Low-orbit satellite multi-source sensor data fusion processing method and system

Abstract

The invention discloses a low-earth-orbit satellite multi-source sensor data fusion processing method and system, and belongs to the technical field of data processing, and the method comprises the following steps: carrying out the dynamic evaluation of bandwidth efficiency based on the residual resource quantity of a low-earth-orbit satellite according to the sensor characteristic parameters of a multi-source sensor and the individual bandwidth occupancy rate; performing adaptability screening on the original data of the multi-source sensor of the low-orbit satellite based on a task demand, and obtaining effective data and a corresponding to-be-fused sensor based on an adaptability screening result; performing data compression dynamic processing on the effective data based on the bandwidth efficiency dynamic evaluation result and the link bandwidth occupancy rate; and carrying out fusion weight dynamic adjustment based on the adaptability parameters of the to-be-fused sensor and the adaptability screening result, carrying out data fusion based on the adjusted fusion weight of the sensor and effective data, and carrying out self-optimization of the fusion weight based on the fusion effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a low-orbit satellite multi-source sensor data fusion processing method and system. BACKGROUND

[0002] Low-orbit satellite constellations have become the core platform for Earth observation and global information acquisition due to their global coverage, low latency, and high revisit rate. These satellites typically carry multiple sensors, including high-resolution optical cameras, synthetic aperture radars, infrared and spectrometers, which can simultaneously acquire multi-dimensional and multi-scale information of the Earth's surface. However, single data sources have limitations, so data fusion technology is crucial. This technology processes heterogeneous data from different satellites and different sensors to complement and enhance information at the information level, generating more complete, accurate, and reliable analysis results. This technology is driving the decision-making model in fields such as Earth science, environmental monitoring, disaster prevention and mitigation, smart cities, and national security towards real-time, global, and intelligent transformation.

[0003] Existing low-orbit satellite multi-source sensor data fusion processing technology begins with data-level preprocessing, including radiometric correction, geometric precision correction of optical, SAR, infrared, and other different source data, and using high-precision ephemeris and attitude data to align different source data to the same space-time reference through space-time registration, which is the prerequisite for effective fusion. Subsequently, feature-level and decision-level fusion are performed: through artificial intelligence and deep learning algorithms, common features (such as target outlines and change information) are automatically extracted and associated in different data, or multi-source information is integrated based on physical models and statistical methods to form more reliable and complete thematic information products than single data. Finally, on-board edge computing and satellite-ground collaborative processing architectures are becoming increasingly important, with preliminary fusion and screening performed on the satellite side to only transmit key information, greatly improving the timeliness of data processing and reducing the pressure on ground stations to achieve rapid transformation from massive raw data to real-time usable knowledge.

[0004] For example, the Chinese invention patent with publication number CN119538579B discloses a fusion positioning algorithm based on low-orbit satellite Doppler measurement, DME, and VOR, which includes: obtaining the Doppler frequency shift of the low-orbit satellite, the slant range information of the DME, and the azimuth angle data of the VOR, constructing a combined observation equation, and obtaining the corresponding Jacobian matrix; based on the geometric dilution of precision, selecting the station and the satellite through a genetic algorithm to optimize the selection of DME, VOR, and low-orbit satellites to obtain the optimal navigation source combination; constructing a motion model under different flight modes, and using an interactive multiple model Kalman filter to dynamically estimate the state of the aircraft to update the position information in real time.

[0005] For example, the Chinese invention patent with publication number CN119089377B discloses a data fusion method and system based on ERA5 reanalysis data and satellite observation data, which includes: first, converting the height coordinates of the reanalysis data set from pressure values to corresponding altitude values to facilitate subsequent matching and fusion; then, dividing the reanalysis data set after height coordinate conversion into a gridded model according to a predetermined resolution, and calculating and fusing the reanalysis data and observation data parameters according to the space-time grid; for each space-time grid, first calculate the reanalysis data mean and reanalysis data standard deviation of the space-time grid, then use the matched satellite observation data to calculate the observation data mean and observation data standard deviation, and finally determine the fusion data gridded model.

