in the T minutes before the tth moment, is the maximum value of the data collected by the jth sensor within T minutes before the tth moment, It is the minimum value of the data collected by the j-th sensor within T minutes before the t-th time.

[0021] The effect is that by calculating the relative prominence of the collected data, it is possible to understand the changes in the data collected by the corresponding sensors.

[0022] Preferably, the weight of the sensor data is expressed as:

[0023]

[0024] In the formula, is the weight of the data collected by the jth sensor at the tth time, is the statistical characteristic consistency of the data collected by the jth sensor at the tth time, is the independent correlation of the jth sensor at the tth time, is an exponential function with base e.

[0025] Preferably, the weight of the sensor data is expressed as:

[0026]

[0027] In the formula, is the weight of the data collected by the jth sensor at the tth time, is the statistical characteristic consistency of the data collected by the jth sensor at the tth time, is the independent correlation of the jth sensor at the tth time, is an exponential function with base e.

[0028] The effect is that by calculating the weight of the collected data, it is easy to fuse the data collected by multiple sensors, thereby improving the accuracy of data fusion.

[0029] Preferably, the expression of data fusion is:

[0030]

[0031] In the formula, is the data after the fusion of the collected data of multiple sensors at the tth moment, is the data collected by the jth sensor at the tth time, is the weight of the data collected by the jth sensor at the tth time, is the weight of the data collected by the rth sensor at the tth time, and m is the number of sensors.

[0032] The effect is that the weight of the sensor data is used to fuse the collected data, thereby improving the accuracy of the fused data.

[0033] In a second aspect, the present invention provides a data fusion system based on multiple sensors on an aircraft, which adopts the following technical solution:

[0034] A data fusion system based on multiple sensors on an aircraft side, a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a data fusion method based on multiple sensors on an aircraft side is implemented.

[0035] The effect is that the above-mentioned data fusion method based on multiple sensors on the aircraft side is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a system based on the memory and the processor for easy use.

[0036] The present invention has the following technical effects:

[0037] In the data fusion process of multiple sensors on the aircraft side, weights are assigned to the data collected by each sensor, and the multi-sensor data at the same moment are fused based on the data collected by each sensor at the same moment and its corresponding weights. Compared with the traditional data fusion based on the fixed fusion weights of the sensors, this reduces the dependence on the sensors themselves and puts the focus of the fusion weights on the data itself. At the same time, the fusion weight ratio of the corresponding sensors can be adaptively adjusted according to the collected data, thereby improving the accuracy of the data fusion results. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0039] Figure 1 The present invention is a method flow chart of a data fusion method based on multiple sensors on an aircraft side according to an embodiment of the present invention.

[0040] Figure 2 It is a structural block diagram of a data fusion system based on multiple sensors on the aircraft side according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] 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 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.

[0042] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0043] The embodiment of the present invention discloses a data fusion method based on multiple sensors on an aircraft side, referring to Figure 1 , including the following steps:

[0044] S1: Obtain the data collected by each sensor, calculate the correlation between the data collected by any two sensors, and construct a correlation matrix based on the correlation, with each sensor corresponding to a row vector.

[0045] For multiple sensors on the aircraft, the same acquisition frequency is set, and multiple sensors collect corresponding data at the same time. The DTW algorithm is used to calculate the correlation between the data collected by any two sensors, and then a correlation matrix is ​​constructed. The expression of the correlation matrix is:

[0046]

[0047] In the formula, is the correlation matrix, is the correlation between the data collected by the first sensor and the first sensor at the tth moment, is the correlation between the data collected by the 1st sensor and the mth sensor at the tth time, is the correlation between the data collected by the mth sensor and the first sensor at the tth time, is the correlation between the data collected by the mth sensor and the mth sensor at the tth moment. Through the correlation matrix, we can intuitively understand the correlation between any two sensors, and thus understand the data relationship between the sensors. Each row in the correlation matrix is ​​regarded as a row vector. It can be seen from the correlation matrix that each sensor corresponds to a row vector. For example, the first sensor corresponds to the first row vector in the correlation matrix, and the mth sensor corresponds to the mth row vector in the correlation matrix.

[0048] S2: Calculate the cumulative sum of the squares of the differences between the elements in the row vector corresponding to the sensor and the element mean, and calculate the independent correlation of the data collected by each sensor. The independent correlation is positively correlated with the cumulative sum of the squares of the differences.

