Distributed asynchronous consistency filtering method, electronic device and computer storage medium

Abstract

The present invention discloses a distributed asynchronous consistency filtering method, an electronic device and a computer storage medium, the method comprising: each sensor node in an asynchronous sensor network obtains its own measurement value through sampling; each sensor node performs local filtering at its own asynchronous sampling moment based on the measurement value obtained by its own sampling using a preset filtering algorithm to obtain its own local estimation; at the consistent fusion moment, based on the local estimation of each sensor node, a weighted average consistency iteration is performed on each sensor node, and after a preset number of iterations, a consistent filtering result is obtained. The distributed asynchronous consistency filtering method proposed in the present invention overcomes the problem that the traditional consistency filtering algorithm requires strict synchronization of sensor sampling in time, and is applicable to asynchronous situations such as different sampling rates and different initial sampling moments of sensor nodes in the network. Moreover, the method proposed in the present invention is also applicable to situations where sensor nodes are non-uniformly sampled.

Description

Technical Field

[0001] The present invention relates to the technical field of distributed consistency filtering of asynchronous sensor networks, and in particular to a distributed asynchronous consistency filtering method, electronic equipment and computer storage medium. Background Art

[0002] A sensor network is an environmental perception network composed of a large number of sensor nodes deployed in different spatial locations. Each sensor node is connected through a network, offering advantages such as flexible structure, strong scalability, wide spatial coverage, and high reliability. In recent years, with technological advances and reduced sensor costs, wireless sensor network technology has experienced unprecedented development and has been widely used in fields such as battlefield surveillance, traffic control, environmental monitoring, and machine health monitoring.

[0003] Distributed filtering is a fundamental problem in wireless sensor networks. It studies how sensor nodes can accurately estimate the state of the measured object based on their own information and that of their neighbors. Distributed filtering in networked systems has garnered widespread attention and yielded numerous research results. Existing distributed consistency filtering algorithms for networked systems can be broadly categorized into three main types, depending on the type of information shared within the network: local estimation-based distributed consistency filtering algorithms, measurement-based distributed consistency filtering algorithms, and innovation-based distributed consistency filtering algorithms.

[0004] The aforementioned research on distributed consensus filtering has all been conducted for synchronous sensor networks, where all sensor nodes in the network perform measurements synchronously. In reality, different sensor nodes may have different sampling rates, initial sampling times, and communication delays, resulting in measurements between sensor nodes often being asynchronous rather than synchronous. However, research on distributed asynchronous consensus filtering for networked systems has yet to be conducted.

[0005] Existing distributed consistency filtering algorithms all consider the case of synchronous sensor networks, assuming that all sensor nodes in the network perform measurements synchronously. However, in real sensor networks, sensors at different nodes can be heterogeneous. Even sensors of the same type may have different sampling rates or different start sampling times. Furthermore, information transmission between sensors may also experience varying transmission delays. These factors can lead to asynchronous situations, for which no suitable distributed consistency filtering solutions currently exist. Summary of the Invention

[0006] The present invention provides a distributed asynchronous consistency filtering method, electronic device and computer storage medium to solve the technical problem that existing distributed consistency filtering algorithms only consider the situation of synchronous sensor networks, and the research on distributed consistency filtering problems of asynchronous sensor networks has not yet been seen.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In one aspect, the present invention provides a distributed asynchronous consistency filtering method applicable to asynchronous sensor networks, the distributed asynchronous consistency filtering method comprising:

[0009] Each sensor node in an asynchronous sensor network obtains its own measurement value through sampling;

[0010] Each sensor node uses a preset filtering algorithm to perform local filtering at its own asynchronous sampling time based on the measurement value obtained by its own sampling to obtain its own local estimate;

[0011] At the time of consistent fusion, the local estimation results of the sensor nodes closest to the consistent fusion moment are synchronized to the consistent fusion moment; and based on the local estimation of each sensor node after synchronization, weighted average consistency iteration is performed on each sensor node. After a preset number of iterations, a consistent filtering result is obtained.

[0012] Furthermore, each sensor node performs local filtering at its own asynchronous sampling moment using a preset filtering algorithm based on its own sampled measurement values ​​to obtain its own local estimation, including:

[0013] Each sensor node uses the Kalman filter algorithm to perform local filtering at its own asynchronous sampling time based on the measurement value obtained by its own sampling to obtain its own local estimation.