[0006] However, in the process of implementing the technical scheme of the embodiments of the present application, the applicant found that the above-mentioned technology at least has the following technical problems: In the prior art, in the on-orbit target identification process of low-orbit satellites, when resources are abundant, partial sensor data redundancy transmission may occur due to lack of dynamic regulation of bandwidth allocation, interfering with effective feature extraction; when resources are scarce, data may be excessively compressed due to limited bandwidth, causing loss of target key features, resulting in invalid data mixed into the fusion process, and also limiting the calculation depth of dynamic adjustment of fusion weights, making it difficult to optimize weight allocation according to real-time sensor performance, resulting in unbalanced contribution of each sensor data, and low data fusion accuracy. SUMMARY

[0007] To solve the technical problem of the prior art that the multi-source data fusion accuracy may be reduced when the on-orbit target identification process of low-orbit satellites is in a state of dynamic limitation of resource differences, the embodiments of the present application provide a low-orbit satellite multi-source sensor data fusion processing method and system. The technical scheme is as follows: On the one hand, a low-orbit satellite multi-source sensor data fusion processing method is provided, which dynamically evaluates the bandwidth efficiency based on the remaining resource amount of the low-orbit satellite and the sensor characteristic parameters and individual bandwidth occupancy rate of the multi-source sensors; adaptively filters the original data of the low-orbit satellite multi-source sensors based on the task requirements, and obtains effective data and its corresponding to-be-fused sensors based on the adaptive filtering result; dynamically processes the data compression of the effective data based on the bandwidth efficiency dynamic evaluation result and the link bandwidth occupancy rate; dynamically adjusts the fusion weight based on the adaptability parameters and the adaptive filtering result of the to-be-fused sensors, performs data fusion based on the adjusted sensor fusion weight and the effective data, and optimizes the fusion weight based on the fusion effect: In another aspect, a low-orbit satellite multi-source sensor data fusion processing system is provided, comprising: an efficiency evaluation module, a data screening module, a data compression module, and a data fusion module; wherein the efficiency evaluation module is configured to perform dynamic bandwidth efficiency evaluation based on the residual resource amount of the low-orbit satellite and the sensor characteristic parameters and individual bandwidth occupancy rate of the multi-source sensor; the data screening module is configured to perform adaptability screening on the original data of the multi-source sensor of the low-orbit satellite based on the task demand, and obtain effective data and corresponding to-be-fused sensors based on the adaptability screening result; the data compression module is configured to perform dynamic data compression processing on the effective data based on the dynamic bandwidth efficiency evaluation result and the link bandwidth occupancy rate; and the data fusion module is configured to perform dynamic adjustment of fusion weights based on the adaptability parameters and the adaptability screening result of the to-be-fused sensors, perform data fusion based on the adjusted sensor fusion weights and the effective data, and self-optimize the fusion weights based on the fusion effect.

[0008] Advantages The technical scheme provided by the embodiments of the present application has at least the following advantages: 1. The present application integrates the dynamic evaluation of on-board resources and communication states, multi-dimensional data screening oriented to task objectives, differentiated compression strategies based on link adaptation, and fusion weight adjustment and self-optimization mechanisms based on data quality into a cooperative closed-loop processing flow, thereby intelligently guaranteeing the transmission quality and integrity of high-value data under resource-constrained conditions, and ensuring that the data on which the fusion algorithm is based have high task relevance and reliability, thereby realizing the simultaneous optimization and improvement of the accuracy, reliability, and overall resource utilization efficiency of multi-source sensor data fusion in a low-orbit satellite dynamic constraint environment.

[0009] 2. The present application combines the inherent characteristics of the sensor with the dynamic satellite residual resources and real-time bandwidth occupancy rate, and uses a series of preset mapping tables for quantitative matching and coupling calculation, thereby generating personalized, dynamically changing bandwidth occupancy rate thresholds for each sensor based on the system state, and distinguishing between the first and second states according to the thresholds; thereby realizing fine and differentiated perception and identification of the bandwidth occupancy efficiency of the multi-source sensor, and providing accurate and objective decision-making basis for subsequent implementation of differentiated data compression or bandwidth regulation strategies for different types of sensors in resource-constrained conditions.