[0049] In one embodiment, the expression of the independent correlation is:

[0050]

[0051] In the formula, The independent correlation of the data collected by the jth sensor at the tth time, is the correlation between the jth sensor and the rth sensor at the tth moment, is the mean of the elements in the row vector corresponding to the jth sensor in the correlation matrix, and m is the number of sensors.

[0052] In other embodiments, another method for calculating the independent correlation different from that described in the above step S2 is provided, and the expression of the independent correlation is:

[0053]

[0054] In the formula, The independent correlation of the jth sensor at time t, is the correlation between the jth sensor and the rth sensor at the tth moment, is the mean of the elements in the row vector corresponding to the jth sensor in the correlation matrix, and m is the number of sensors.

[0055] S3: Construct the statistical feature vector of each sensor.

[0056] Calculate the rate of change and relative prominence of the data collected by the sensor:

[0057]

[0058]

[0059] In the formula, is the rate of change of the data collected by the jth sensor at the tth time, is the data collected by the jth sensor at the tth time, is the data collected by the jth sensor at time t-1, is the relative prominence of the data collected by the jth sensor at the tth time among all the data collected within the preset time period. The relative prominence indicates the difference from other data. is the median of the data collected by the jth sensor in the T minutes before the tth moment, is the maximum value of the data collected by the jth sensor within T minutes before the tth moment, It is the minimum value of the data collected by the j-th sensor within T minutes before the t-th time.

[0060] in, It is the difference between the characteristic value of the data collected by the jth sensor at the tth time and the data collected in the previous T minutes. Here, the characteristic value is the median of the collected data. The larger the value, the greater the relative prominence of the corresponding collected data in the corresponding time period. It is the distribution range of the data collected in the corresponding time period. The smaller the value is, the greater the relative prominence of the corresponding data collected is.

[0061] The statistical feature vector is constructed according to the change rate and relative prominence of the data collected by the sensor. The expression of the statistical feature vector is:

[0062]

[0063] Indicates Moment The statistical feature vector of the data collected by the sensor is represents the rate of change of the data collected by the jth sensor at the tth time, It represents the relative prominence of the data collected by the jth sensor at the tth time.

[0064] S4: Calculate the statistical feature consistency of the sensor collected data.

[0065] A sensor difference vector is constructed, wherein the elements in the difference vector are the difference between the modulus of the statistical feature vector of the corresponding sensor and the modulus of the statistical feature vectors of other sensors, and the modulus of the difference vector is used as the statistical feature consistency of the sensor collected data.

[0066] The expression for the difference vector is:

[0067] In the formula, is the difference vector of the data collected by the jth sensor at the tth time, is the statistical feature vector of the data collected by the jth sensor at the tth time The model, is the statistical feature vector of the data collected by the first sensor at time t The model, is the statistical feature vector of the data collected by the second sensor at time t The model, is the statistical feature vector of the data collected by the m-1th sensor at the tth time Model.

[0068] The expression of statistical feature consistency is: , where represents the statistical consistency of the data collected by the jth sensor at the tth time, Represents the modulus of the difference vector of the data collected by the jth sensor at the tth time.

[0069] Different sensors may provide information about different dimensions of the same target. When there is a strong correlation between sensor data, it can be considered that the data has higher quality and reliability and can provide more accurate data.

[0070] S5: Calculate the weight of the sensor collected data. The weight is positively correlated with the consistency of statistical features and the sum of independent correlation.

[0071] In one embodiment, the weight of the sensor collected data is expressed as:

[0072]

[0073] In the formula, is the weight of the data collected by the jth sensor at the tth time, is the statistical characteristic consistency of the data collected by the jth sensor at the tth time, is the independent correlation of the jth sensor at the tth time, is an exponential function with base e.

[0074] In other embodiments, the weight of the sensor collected data is expressed as:

[0075]

[0076] In the formula, is the weight of the data collected by the jth sensor at the tth time, is the statistical characteristic consistency of the data collected by the jth sensor at the tth time, is the independent correlation of the jth sensor at the tth time, is an exponential function with base e.

[0077] S6: Perform weighted summation on the collected data according to the weight ratio of each sensor to obtain the data fusion result of multiple sensors.

[0078] The expression of data fusion is:

[0079]

[0080] In the formula, is the data after the fusion of the collected data of multiple sensors at the tth moment, is the data collected by the jth sensor at the tth time, is the weight of the data collected by the jth sensor at the tth time, is the weight of the data collected by the rth sensor at the tth time, and m is the number of sensors.

