A multi-aircraft clock bias calibration synchronization method, system and apparatus

Abstract

The application discloses a kind of multi-aircraft clock deviation calibration synchronization method, system and device, the method includes: matching measurement information with the state information of target, and carrying out target real-time state estimation;Obtain the motion parameters of cooperative target and form target motion template database;The target state estimation is matched with the target motion template database, and the optimization function of absolute time deviation is constructed using associated target parameter, and the absolute time deviation is solved;The state estimation of different aircrafts to the same target is used, and the relative clock deviation of adjacent nodes is estimated according to the pre-constructed estimation motion model;Absolute time deviation and the relative clock deviation of adjacent nodes are used for clock deviation fusion, and global clock synchronization is completed.The system includes: target state estimation module, absolute time deviation calibration module based on cooperative target, relative time deviation calibration module based on non-cooperative target and airborne real-time clock synchronization module.The device includes memory and processor for executing the above-mentioned multi-aircraft clock deviation calibration synchronization method.By using the application, the precise clock synchronization of multi-aircraft cluster in real complex environment is realized.The application can be widely applied to flight control field.

Description

Technical Field

[0001] This invention relates to the field of flight control, and in particular to a method, system, and apparatus for calibrating and synchronizing clock deviations of multiple aircraft. Background Technology

[0002] Aircraft are well-suited for performing demanding tasks in complex and hazardous environments, thus possessing vast application prospects in both military and civilian fields. As the application areas of aircraft continue to expand, the performance requirements for aircraft are becoming increasingly stringent, making it increasingly difficult for a single aircraft to meet mission demands. The reasons for this are threefold: first, the limited flight range of a single aircraft leads to low mission efficiency; second, the reliability of single-aircraft systems is relatively low, and missions cannot be completed when the aircraft malfunctions; third, for certain high-precision mission scenarios, single aircraft require high-precision components, such as high-precision sensors, which increases manufacturing costs. Therefore, multi-aircraft collaborative systems have emerged. Multiple aircraft can complete tasks quickly and efficiently through task allocation, and by networking, they can form wireless sensor networks. Even multi-aircraft networks equipped with low-precision sensors can achieve high-precision mission requirements, such as multi-aircraft search and target tracking.

[0003] One of the key challenges in wireless sensor networks is time synchronization. Dynamic sensor networks composed of multiple aircraft are widely used for target tracking, target localization, and search missions in unknown environments. In these applications, it is crucial that nodes operate in a coordinated and synchronized manner, requiring global clock synchronization to provide a precise time reference. However, current clock synchronization schemes are insufficient for aircraft performing missions in variable environments, as they do not account for all unknown parameters, namely the skew and offset of all clocks and communication links.

[0004] Currently, aircraft clock calibration mainly relies on timing information provided by the Global Positioning System (GNSS). In severely obstructed outdoor and indoor environments, GNSS cannot function properly and cannot provide stable timing information. This directly results in multiple aircraft networks not having the same global clock, making it impossible to complete collaborative tasks.

[0005] In summary, it is necessary to study highly reliable clock synchronization algorithms to achieve accurate clock synchronization of multi-aircraft clusters in complex real-world environments. Summary of the Invention

[0006] In view of this, in order to solve the technical problem that existing aircraft clock synchronization methods cannot achieve accurate clock synchronization when GNSS signals are limited or denied, this invention proposes a multi-aircraft clock deviation calibration and synchronization method, which includes the following steps:

[0007] The measurement information is matched with the target's state information, and the target's real-time state is estimated.

[0008] Obtain the motion parameters of the cooperative targets and construct a target motion template database;

[0009] The target state estimate is matched with the target motion template database, and an optimization function for the absolute time deviation is constructed using the associated target parameters. The absolute time deviation is then obtained by solving the function.

[0010] By using state estimates of the same target from different aircraft, the relative clock deviation of adjacent nodes is estimated based on a pre-built estimation motion model;

[0011] Global clock synchronization is achieved by weighted fusion of clock deviations based on absolute time deviations and relative clock deviations of adjacent nodes.