[0014] Furthermore, each sensor node performs local filtering at its own asynchronous sampling moment using a Kalman filter algorithm based on its own sampled measurement values ​​to obtain its own local estimation, including:

[0015] The estimated state value predicted by each sensor node is calculated according to the following formula and the covariance matrix of the estimation error

[0016]

[0017] in, Indicates from arrive The state transition matrix of the system at time t, represents the covariance matrix of the process noise, Indicates that the i-th sensor is at the k-th filtering time interval (t k-1 ,t k ] the s-1th local estimate within, Indicates that the i-th sensor is at the k-th filtering time interval (t k-1 ,t k ] the covariance matrix of the s-1th local estimation error within ;

[0018] Calculate the filter gain matrix of each sensor node according to the following formula:

[0019]

[0020] in, represents the measurement matrix, Represents the covariance matrix of measurement noise; calculates the updated state estimate of the local sensor node and the error covariance matrix

[0021]

[0022] in, Indicates that the i-th sensor node is in the k-th filtering time interval (t k-1 ,t k ] is the sth measurement value within Indicates the status of the system.

[0023] Furthermore, the weighted average consistency iteration is performed on each sensor node based on the local estimation after synchronization, and a consistent filtering result is obtained after a preset number of iterations, including:

[0024] Each sensor node sends the synchronized local estimation results to its neighboring nodes, receives the local estimation results of the neighboring nodes at the same time, and performs weighted averaging iteration on the received local estimation results of the neighboring nodes and its own local estimation results to obtain a consistent filtering result.

[0025] Furthermore, at the consistent fusion moment, based on the measurement values ​​of each sensor node closest to the consistent fusion moment, synchronizing the local estimation results thereof to the consistent fusion moment includes:

[0026] Based on the measurement value of the sensor node closest to the consistent fusion moment, its local estimation result is synchronized to the consistent fusion moment. The conversion formula is as follows:

[0027]

[0028] in, Indicates that the i-th sensor node is in the consistent fusion t k The estimated state value at time t, Indicates from to t k The state transition matrix at time , Indicates that the i-th sensor node is in the consistent fusion t k The covariance matrix of the estimation error at each moment, Represents the covariance matrix of the process noise. Furthermore, each sensor node sends the synchronized local estimation results to its neighboring nodes, receives the local estimation results of the neighboring nodes at the same time, and performs weighted averaging iteration on the received local estimation results of the neighboring nodes and its own local estimation results to obtain a consistent filtering result, including:

[0029] Each sensor node broadcasts the synchronized local estimation results to all its neighboring nodes, and receives the local estimation results of all neighboring nodes at the same time. It then performs weighted consistency iteration on the received local estimation results of neighboring nodes and its own local estimation results as shown in the following formula:

[0030]

[0031]

[0032] Repeat the above iterative process to obtain the estimated result after L iterations, and set it as:

[0033]

[0034] Among them, π i,j represents the weighting coefficient, and N i represents the set of neighbor nodes of the i-th sensor node, i represents the i-th sensor node, and They represent the i-th sensor node at t k The covariance matrix of the estimated value and the estimated error after l+1 weighted consistency iterations at time t, and They represent the i-th sensor node at t k The covariance matrix of the estimated value and the estimated error after l weighted consistency iterations at time , and They represent the i-th sensor node at t k The covariance matrix of the estimated value and estimation error after L weighted consistency iterations at time.

[0035] On the other hand, the present invention further provides an electronic device, comprising a processor and a memory; wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the above method.

[0036] In yet another aspect, the present invention further provides a computer-readable storage medium, wherein the storage medium stores at least one instruction, and the instruction is loaded and executed by a processor to implement the above method.