[0081] The collected data are fused according to the weights of each collected data to obtain the final fusion result. In the process of calculating the weights of the collected data, factors such as the statistical characteristic consistency and independent correlation of the collected data are referred to. Compared with the traditional method of fusing data using fixed weights of sensors, this method puts the focus of the weight on the collected data itself, avoiding dependence on the sensor itself, and when one of the sensors fails, it can adaptively adjust the weight ratio of the data collected by the remaining sensors, thereby improving the accuracy of the fusion result and having strong robustness.

[0082] Combination Figure 2 As shown, an embodiment of the present invention also discloses a data fusion system based on multiple sensors on an aircraft side, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a data fusion method based on multiple sensors on an aircraft side according to the present invention is implemented.

[0083] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0084] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high bandwidth memory HBM (High Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0085] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0086] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A data fusion method based on multiple sensors on an aircraft, characterized in that: Includes steps: Get the data collected by each sensor, calculate the correlation between the data collected by any two sensors, and build a correlation matrix based on the correlation, with each sensor corresponding to a row vector; Calculate the cumulative sum of the squares of the differences between the elements in the row vector corresponding to the sensor and the element mean, and calculate the independent correlation of the data collected by each sensor. The expression of independent correlation is: In the formula, The independent correlation of the data collected by the jth sensor at the tth time, is the correlation between the jth sensor and the rth sensor at the tth moment, is the mean of the elements in the row vector corresponding to the jth sensor in the correlation matrix, and m is the number of sensors; construct the statistical feature vector of each sensor; Construct the difference vector of the sensor. The elements in the difference vector are the difference between the modulus of the statistical feature vector of the corresponding sensor and the modulus of the statistical feature vectors of other sensors. The modulus of the difference vector is used as the statistical feature consistency of the sensor collected data. Calculate the weight of the sensor collected data. The weight is positively correlated with the sum of the statistical feature consistency and independent correlation. According to the weight ratio of each sensor, the collected data is weighted and summed to obtain the data fusion result of multiple sensors. Among them, the expression of the statistical eigenvector is: , Represents the statistical feature vector of the data collected by the jth sensor at the tth time, represents the rate of change of the data collected by the jth sensor at the tth time, Represents the relative prominence of the data collected by the jth sensor at the tth time. The relative prominence represents the difference from other data. The expression of relative prominence is: In the formula, is the relative prominence of the data collected by the jth sensor at the tth time among all the data collected within the preset time period, is the data collected by the jth sensor at the tth time, is the median of the data collected by the jth sensor in the T minutes before the tth moment, is the maximum value of the data collected by the jth sensor within T minutes before the tth moment, It is the minimum value of the data collected by the jth sensor within T minutes before the tth moment.

2. The data fusion method based on aircraft-side multi-sensor according to claim 1 is characterized in that: The expression of the correlation matrix is: In the formula, is the correlation matrix, is the correlation between the data collected by the first sensor and the first sensor at the tth moment, is the correlation between the data collected by the 1st sensor and the mth sensor at the tth time, is the correlation between the data collected by the mth sensor and the first sensor at the tth time, It is the correlation between the data collected by the mth sensor and the mth sensor at the tth time.

3. The data fusion method based on aircraft-side multi-sensor according to claim 1 is characterized in that: The expression of the weight of the sensor collected data is: In the formula, is the weight of the data collected by the jth sensor at the tth time, is the statistical characteristic consistency of the data collected by the jth sensor at the tth time, is the independent correlation of the jth sensor at the tth time, is an exponential function with base e.

4. The data fusion method based on aircraft-side multi-sensor according to claim 1 is characterized in that: The expression of the weight of the sensor collected data is: In the formula, is the weight of the data collected by the jth sensor at the tth time, is the statistical characteristic consistency of the data collected by the jth sensor at the tth time, is the independent correlation of the jth sensor at the tth time, is an exponential function with base e.

5. The data fusion method based on aircraft-side multi-sensor according to claim 1 is characterized in that: The expression of data fusion is: In the formula, is the data after the fusion of the collected data of multiple sensors at the tth moment, is the data collected by the jth sensor at the tth time, is the weight of the data collected by the jth sensor at the tth time, is the weight of the data collected by the rth sensor at the tth time, and m is the number of sensors.

6. A data fusion system based on multiple sensors on an aircraft, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a data fusion method based on aircraft-side multi-sensors according to any one of claims 1 to 5 is implemented.

Images