[0010] 3. By establishing multi-dimensional quantitative matching standards for time, space, and observation angle, and calculating the actual matching degree parameters of each sensor data based on specific task objectives, observation data that simultaneously meets basic quality requirements and is highly relevant to the current task is selected at the source, and the sensors providing these data are identified as units to be fused. This achieves task adaptability purification and refinement of multi-source heterogeneous input data, effectively eliminating interference from invalid or low-relevance information, and establishing a high-quality, highly relevant input set for subsequent data compression and fusion processing, laying the data foundation for improving the accuracy and reliability of the final fusion results.

[0011] 4. This invention establishes a multi-level link bandwidth occupancy threshold linked to the system's operating mode, and adopts differentiated response strategies based on different occupancy ranges. This includes optimizing the transmission mode when resources are plentiful, limiting the bandwidth of non-critical sensors and calculating the overload rate when resources are strained, and then dynamically calculating and constraining the data compression ratio of each sensor by combining sensor priority and overload rate. This enables differentiated and refined compression control of data from sensors of different importance when link resources are scarce. Consequently, it maximizes the overall bandwidth utilization while strictly ensuring the accuracy of high-priority sensor data, ensuring reliable and efficient transmission of key observation information under limited channel conditions, and providing quality-controlled data input for backend fusion.

[0012] 5. This invention comprehensively considers the multi-dimensional fit between sensor data and the current task, its long-term reliability, and its predetermined importance. It quantifies this data into fusion weight adjustment values ​​and coefficients using a mapping mechanism, and dynamically synthesizes the baseline weights to assign appropriate initial weights for each fusion task. Furthermore, by establishing an adaptive feedback adjustment mechanism based on the consistency of fused data and using abnormal fluctuations in weights or data as diagnostic clues, the system can automatically identify and correct weight allocation inaccuracies caused by instantaneous sensor performance fluctuations or model mismatches. This closed-loop process ultimately achieves full-cycle dynamic and fine-grained management of fusion weights from initial allocation and real-time adjustment to continuous optimization, significantly improving the adaptability, robustness, and overall reliability of the output results of the multi-source data fusion system. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 A flowchart of a low-orbit satellite multi-source sensor data fusion processing method provided in this application embodiment; Figure 2 This is a flowchart of the dynamic data compression processing method for the low-orbit satellite multi-source sensor data fusion processing method provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of a low-orbit satellite multi-source sensor data fusion processing system provided in an embodiment of this application. Detailed Implementation

[0015] The technical solution provided in this application will now be described with reference to the accompanying drawings.

[0016] To facilitate understanding of the embodiments of this application, the following description is provided first: In this application, the use of prefixes such as "first" and "second" is solely for the purpose of distinguishing different things belonging to the same category, and does not constrain the order, size, or quantity of things. For example, "first message" and "second message" are simply different messages, and there is no chronological, size, or priority relationship between them.

[0017] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0018] like Figure 1 The diagram shows a flowchart of a low-Earth orbit (LEO) satellite multi-source sensor data fusion processing method provided in this application embodiment. The method includes the following steps: Based on the remaining resources of the LEO satellite, dynamically assess bandwidth efficiency according to the sensor characteristic parameters and individual bandwidth occupancy rates of the multi-source sensors. This dynamic bandwidth efficiency assessment represents a dynamic evaluation of the bandwidth occupancy of each sensor under satellite resource conditions, providing a performance basis for subsequent data compression strategies. Based on mission requirements, perform adaptability screening on the raw data from the LEO satellite multi-source sensors, and obtain valid data and their corresponding sensors to be fused based on the adaptability screening results. This adaptability screening represents the selection of valid data that meets the mission objectives and the corresponding sensors to be fused. The system integrates sensors to establish a high-quality input foundation for subsequent data compression strategies. Based on dynamic bandwidth efficiency assessment results and link bandwidth occupancy, it performs dynamic data compression processing on the effective data. This dynamic data compression processing represents a real-time processing mechanism that dynamically adjusts the data compression strategy to match the current low-Earth orbit satellite communication channel conditions and mission requirements. Based on the adaptability parameters and adaptability screening results of the sensors to be fused, the system dynamically adjusts the fusion weights. Data fusion is then performed based on the adjusted sensor fusion weights and the effective data, and the fusion weights are self-optimized based on the fusion effect. This dynamic adjustment of fusion weights represents dynamically adjusting the fusion weights of the sensors to be fused to optimize the overall quality and reliability of the fused data.