[0037] The beneficial effects brought about by the technical solution provided by the present invention include at least:

[0038] In the distributed asynchronous consistency filtering method proposed in the present invention, the sensor nodes first perform local filtering at their respective asynchronous sampling moments, and synchronize their local estimation results to the consistent fusion moment based on the measurement values ​​of the sensor nodes closest to the consistent moment. Then, at the consistent fusion moment, consistency iteration is performed by sharing the synchronized local estimates to obtain a consistent filtering result that is better than the local estimation. This overcomes the problem that traditional consistency filtering algorithms require strict temporal synchronization of sensor sampling. The distributed asynchronous consistency filtering method proposed in the present invention is applicable to situations where sensor nodes in the network have different sampling rates, different initial sampling moments, and other asynchronous situations. Moreover, the method of the present invention is also applicable to situations where sensor nodes are non-uniformly sampled. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 1 is a flow chart of a distributed asynchronous consistency filtering method provided by the first embodiment of the present invention;

[0041] Figure 2 This is a topological diagram of asynchronous sensor network communication consisting of four sensors;

[0042] Figure 3 is a sampling graph of an asynchronous sensor network;

[0043] Figure 4 2 is a flow chart of a distributed asynchronous consistency filtering method provided by the second embodiment of the present invention;

[0044] Figure 5 is an estimation curve graph obtained by each sensor node in the asynchronous sensor network using the distributed asynchronous consistency filtering method provided by the second embodiment of the present invention;

[0045] Figure 6 This is a root mean square error curve diagram of the third sensor node in the asynchronous sensor network after adopting the distributed asynchronous consistency filtering method provided by the second embodiment of the present invention and standard Kalman filtering. DETAILED DESCRIPTION

[0046] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0047] First embodiment

[0048] In view of the problem that the existing distributed consistency filtering method is only applicable to synchronous sensor networks, this embodiment provides a distributed asynchronous consistency filtering method, which is applicable to asynchronous sensor networks. The execution process is as follows: Figure 1 As shown, the following steps are included:

[0049] S1, each sensor node in the asynchronous sensor network obtains its own measurement value through sampling;

[0050] S2, each sensor node performs local filtering based on the measurement values ​​obtained by its own sampling using a preset filtering algorithm at its own asynchronous sampling time to obtain its own local estimate;

[0051] Specifically, in this embodiment, the filtering algorithm used is the Kalman filtering algorithm. The local filtering is implemented by the Kalman filtering algorithm, and the process of obtaining the local estimation is as follows:

[0052] One-step prediction: Calculate the estimated state value of each sensor node according to the following formula: and the covariance matrix of the estimation error

[0053]

[0054] in, Indicates from arrive The state transition matrix of the system at time t, represents the covariance matrix of the process noise, Indicates that the i-th sensor is at the k-th filtering time interval (t k-1 ,t k ] the s-1th local estimate within, Indicates that the i-th sensor is at the k-th filtering time interval (t k-1 ,t k ] the covariance matrix of the s-1th local estimation error within ;

[0055] Measurement update: Calculate the filter gain matrix of each sensor node according to the following formula

[0056]

[0057] in, represents the measurement matrix, Represents the covariance matrix of the measurement noise; Next, calculate the updated state estimate of the local sensor node and the error covariance matrix

[0058]

[0059] in, Indicates that the i-th sensor node is in the k-th filtering time interval (t k-1 ,t k ] is the sth measurement value within Indicates the status of the system.

[0060] S3, at the time of consistent fusion, based on the measurement values ​​of each sensor node closest to the consistent fusion time, synchronize its local estimation results to the consistent fusion time; and based on the local estimation of each sensor node after synchronization, perform weighted average consistency iteration on each sensor node, and obtain a consistent filtering result after a preset number of iterations.

[0061] Specifically, in this embodiment, the above S3 includes the following steps:

[0062] S31, at the consistent fusion moment, synchronize the local estimation results of the sensor nodes closest to the consistent fusion moment to the consistent fusion moment; the conversion formula is as follows:

[0063]

[0064] in, Indicates that the i-th sensor node is in the consistent fusion t k The estimated state value at time t, Indicates from to t k The state transition matrix at time , Indicates that the i-th sensor node is in the consistent fusion t k The covariance matrix of the estimation error at each moment, Represents the covariance matrix of the process noise.