[0012] This method can be applied to cases such as multi-aircraft cooperative search and multi-aircraft cooperative detection. In cases where GNSS signals are denied or restricted, and considering issues such as random communication delays in complex environments and differences in sampling frequencies and start measurement times between sensors in the cluster, the aircraft payload estimates the target state of cooperative or non-cooperative targets in real time by measuring. Based on the maximum likelihood estimation method, it comprehensively processes the posterior state estimates of the same cooperative or non-cooperative target to estimate its own clock deviation in real time, thereby achieving clock synchronization of the multi-aircraft cluster. This solves the problem of large clock synchronization errors caused by the lack of GNSS signals in complex environments with random communication delays, and provides an accurate time reference for subsequent multi-sensor information fusion.

[0013] The pre-constructed target motion model is uniform motion, uniform acceleration, etc.

[0014] In some embodiments, the step of matching the measurement information with the target's state information and performing real-time target state estimation specifically includes:

[0015] The aircraft associates and pairs the sensor measurement information of its own body with the target;

[0016] Target state estimation is propagated between different aircraft, and target association and pairing are performed between two neighboring aircraft;

[0017] State estimation of the target is performed based on nonlinear functions.

[0018] In this embodiment, target state estimation involves matching sensor measurement information with target state information and combining extended Kalman filtering (EKF) or other nonlinear filtering methods to estimate the real-time state of the target in order to obtain an unbiased estimate in the sense of minimum variance, thereby ensuring the data quality of the input solution process and improving the accuracy and reliability of clock deviation calculation.

[0019] In some embodiments, the step of matching the target state estimate with the target motion template database specifically includes:

[0020] Construct an estimated trajectory chain based on the target state estimate;

[0021] Based on the multi-attribute weighted association matching model method, the estimated trajectory chain and the trajectory template in the target motion template database are matched.

[0022] In this embodiment, the trajectory chain of the given target state estimate is matched with the target motion template database, and an optimization function for the absolute time deviation is constructed using the associated target parameters. The absolute time deviation is then solved by numerical iteration or numerical calculation.

[0023] In some embodiments, the optimization function for the absolute time deviation is expressed by the following formula:

[0024]

[0025] In the above formula, Let A represent the state estimate of the target, θ represent the known motion parameter matrix, and β represent the time matrix. i This indicates the initial clock deviation.

[0026] In some embodiments, the step of estimating the relative clock offset between adjacent nodes based on a pre-built estimation motion model using state estimation of the same target from different aircraft specifically includes:

[0027] The sliding window coarse detection process is used to process the unpaired target state estimate and estimate the initial relative clock offset;

[0028] Based on the maximum likelihood estimation method, the initial relative clock deviation is processed twice to estimate the relative clock deviation of adjacent nodes.

[0029] In this embodiment, based on the uniqueness and determinism of the target motion state being associated with time, the state estimates of the same target from different aircraft in the cluster are used, and relative time calibration is performed based on the established estimated motion model.

[0030] In some embodiments, the step of fusing clock deviations based on absolute time deviations and relative clock deviations of adjacent nodes to achieve global clock synchronization specifically includes:

[0031] For each state, time deviation is estimated to obtain a time deviation measurement of random noise containing an upper bound of uniform variance;

[0032] The time deviation measurement is estimated based on the maximum likelihood estimation method to obtain the unbiased relative time deviation estimate in the sense of minimum mean square error.

[0033] Based on the unbiased relative time deviation estimate in the sense of minimum mean square error and the weighted fusion of absolute time deviation, clock synchronization is achieved using a distributed linear iterative scheme based on average consistency.

[0034] In this embodiment, the absolute time deviation is calibrated using relative deviation estimation based on motion model estimation. The calibration of local relative clock deviation for non-cooperative targets is combined with the calibration of absolute deviation for cooperative targets. By using the consistency principle and comparing the two, the model error of local relative clock deviation estimation caused by sampling time difference is reduced. Multi-source information fusion improves the accuracy of clock deviation calibration.

[0035] This invention also proposes a multi-aircraft clock deviation calibration and synchronization system, the system comprising:

[0036] The target state estimation module matches the measurement information with the target's state information and performs real-time target state estimation.

[0037] The absolute time deviation calibration module based on cooperative targets acquires the motion parameters of cooperative targets and forms a target motion template database; it matches the target state estimate with the target motion template database, and uses the associated target parameters to construct an optimization function for the absolute time deviation, and solves for the absolute time deviation.

[0038] The relative time deviation calibration module based on non-cooperative targets uses the state estimation of the same target by different aircraft to estimate the relative clock deviation of adjacent nodes according to the pre-built estimation motion model.