[0019] In this embodiment, when data fusion is performed based on the adjusted sensor fusion weight and effective data, first, the effective data of each sensor to be fused is preprocessed, and through time synchronization calibration, spatial registration and data format standardization, the fusion obstacles caused by data heterogeneity are eliminated, and a consistent foundation is laid for subsequent fusion; taking the dynamically adjusted and optimized fusion weight as the core basis, combined with the type characteristics of effective data, the adaptive fusion algorithm is selected: for numerical monitoring data (such as temperature, humidity), weighted fusion method is adopted, and data contribution is allocated according to the weight proportion of each sensor, the higher the weight of the sensor data, the greater the weight coefficient in the fusion calculation; for characteristic data (such as target contour, spectral characteristics), feature layer fusion algorithm is adopted, and the core features of high-weight sensor data are extracted as the dominant of fusion through weight guidance, and the features of low-weight sensor data are used as supplementary verification; for decision-making data (such as target recognition result), D-S evidence theory algorithm is adopted, and weight is converted into evidence credibility to realize weighted fusion of multi-source decision-making; in the fusion execution process, the observation matching degree comprehensive index and historical accuracy of each sensor data are introduced in real time as dynamic correction factors, if the real-time fluctuation of certain sensor data exceeds the threshold, the weight proportion can be temporarily adjusted to reduce interference; finally, the fusion result is double-checked, on the one hand, the deviation value of the fusion result and the effective data of each sensor is calculated to ensure that the deviation is within the preset reasonable interval, on the other hand, the practicability of the fusion result is verified combined with the task target demand (such as whether the target positioning accuracy and monitoring data error meet the requirements), if it does not meet the requirements, the weight adjustment logic and data preprocessing link are traced back, until the fusion data with accuracy, reliability and task adaptability is output, providing final data support for low-orbit satellite on-orbit target recognition and other core tasks.The application is based on the dynamic evaluation of bandwidth efficiency of satellite residual resource quantity and sensor characteristic parameters and individual bandwidth occupation rate, breaks the limitation of traditional fixed bandwidth allocation mode, can quantize the bandwidth occupation rationality of each sensor under different resource conditions in real time, provides accurate efficiency basis for subsequent compression strategy, effectively avoids the problems of bandwidth waste when resources are abundant and insufficient bandwidth of key sensors when resources are tight, improves the fine scheduling level of on-board resources; the adaptability screening mechanism based on task demand extracts effective data and corresponding to-be-fused sensors by focusing on task targets, eliminates irrelevant redundant information and interference data from the data source, establishes a high-quality data input basis for subsequent processing links, significantly reduces the negative influence of invalid data on fusion results, and reduces the calculation overhead of subsequent compression and fusion processing; combined with the dynamic compression processing of bandwidth efficiency evaluation results and link bandwidth occupation rate, the compression strategy can be flexibly adjusted according to the real-time state of satellite communication channel and task demand, low compression ratio is adopted to ensure data integrity when link resource is abundant, reasonable compression ratio is adopted to balance data quantity and key feature reservation degree when link load is high or resource is tight, the problems of excessive compression feature loss or insufficient compression bandwidth occupation caused by fixed compression strategy are avoided, and the dynamic adaptation of data transmission and channel condition is realized; based on the dynamic adjustment of fusion weight of to-be-fused sensor adaptability parameters and screening results, combined with the weight self-optimization mechanism driven by fusion effect, the weight distribution can be adjusted in real time according to the matching degree and data reliability of each sensor data and task, so that the sensor data with high adaptability and high accuracy occupies higher contribution in fusion, and the weight parameters are continuously optimized through fusion effect feedback, effectively solving the disadvantages of equal treatment of good and bad data in traditional fixed weight fusion, significantly improving the overall quality, reliability and matching degree with task target of fusion data, finally realizing the efficient processing and accurate fusion of multi-source sensor data of low-orbit satellite under complex resource conditions, providing stable and reliable data support for on-orbit target identification and other core tasks.