[0065] In step S32, each sensor node sends the synchronized local estimation result to its neighboring node, receives the local estimation result of the neighboring node, and performs weighted averaging iteration on the received local estimation result of the neighboring node and its own local estimation result to obtain a consistent filtering result. The formula is as follows:

[0066]

[0067]

[0068] Repeat the above iterative process to obtain the estimated result after L iterations, and set it as:

[0069]

[0070] Among them, π i,j represents the weighting coefficient, and N i represents the set of neighbor nodes of the i-th sensor node, i represents the i-th sensor node, and They represent the i-th sensor node at t k The covariance matrix of the estimated value and the estimated error after l+1 weighted consistency iterations at time t, and They represent the i-th sensor node at t k The covariance matrix of the estimated value and the estimated error after l weighted consistency iterations at time, and They represent the i-th sensor node at t k The covariance matrix of the estimated value and estimation error after L weighted consistency iterations at time.

[0071] In summary, in the method of this embodiment, the sensor nodes first perform local filtering at their respective asynchronous sampling moments, and synchronize their local estimation results to the consistent fusion moment based on the measurement values ​​of the sensor nodes closest to the consistent moment. Then, at the consistent fusion moment, consistency iteration is performed by sharing the synchronized local estimates to obtain a consistent filtering result that is better than the local estimation. This overcomes the problem that traditional consistency filtering algorithms require strict temporal synchronization of sensor sampling. The distributed asynchronous consistency filtering method proposed in this embodiment is applicable to asynchronous situations in which sensor nodes in the network have different sampling rates, different initial sampling moments, and so on. Moreover, the method provided in this embodiment is also applicable to situations where sensor nodes are non-uniformly sampled.

[0072] Second embodiment

[0073] To address the problem that existing distributed consistency filtering methods are only applicable to synchronous sensor networks, this embodiment provides a distributed asynchronous consistency filtering method, which is applicable to asynchronous sensor networks.

[0074] Among them, asynchronous sensor network refers to the situation where sensor nodes in the network have different sampling rates, different initial sampling times, etc. The asynchronous sensor network communication topology composed of four sensors is as follows: Figure 2 As shown, the sampling of asynchronous sensor networks is as follows Figure 3 The execution process of the distributed consistency filtering method of this embodiment is as follows Figure 4 As shown, the following steps are included:

[0075] Step 1: Given the initial state estimation value and error covariance matrix of each sensor node in the asynchronous sensor network.

[0076] Step 2: Each sensor node obtains its own asynchronous measurement value through sampling.

[0077] Step 3: Each sensor node obtains its own asynchronous local estimate based on its own asynchronous measurement using the Kalman filter algorithm.

[0078] Step 4: Determine whether the consistent fusion time has been reached. If the consistent fusion time has not been reached, jump to step 2. If the consistent fusion time has been reached, jump to step 5.

[0079] Step 5: Synchronize the local estimation results of the sensor nodes closest to the consistent time to the consistent fusion time.

[0080] Step 6: Send the synchronized local estimation of the sensor node to the neighboring nodes, receive the information of the neighboring nodes at the same time, and perform weighted averaging iteration together with the estimation result of the node itself, and finally obtain the consistent estimation.

[0081] Specifically, in this embodiment, the specific implementation of step 1 is as follows:

[0082] Assume that the initial state of each sensor in the asynchronous sensor network is The mean is The covariance matrix is Gaussian distribution, then this embodiment will and They are used as the initial state estimation value and error covariance matrix of each sensor node respectively.

[0083] The specific implementation of step 2 is as follows:

[0084] In asynchronous sensors, is the value of the i-th sensor at the k-th filtering time interval (t k-1 ,t k ] is the sth measurement value, is the value of the i-th sensor at the k-th filtering time interval (t k-1 ,t k] the number of measured values ​​within, is the value of the i-th sensor at the k-th filtering time interval (t k-1 ,t k ]At the moment of the sth measurement value, the measurement equation is:

[0085]

[0086] in, Indicates the status of the system, Represents measurement noise, which has a mean of 0 and a covariance matrix of Gaussian distribution, represents the measurement matrix.