[0039] The airborne real-time clock synchronization module fuses clock deviations based on absolute time deviations and relative clock deviations of adjacent nodes to achieve global clock synchronization.

[0040] This invention also proposes a multi-aircraft clock deviation calibration and synchronization device, comprising:

[0041] At least one processor;

[0042] At least one memory for storing at least one program;

[0043] When the at least one program is executed by the at least one processor, the at least one processor implements a multi-aircraft clock deviation calibration synchronization method as described above.

[0044] Based on the above scheme, the present invention provides a multi-aircraft clock deviation calibration synchronization method, system and device. On the one hand, by obtaining target parameters from cooperative targets, a time-state matching template library is established. The aircraft's real-time state estimation of cooperative targets is matched with the template library to achieve its own clock calibration. On the other hand, based on the state estimation of non-cooperative targets, the relative time deviation between node pairs is calculated in real time. Finally, by utilizing the consistency principle and comparing the two, the model error of local relative clock deviation estimation caused by sampling time difference is reduced. Multi-source information fusion improves the accuracy of clock deviation calibration. Attached Figure Description

[0045] Figure 1 This is a flowchart of the steps of a multi-aircraft clock deviation calibration and synchronization method according to the present invention;

[0046] Figure 2 This is a schematic diagram of the data flow of a multi-aircraft clock deviation calibration and synchronization method according to the present invention;

[0047] Figure 3 This is a schematic diagram of the clock calibration and synchronization scheme of the present invention;

[0048] Figure 4 This is a schematic diagram illustrating the association between the observation set and the target set in this invention;

[0049] Figure 5 This is a schematic diagram of the inter-machine target set association of the present invention. Detailed Implementation

[0050] To address the shortcomings mentioned in the background art, this invention provides a clock synchronization scheme for multi-aircraft clusters in complex environments where GNSS signals are restricted or denied and inter-aircraft communication has random time delays. This scheme utilizes state estimation of cooperative and non-cooperative targets and employs maximum likelihood estimation for absolute time calibration and relative deviation calibration. It has the advantages of flexible configuration, strong versatility, and good scalability. Furthermore, this scheme can perform real-time data processing through an onboard data processing unit, increasing the timeliness and security of the solution.

[0051] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0052] It should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0053] It should be understood that the terms "system," "apparatus," "unit," and / or "module" used in this application are a method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0054] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "a," and / or "the" are not specifically singular and may include the plural. Generally, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements. An element defined by the phrase "comprising an..." does not exclude the presence of other identical elements in the process, method, product, or apparatus that includes the element.

[0055] In the description of the embodiments of this application, "a plurality of" refers to two or more. The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0056] Furthermore, flowcharts are used in this application to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, the steps can be processed in reverse order or simultaneously. Additionally, other operations can be added to these processes, or one or more steps can be removed from them.

[0057] Reference Figure 1 , Figure 2 and Figure 3 This is a flowchart illustrating an optional example of the multi-aircraft clock deviation calibration and synchronization method proposed in this invention. This method can be applied to computer equipment, and the imaging method proposed in this embodiment may include, but is not limited to, the following steps:

[0058] Step S1: Match the measurement information with the target's state information and perform real-time target state estimation;

[0059] In step S1, target state estimation needs to be performed, which specifically includes:

[0060] The input information for this step mainly comes from the broadcast information of sensors and neighboring nodes. State estimation includes self-data association, inter-machine data association, and state estimation.

[0061] Self-data association: Since the aircraft is unaware of the correlation between its observations and the target, mismatches between the observation set and the target set can lead to divergence in target state estimation. Therefore, data association becomes essential before state estimation. A schematic diagram of the association between the observation set and the target set is shown below. Figure 4 The self-data association step is the process by which an aircraft correlates the observations from its own sensors with those of a target.

[0062] At the start of the iteration, we assume that the aircraft node C i Having N i Observations The prior state estimates for all targets can be obtained from the previous iteration. The prior state estimate at the k-th iteration is: Therefore, the forms of the prior state estimate and its covariance in the observation space are as follows: and Let the covariance of the nth observation in the spacecraft node be defined as Then, we use Mahalanobis distance to calculate the data associations within itself:

[0063]

[0064] in, Let be the error value for pairing the nth observation with the jth target. This indicates that the i-th observation is paired with the j-th target. Let T represent the nth observation of the i-th spacecraft, and let T denote the matrix transpose. Let be the covariance of the error values ​​between the nth observation and the jth target. We will now use the Hungarian algorithm to match between the observation set and the target set. This is a binary matching algorithm that finds the association between two sets by minimizing the sum of distances on all relevant pairs, and we set a distance threshold α, ignoring matching pairs whose distance exceeds a certain threshold.