[0020] Further, the sensor characteristic parameters include data dimension, data transmission amount and sensor priority level; and based on the residual resource amount of the low earth orbit satellite, the step of performing bandwidth efficiency dynamic evaluation according to the sensor characteristic parameters and individual bandwidth occupancy rate of the multi-source sensors comprises: matching the sensor priority level of each sensor with a preset standard occupancy rate mapping table to obtain a corresponding reference bandwidth occupancy rate threshold, the standard occupancy rate mapping table defining the reference bandwidth occupancy rate threshold corresponding to different sensor priority levels; matching the residual resource amount of the low earth orbit satellite with a preset operation mode mapping table to obtain a corresponding system operation mode, the operation mode mapping table defining the system resource scheduling strategy mode corresponding to different residual resource amount intervals, and obtaining a corresponding preset bandwidth adjustment ratio based on the system operation mode; matching the data dimension and data transmission amount of each sensor with a preset bandwidth first adjustment mapping table and a bandwidth second adjustment mapping table respectively to obtain a corresponding bandwidth occupancy rate first adjustment value and a bandwidth occupancy rate second adjustment value, the bandwidth first adjustment mapping table and the bandwidth second adjustment mapping table respectively defining the influence amount of different data dimension and different data transmission amount levels on bandwidth occupancy; performing weighting processing on the coupling processing result of the bandwidth adjustment ratio on the bandwidth occupancy rate first adjustment value and the bandwidth occupancy rate second adjustment value to obtain the bandwidth occupancy rate adjustment value of each sensor; performing sum processing on the corresponding reference bandwidth occupancy rate threshold respectively by using the bandwidth occupancy rate adjustment value of each sensor to obtain the individual occupancy rate threshold of each sensor; and performing bandwidth efficiency dynamic evaluation based on the bandwidth occupancy rate adjustment value and the individual occupancy rate threshold of each sensor.

[0021] In the embodiment, the application determines the reference bandwidth occupancy threshold value by matching the sensor priority level with the standard occupancy ratio mapping table, constructs a priority-oriented bandwidth allocation basic logic, ensures that high-priority sensors, such as sensors undertaking core target identification tasks, obtain reasonable bandwidth resource guarantee, avoids that key tasks are affected due to unbalanced bandwidth allocation, and meets the core task demand from the source of resource allocation; relies on the satellite residual resource quantity to match the operation mode mapping table to determine the system operation mode and the corresponding bandwidth adjustment ratio, deeply binds the bandwidth evaluation with the actual resource state of the satellite, breaks the disadvantages of the traditional fixed bandwidth evaluation and resource disconnection, moderately relaxes the bandwidth limit when the satellite residual resource is sufficient to guarantee the data transmission integrity, and tightens the evaluation standard through the adjustment ratio when the resource is tight, realizes the dynamic adaptation of the bandwidth evaluation and the resource state, and improves the flexibility and pertinence of the on-board resource utilization; the sensor data dimension and data transmission quantity are matched with the exclusive adjustment mapping table respectively to obtain the quantitative adjustment value, realizes the accurate measurement of the data characteristics of the sensors themselves, solves the problem of high priority and light data characteristics in the traditional evaluation, for example, gives targeted adjustment to the sensors with high data dimension and large transmission quantity, avoids that the sensors are not identified due to abnormal bandwidth occupancy caused by their own data properties, and makes the bandwidth evaluation more in line with the actual operation demand of each sensor; the coupling processing result of the two adjustment values is weighted through the bandwidth adjustment ratio, and then the individual occupancy threshold value is obtained by summing the reference threshold value, finally the differential evaluation standard of the reference threshold value and the individualized adjustment is formed, so that each sensor has exclusive quantitative basis for bandwidth occupancy evaluation, which replaces the extensive mode of the traditional unified threshold evaluation, can accurately identify the first type of sensors with individual bandwidth occupancy rate exceeding the threshold value and the second type of sensors complying with the threshold value, provides clear and quantitative performance judgment basis for subsequent bandwidth optimization and data compression strategy; the whole evaluation process combines multiple mapping tables and quantitative calculation, converts abstract factors such as satellite resource state, sensor priority, data characteristics, etc. into specific parameters that can be calculated and compared, upgrades the bandwidth performance evaluation from experience judgment to data-driven, significantly improves the objectivity, accuracy and reliability of the evaluation result, provides solid performance support for the formulation of subsequent data compression strategy, effectively reduces the problems of excessive or insufficient compression caused by inaccurate bandwidth evaluation, and lays an important foundation for the efficient operation of the whole multi-source sensor data processing flow.