[0087] The specific implementation of step 3 is as follows:

[0088] Based on the asynchronous measurement values ​​obtained in step 2, the Kalman filtering algorithm is used to perform local filtering on the sensor nodes in the asynchronous sensor network, including the following process:

[0089] One-step prediction: Calculate the estimated state value of each sensor node according to the following formula: and the covariance matrix of the estimation error

[0090]

[0091] in, represents the system state transition matrix, is the covariance matrix of the process noise, satisfying for When s=1,k≠1, where t k-1 Represents the last filtering fusion moment; when s=1,k=1, When s≠1, Represents the i-th sensor at the k-th filtering time interval (t k-1 ,t k ] is the moment at which the s-1th measurement value is located.

[0092] Measurement update: Calculate the filter gain matrix of each sensor node according to the following formula

[0093]

[0094] Next, the updated state estimate of the local sensor node is calculated and the error covariance matrix

[0095]

[0096] The specific implementation of step 5 is as follows:

[0097] Based on the measurement value of the sensor node closest to the consistent time, its local estimation result is synchronized to the consistent fusion time. The conversion formula is as follows:

[0098]

[0099] The specific implementation of step 6 is as follows:

[0100] The conversion filtering results obtained by each sensor node in the asynchronous sensor network in step 5 are set as the initial value of the asynchronous consistency weighted average filtering algorithm:

[0101]

[0102] For each sensor node, weighted average consistency iteration is implemented, and the iteration variable is l∈{0,1,…,L-1}.

[0103] First, the synchronized local estimate and Broadcast to all its adjacent nodes j∈N i , and receive filtering information from all adjacent nodes at the same time and The received filtering information and Same as the node's own filtering information and Perform weighted consistency iteration as shown below

[0104]

[0105]

[0106] Among them, π i,j is the weighting coefficient, and

[0107] Repeat the above iterative process to obtain the estimated result after L iterations and set it as

[0108]

[0109] Return to step 3 and repeat the cycle.

[0110] The estimated curve obtained by each sensor node in the asynchronous sensor network using the distributed asynchronous consistency filtering method provided by this embodiment is as follows: Figure 5As shown; the root mean square error curve after the third sensor node in the asynchronous sensor network adopts the distributed asynchronous consistency filtering method provided by this embodiment and the standard Kalman filtering is as shown Figure 6 As shown in the figure, the distributed asynchronous consistency filtering method provided by this embodiment overcomes the problem that traditional consistency filtering algorithms require strict synchronization of sensor sampling. The distributed asynchronous consistency filtering method proposed in this embodiment is applicable to situations where sensor nodes in the network have different sampling rates, different initial sampling times, and other asynchronous situations. It is also applicable to situations where sensor nodes have non-uniform sampling.

[0111] Third embodiment

[0112] This embodiment provides an electronic device, which includes a processor and a memory; wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method of the first embodiment.

[0113] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) and one or more memories, wherein the memories store at least one instruction, which is loaded by the processor and executes the above method

[0114] Fourth embodiment

[0115] This embodiment provides a computer-readable storage medium that stores at least one instruction. The instruction is loaded and executed by a processor to implement the method of the first embodiment described above. The computer-readable storage medium may be a ROM, random access memory, CD-ROM, magnetic tape, floppy disk, or optical data storage device. The instructions stored therein can be loaded by a processor in a terminal to execute the method described above.

[0116] Furthermore, it should be noted that the present invention may be provided as a method, apparatus, or computer program product. Thus, embodiments of the present invention may take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media containing computer-usable program code.

[0117] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the 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 the 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, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the process 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.

[0118] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for implementing the process in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0119] It should also be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or terminal device comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or terminal device comprising the element.

[0120] Finally, it should be noted that the above is a preferred embodiment of the present invention. It should be noted that although the preferred embodiment of the present invention has been described, it is clear that those skilled in the art, once they understand the basic inventive concept of the present invention, can make various improvements and modifications without departing from the principles of the present invention. Such improvements and modifications should also be considered as within the scope of protection of the present invention. Therefore, the appended claims are intended to be interpreted as including the preferred embodiment and all changes and modifications that fall within the scope of the embodiments of the present invention.