[0065] Inter-vehicle data association: When two nodes exchange information about multiple targets via communication, they should be able to associate different targets more reliably than when there is no communication. The inter-vehicle data association step in a multi-vehicle cluster involves associating different targets between two neighboring aircraft. After an aircraft node broadcasts its state estimate, state covariance, and measurement information, we need to associate targets across aircraft. If the aircraft maintain a consistent ranking scheme for the targets, this step is unnecessary. If this ranking information is unavailable, the aircraft can use a cross-aircraft data association scheme to estimate the association between the target set and the observation set. Inter-vehicle target set association refers to... Figure 5 .

[0066] As in the case of data association, Mahalanobis distance is used as the criterion for determining data association:

[0067]

[0068] in, Let be the error value for pairing the j-th target in the i-th aircraft with the j'-th target in the i'-th aircraft. The covariance of the error values ​​for pairing the j-th target in the i-th aircraft with the j'-th target in the i'-th aircraft. We still use the Hungarian algorithm to match target sets between different aircraft. To simplify the representation, we assume that after performing cross-aircraft data association, the information from each neighboring aircraft is sorted such that T i j The information is represented as all data with index j in the i-th aircraft.

[0069] State estimation: Consider a multi-aircraft swarm consisting of N nodes distributed in a certain region of space. At any time k, each node i can obtain the target state vector. Relevant measurement values

[0070] Assume that the measurement and target state vectors satisfy the following state-space model:

[0071]

[0072] in, It is the state of the target; It is the state transition matrix; h(·) is a local measurement value; h(·) is a local nonlinear measurement function. The linearized Jacobi matrix; These are process noise and local measurement noise, respectively. It is usually assumed that measurement noise is uncorrelated between sensor nodes; however, the nonlinear measurement function will differ for different types of sensors.

[0073] The Extended Kalman Filter (EKF) process consists of two parts: time update and measurement update. The time update is the prediction step.

[0074]

[0075]

[0076] Represents the state variable for one-step prediction. This represents the covariance of the state variables in one step of prediction. This represents the variance of process noise.

[0077] Measurement update is the update step:

[0078]

[0079]

[0080]

[0081] This represents the Kalman gain.

[0082] Step S2: Obtain the motion parameters of the cooperative target and form a target motion template database;

[0083] Step S3: Match the target state estimate with the target motion template database, and use the associated target parameters to construct an optimization function for the absolute time deviation, and solve for the absolute time deviation;

[0084] In steps S2 and S3, absolute time deviation calibration based on cooperative targets is performed: absolute time deviation calibration based on cooperative targets involves forming a target motion template database from the motion parameters obtained from the cooperative targets, matching the trajectory chain of the given target state estimate with the target motion template database, constructing an optimization function for absolute time deviation using the associated target parameters, and solving for the absolute time deviation through numerical iteration or numerical calculation methods.

[0085] First, based on the motion parameters sent by the cooperative target, a target motion template library is constructed. Then, using the trajectory chains formed by associating the target with other targets in the state estimation module, a multi-attribute weighted association matching model is employed to perform real-time trajectory and trajectory template matching. A target motion prediction model is then established.

[0086] x(t)=Aθ(t) (9)

[0087] Where A is a known matrix of motion parameters. θ∈R m Given an unknown time matrix, θ(t) = [t m-1 …t 1] T t, T, and m represent the relative motion time, matrix device symbol, and highest order of the motion model, respectively. Then, using the matched target motion model with matching parameters, a real-time absolute time deviation optimization function is constructed:

[0088]

[0089] in, This represents the estimated state of the target. make Equation (10) is equivalent to solving a system of nonlinear equations. The clock skew can be solved using the Newton iteration method.

[0090] Given a discrete-time posterior state sequence X = {x} k The initial clock deviation can be solved using Newton's iterative method for the interval k = 0, 1, 2, ... Clock initial deviation β i Maximum likelihood estimation The WLS solution is

[0091]

[0092] The measurement matrix H = {1}. Here, j, T, and n represent the single deviation measurement sequence number, the matrix transpose symbol, and the total number of deviation measurements, respectively.

[0093] Step S4: Using state estimation of the same target from different aircraft, estimate the relative clock deviation of adjacent nodes based on the pre-built target motion model;

[0094] In step S4, absolute time deviation calibration based on non-cooperative targets is performed: relative time deviation calibration based on non-cooperative targets is performed. Based on the uniqueness and determinism of the correlation between target motion state and time, the relative time calibration is performed using state estimates of the same target from different aircraft in the cluster, according to the established estimated motion model. When aircraft measurement equipment observes targets, state estimation of non-cooperative targets is usually required in the target tracking mission area. Since the motion parameters of these non-cooperative targets cannot be obtained through communication, the relative clock deviation can only be calculated by registering the state estimates of the same target between aircraft. Because the sampling frequency and starting measurement time of sensors between aircraft may differ, a sliding window coarse detection process is used to estimate the relative clock deviation in order to ensure the accuracy of the relative clock estimation, thereby determining the proximity relationship of state estimates between neighboring aircraft.

[0095] First, a sliding window coarse detection process is used to estimate the initial relative clock offset, and then an optimization function for the nearest neighbor estimated sequence number is constructed:

[0096]

[0097] Where d is the window length; k new This is the sequence number of the latest sampling time. Let represent the c-th state estimate of the i-th aircraft towards the target. Considering the differences in sampling frequency and initial sampling time of the inter-vehicle sensors, as well as the uncertainty of the posterior state estimate, we propose a sliding window coarse detection algorithm based on equation (12), adding a coarse correction step for time deviation to determine and Related nearest neighbor valuation,

[0098]

[0099] in, For the estimation of time bias in coarse detection, a j (l r () represents the clock reading of aircraft j during the r-th match between aircraft i and j, where r = 1, 2, ..., n is the sampling sequence within the initial window. When the sampling frequency is sufficiently high, the state estimate... and The corresponding real time difference is a small amount of dt. η ≤1 / 2f si dt η It can be solved by the following formula:

[0100]

[0101] Therefore, the time deviation of the i-th and j-th aircraft for:

[0102]

[0103] Time Deviation Error covariance Therefore, given independently distributed measurements... Solvable time bias estimation Online real-time update of time deviation recursive estimation:

[0104]

[0105] (Q ij (l)) -1 This represents the inverse of the time bias estimation error covariance matrix after the l-th iteration.

[0106] Step S5: Perform clock deviation fusion based on the absolute time deviation and the relative clock deviation of adjacent nodes to complete global clock synchronization.

[0107] In step S5, real-time clock synchronization consistency is performed. The time deviation estimate based on each state estimate is equivalent to a time deviation measurement with random noise containing a uniform variance upper bound obtained from direct observation of time. The relative time deviation measurement can be estimated using the maximum likelihood estimation method to obtain an unbiased estimate in the sense of minimum mean square error. Based on the above unbiased estimate, we use a distributed linear iterative scheme based on average consistency to achieve clock synchronization. This scheme does not involve explicit point-to-point messaging or routing; instead, it propagates information in the network by updating the data of each node with a weighted average of the neighbor data. In each iteration, each node can calculate a local weighted deviation estimate, which eventually converges to the global average. This scheme is robust to unreliable communication links (e.g., due to the mobility of multi-aircraft swarms, sensor signal fading, or power limitations).

[0108] We use a distributed linear iterative method to compute the average value. We model the topology of the cluster network using an undirected graph. Let... Defined as having a vertex set and edge set The graph is an undirected graph where each edge (i,j) is a pair of unordered distinct nodes. At t=0, each node initializes its state to α. i (0) = 0. In each subsequent step, each node updates its state using a linear combination of its own state and the states of its instantaneous neighbors:

[0109]

[0110] Among them, W ij (t) is the α in node i j linear weights, Let j be the set of neighboring nodes of node i at time t, where j represents the set of target neighboring nodes. Let be any neighboring node of , and 'a' represent the clock offset compensation coefficient of the node. Note that the weight matrix W(t) ∈ R n×n It should satisfy the pattern of a communication undirected graph. If Then W ij (t) = 0. We hope that the weight matrix will cause the time deviations of all nodes to converge to the global average:

[0111]

[0112] Where β represents the initial time deviation and n represents the total number of clusters.