Claims

1. A distributed asynchronous consistency filtering method, suitable for asynchronous sensor networks, characterized in that: The distributed asynchronous consistency filtering method comprises: Each sensor node in an asynchronous sensor network obtains its own measurement value through sampling; Each sensor node uses a preset filtering algorithm to perform local filtering at its own asynchronous sampling time based on the measurement value obtained by its own sampling to obtain its own local estimation; At the time of consistent fusion, based on the measurement values ​​of each sensor node closest to the consistent fusion time, its local estimation results are synchronized to the consistent fusion time; and based on the local estimation of each sensor node after synchronization, weighted average consistency iteration is performed on each sensor node, and after a preset number of iterations, a consistent filtering result is obtained; Each sensor node performs local filtering at its own asynchronous sampling time using a preset filtering algorithm based on the measurement value obtained by its own sampling to obtain its own local estimation, including: Each sensor node uses the Kalman filter algorithm to perform local filtering at its own asynchronous sampling time based on the measurement value obtained by its own sampling to obtain its own local estimation; Each sensor node performs local filtering at its own asynchronous sampling time based on the measurement value obtained by its own sampling using the Kalman filtering algorithm to obtain its own local estimation, including: The estimated state value predicted by each sensor node is calculated according to the following formula and the covariance matrix of the estimation error in, Indicates from arrive The state transfer matrix of the system at time t, represents the covariance matrix of the process noise, It means that the i-th sensor is in the k-th filtering time interval (t k-1 ,t k ] is the s-1th local estimate within It means that the i-th sensor is in the k-th filtering time interval (t k-1 ,t k ] is the covariance matrix of the s-1th local estimation error within ; The filter gain matrix of each sensor node is calculated according to the following formula: in, represents the measurement matrix, represents the covariance matrix of the measurement noise; Calculate the updated state estimate of the local sensor node and the error covariance matrix in, It means that the i-th sensor node is in the k-th filtering time interval (t k-1 ,t k ] is the sth measurement value within Indicates the status of the system.

2. The distributed asynchronous consistency filtering method according to claim 1, characterized in that: The method of performing weighted average consistency iteration on each sensor node based on the local estimation after synchronization of each sensor node, and obtaining a consistent filtering result after a preset number of iterations, includes: Each sensor node sends the synchronized local estimation result to its neighboring nodes, receives the local estimation results of the neighboring nodes at the same time, and performs weighted average iteration on the received local estimation results of the neighboring nodes and its own local estimation results to obtain a consistent filtering result.

3. The distributed asynchronous consistency filtering method according to claim 2, characterized in that: At the consistent fusion moment, based on the measurement values ​​of each sensor node closest to the consistent fusion moment, synchronizing the local estimation results to the consistent fusion moment includes: Based on the measurement value of the sensor node closest to the consistent fusion time, its local estimation result is synchronized to the consistent fusion time. The conversion formula is as follows: in, Indicates that the i-th sensor node is in the consistent fusion t k The estimated state value at time Indicates from to k The state transfer matrix at time , Indicates that the i-th sensor node is in the consistent fusion t k The covariance matrix of the estimation error at each moment, Represents the covariance matrix of the process noise.

4. The distributed asynchronous consistency filtering method according to claim 3, characterized in that: Each sensor node sends the synchronized local estimation result to its neighbor node, receives the local estimation result of the neighbor node, and performs weighted average iteration on the received local estimation result of the neighbor node and its own local estimation result to obtain a consistent filtering result, including: Each sensor node broadcasts the synchronized local estimation results to all its neighboring nodes, and receives the local estimation results of all neighboring nodes at the same time, and performs weighted consistency iteration on the local estimation results of the received neighboring nodes and the local estimation results of the node itself as shown in the following formula: Repeat the above iterative process to obtain the estimated result after L iterations, and set it as: Among them, π i,j represents the weighting coefficient, and N i represents the set of neighbor nodes of the i-th sensor node, i represents the i-th sensor node, and They represent the i-th sensor node at t k The covariance matrix of the estimated value and the estimated error after l+1 weighted consistency iterations at time , and They represent the i-th sensor node at t k The covariance matrix of the estimated value and the estimated error after l weighted consistency iterations at time, and They represent the i-th sensor node at t k The covariance matrix of the estimated value and the estimation error after L weighted consistency iterations at time.

5. An electronic device, comprising at least one processor and at least one memory; characterized in that: At least one instruction is stored in the memory, and the instruction is loaded and executed by the processor to implement the method according to any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method according to any one of claims 1 to 4.

Images