[0113] The necessary and sufficient conditions for equation (19) to hold are:

[0114] 1 T W=1 T ,W1=1,ρ(W-11 T / n)<1 (19)

[0115] Where ρ represents the spectral radius of the matrix.

[0116] In a time-invariant communication topology (communication graph constant), the weight matrix that satisfies the constraints (19) and the communication topology constraints can be either the maximum degree matrix or the Metropolis matrix.

[0117] Maximum degree matrix:

[0118]

[0119] Where d represents the connectivity of a node.

[0120] Metropolis matrix:

[0121]

[0122] Where ε(t) represents the edge set. The maximum degree matrix and the Metropolis matrix are both random, symmetric and irreducible matrices, satisfying the constraint (19). When the communication graph is a jointly connected dynamic topology graph, it can still be proven that the global deviation can converge to the global average.

[0123] This invention utilizes the unique real-time absolute position and other motion state information of cooperative or non-cooperative moving targets in the geodetic coordinate system, fully considering the one-to-one correspondence between motion state information and time information. The scheme addresses key technologies in multi-aircraft cluster clock synchronization, including absolute coordinate system calibration, registration of cooperative target state estimation with its real-time absolute position, and time calibration based on the asynchronous states of non-cooperative targets among different aircraft. It conducts research on linearized modeling of target motion state and time, multi-sensor fusion estimation, rapid matching and optimization of single-aircraft target state estimation with motion templates, and clock deviation fusion estimation based on the maximum likelihood estimation method.

[0124] By combining different types of sensors (monocular cameras, lidar, and infrared sensors, etc.), the measurement data of moving targets can be processed using Kalman filtering or its extended filtering techniques to achieve complementarity and full utilization of information from different types of sensors. The fusion technology includes not only Kalman filtering but also its variants, particle filtering, Bayesian probabilistic fusion techniques, etc.

[0125] Unlike traditional consensus-based clock synchronization control algorithms, this algorithm does not require a specific real-time synchronization communication scheme for multiple aircraft. Even in asynchronous communication schemes with delays, it can still effectively estimate relative clock deviations, thereby achieving clock synchronization between clusters based on asynchronous communication.

[0126] This invention combines local relative clock deviation calibration for non-cooperative targets with absolute deviation calibration for cooperative targets. By utilizing the consistency principle and comparing the two, the model error of local relative clock deviation estimation caused by sampling time difference is reduced, and the multi-source information fusion improves the accuracy of clock deviation calibration.

[0127] This solution addresses the problems of complex modeling and lack of consideration for communication delay and synchronization in existing clock synchronization control protocols. It fully utilizes the uniqueness of the target motion state to establish a simple and effective clock deviation estimation optimization model, and realizes rapid calibration of time deviation estimation for multi-aircraft distributed clock synchronization based on the maximum likelihood estimation method.

[0128] In complex combat environments where GNSS signals are restricted or denied, this solution can replace GNSS solutions to provide precise clock synchronization for aircraft within a short period, providing an accurate synchronization time reference for subsequent missions and functions of multi-aircraft clusters.

[0129] This solution offers significant advantages in terms of task requirements and low cost. Its consistency-based distributed architecture is robust to link and node failures, providing high flexibility. Furthermore, its decentralized model greatly reduces the difficulty of maintenance and upgrades.

[0130] A multi-aircraft clock skew calibration and synchronization system includes:

[0131] The target state estimation module matches the measurement information with the target's state information and performs real-time target state estimation.

[0132] The target state estimation module matches sensor measurement information with target state information, and uses extended Kalman filtering (EKF) or other nonlinear filtering methods to estimate the target's real-time state to obtain an unbiased estimate in the sense of minimum variance. This ensures the data quality of the input solution process and improves the accuracy and reliability of clock deviation calculation. It includes data preprocessing, self-data association, inter-aircraft data association, and state estimation. The aircraft's measurement information first undergoes data preprocessing to verify its rationality. Then, self-data association is performed to pair the target and measurement information. The target state estimate is propagated through inter-aircraft communication, and finally, inter-aircraft data association is performed to match the state estimates of the same target from different aircraft.

[0133] The absolute time deviation calibration module based on cooperative targets acquires the motion parameters of cooperative targets and forms a target motion template database; it matches the target state estimate with the target motion template database, and uses the associated target parameters to construct an optimization function for the absolute time deviation, and solves for the absolute time deviation.

[0134] The absolute time deviation calibration module based on cooperative targets assembles a target motion template database from the motion parameters obtained from the cooperative targets, constructs an optimization function for the absolute time deviation, and uses numerical iteration or numerical computation methods to calculate the estimated absolute time deviation at the current moment. Finally, it estimates the absolute time deviation based on the maximum likelihood estimation method. The module's processing flow includes acquiring the motion parameters of the cooperative targets, matching the measured target data with the target template, and calibrating the absolute time deviation based on the maximum likelihood estimation method. Based on the given trajectory chain of the target's state estimate, it matches it with the target motion template database, constructs an optimization function for the absolute time deviation using the associated target parameters, and solves for the absolute time deviation through numerical iteration or numerical computation methods.

[0135] The relative time deviation calibration module based on non-cooperative targets uses the state estimation of the same target by different aircraft to estimate the relative clock deviation of adjacent nodes according to the pre-built estimation motion model.

[0136] The relative time deviation calibration module based on non-cooperative targets, based on the uniqueness and determinism of the correlation between target motion state and time, utilizes the state estimates of the same target from different aircraft in the cluster and performs relative time calibration according to the established estimated motion model. The module's processing flow includes sliding window coarse detection processing, time estimation based on a relative reference, and relative time deviation calibration based on the maximum likelihood estimation method. When aircraft measurement equipment observes a target, state estimation of non-cooperative targets is usually required in the target tracking mission execution area. Since the non-cooperative targets cannot obtain their motion parameters through communication, the relative clock deviation can only be calculated by registering the state estimates of the same target between aircraft. Because the sampling frequency and starting measurement time of sensors between aircraft may differ, a sliding window coarse detection process is used to estimate the relative clock deviation in order to ensure the accuracy of the relative clock estimation and to determine the proximity relationship of the state estimates between neighboring aircraft.

[0137] The airborne real-time clock synchronization module fuses clock deviations based on absolute time deviations and relative clock deviations of adjacent nodes to achieve global clock synchronization.

[0138] The airborne real-time clock synchronization module can directly utilize the output of the absolute time deviation calibration module based on cooperative objectives to achieve distributed cluster clock synchronization. The relative time deviation calibration module based on non-cooperative objectives estimates the relative clock deviations of adjacent nodes, which cannot directly achieve group clock synchronization. However, it can utilize a consensus algorithm to asynchronously average the relative clock deviations of each node to achieve distributed real-time clock synchronization.

[0139] The content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0140] A multi-aircraft clock deviation calibration and synchronization device:

[0141] At least one processor;

[0142] At least one memory for storing at least one program;

[0143] When the at least one program is executed by the at least one processor, the at least one processor implements a multi-aircraft clock deviation calibration synchronization method as described above.

[0144] The content of the above method embodiments is applicable to the device embodiments. The specific functions implemented by the device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0145] A storage medium storing processor-executable instructions, which, when executed by a processor, are used to implement a multi-aircraft clock deviation calibration synchronization method as described above.

[0146] The content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0147] This invention solves the time drift problem in scenarios where GNSS signals are restricted or denied due to the lack of precise clock timing. Simultaneously, it overcomes the impact of existing clock synchronization control protocols on random communication delays, thus improving the synchronization accuracy and efficiency of clock synchronization control protocols. This solution can be applied in many fields, providing a precise time reference for multi-aircraft collaborative missions, enabling synchronized mission execution and spatiotemporal synchronization of multi-source information. In military applications, it can be used for tasks such as airborne early warning, reconnaissance and surveillance, communication relay, target attack, electronic warfare, and intelligence gathering. In civilian applications, the aircraft can be used for meteorological observation, terrain surveying, urban environmental monitoring, artificial rainmaking, forest fire early warning, and aerial photography. This solution can not only be deployed in multi-aircraft clusters to achieve clock synchronization and complete dynamic tasks such as target tracking and mapping, but also play a role in areas where static sensor networks are deployed, such as target positioning systems, surveillance, and intrusion detection, serving as an auxiliary means for network clock synchronization. Finally, the solution is simple and easy to implement, relying solely on node state estimation data to calculate estimated clock deviations, and can be executed asynchronously in a distributed manner, possessing high flexibility. Furthermore, its decentralized model greatly reduces the difficulty of maintenance and upgrades, and will undoubtedly provide a high-precision clock reference for distributed multi-aircraft applications.

[0148] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A method for calibrating and synchronizing clock deviations in multiple aircraft, characterized in that, Includes the following steps: The measurement information is matched with the target's state information, and the target's real-time state is estimated. Obtain the motion parameters of the cooperative targets and construct a target motion template database; The target state estimate is matched with the target motion template database, and an optimization function for the absolute time deviation is constructed using the associated target parameters. The absolute time deviation is then obtained by solving the function. By using state estimations of the same target from different aircraft, the relative clock deviations of adjacent nodes are estimated based on a pre-built target motion model; Global clock synchronization is achieved by weighted fusion of clock deviations based on absolute time deviations and relative clock deviations of adjacent nodes. The optimization function for the absolute time deviation is expressed by the following formula: In the above formula, This represents the state estimate of the target, where A represents the known motion parameter matrix. Represents the time matrix, Indicates the initial clock deviation; The step of estimating the state of the same target using different aircraft and estimating the relative clock deviation of adjacent nodes based on a pre-built target motion model specifically includes: The sliding window coarse detection process is used to process the unpaired target state estimate and estimate the initial relative clock offset; Based on the maximum likelihood estimation method, the initial relative clock deviation is processed twice to estimate the relative clock deviation of adjacent nodes.

2. The multi-aircraft clock deviation calibration and synchronization method according to claim 1, characterized in that, The step of matching the measurement information with the target's state information and performing real-time target state estimation specifically includes: The aircraft associates and pairs the sensor measurement information of its own body with the target; Target state estimation is propagated between different aircraft, and target association and pairing are performed between two neighboring aircraft; State estimation of the target is performed based on nonlinear functions.

3. The multi-aircraft clock deviation calibration and synchronization method according to claim 2, characterized in that, The step of matching the target state estimate with the target motion template database specifically includes: Construct an estimated trajectory chain based on the target state estimate; Based on the multi-attribute weighted association matching model method, the estimated trajectory chain and the trajectory template in the target motion template database are matched.

4. The multi-aircraft clock deviation calibration and synchronization method according to claim 1, characterized in that, The step of performing clock deviation weighted fusion based on absolute time deviation and relative clock deviation of adjacent nodes to complete global clock synchronization specifically includes: For each state, time deviation is estimated to obtain a time deviation measurement of random noise containing an upper bound of uniform variance; The time deviation measurement is estimated based on the maximum likelihood estimation method to obtain the unbiased relative time deviation estimate in the sense of minimum mean square error. Clock synchronization is achieved by weighting and fusing the unbiased relative time deviation estimate in the sense of minimum mean square error with the absolute time deviation, and applying a distributed linear iterative scheme based on average consistency.

5. A multi-aircraft clock deviation calibration and synchronization system, characterized in that, include: The target state estimation module matches the measurement information with the target's state information and performs real-time target state estimation. The absolute time deviation calibration module based on cooperative targets acquires the motion parameters of cooperative targets and forms a target motion template database; it matches the target state estimate with the target motion template database, and uses the associated target parameters to construct an optimization function for the absolute time deviation, and solves for the absolute time deviation. The relative time deviation calibration module based on non-cooperative targets uses the state estimation of the same target by different aircraft to estimate the relative clock deviation of adjacent nodes according to the pre-built estimation motion model. The airborne real-time clock synchronization module fuses clock deviations based on absolute time deviations and relative clock deviations of adjacent nodes to achieve global clock synchronization. The optimization function for the absolute time deviation is expressed by the following formula: In the above formula, This represents the state estimate of the target, where A represents the known motion parameter matrix. Represents the time matrix, Indicates the initial clock deviation; The step of estimating the state of the same target using different aircraft and estimating the relative clock deviation of adjacent nodes based on a pre-built target motion model specifically includes: The sliding window coarse detection process is used to process the unpaired target state estimate and estimate the initial relative clock offset; Based on the maximum likelihood estimation method, the initial relative clock deviation is processed twice to estimate the relative clock deviation of adjacent nodes.

6. A multi-aircraft clock deviation calibration and synchronization device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a multi-aircraft clock deviation calibration synchronization method as described in any one of claims 1-4.

Images