[0022] Further, the step of performing dynamic bandwidth performance evaluation based on the individual bandwidth occupancy threshold value and the sum of the bandwidth occupancy adjustment values of each sensor includes: if the individual bandwidth occupancy of any sensor exceeds the individual occupancy threshold value, the sensor is marked as a first type of sensor; if the individual bandwidth occupancy of any sensor does not exceed the individual occupancy threshold value, the sensor is marked as a second type of sensor.

[0023] In the embodiment, the application converts the abstract quantitative indicators calculated by multi-dimensional parameters into specific sensor classification results that can be directly applied, making the originally complex bandwidth performance evaluation land from the data level to the application level, solving the problem of disconnection between quantitative evaluation results and actual processing actions, and enabling bandwidth performance evaluation to be no longer a simple numerical calculation, but a practical judgment that can directly guide subsequent operations; through the clear binary determination standard, the traditional evaluation of vague reasonable and unreasonable qualitative description is replaced, the subjective bias of human judgment is eliminated, and the consistency and reliability of the evaluation results are significantly improved, avoiding the confusion of subsequent processing caused by unclear classification standards; the classification method of marking the sensors exceeding the threshold as the first type and the compliant sensors as the second type provides a clear target orientation for subsequent bandwidth optimization and data processing strategies, enabling the subsequent operation to accurately focus on the first type of sensors with bandwidth occupation anomalies, avoiding resource waste and low efficiency caused by indiscriminate processing of all sensors, for example, bandwidth compression or resource scheduling optimization can be implemented for the first type of sensors, and the second type of sensors is maintained in the existing state to reduce unnecessary processing overhead; the classification result of the application builds an efficient connection bridge for problem positioning and precise policy implementation for the entire data processing flow, which can quickly filter out sensors with bandwidth occupation risks, so that subsequent bandwidth adjustment, data compression and other strategy formulation no longer need to analyze complex evaluation parameters, but can quickly start targeted processing based on the classification result, greatly improving the response efficiency of the flow; the classification step is essentially a purification and simplification of the bandwidth performance evaluation result, which reduces the information processing cost of the subsequent link while retaining the core evaluation value, enabling technical personnel or automatic processing systems to grasp the bandwidth status of each sensor with extremely low cognitive load, providing simple and reliable decision support for the efficient and orderly operation of the entire low-orbit satellite multi-source sensor data fusion processing flow.

[0024] Further, the step of performing adaptability screening on the raw data of the low-orbit satellite multi-source sensor based on the task requirement and obtaining effective data and corresponding to-be-fused sensors based on the adaptability screening result comprises: acquiring preset observation matching degree evaluation parameters, the observation matching degree evaluation parameters comprising a time matching degree threshold, a space matching degree threshold and an observation angle matching degree threshold; determining task target data and corresponding target observation parameters based on the task requirement, the task target data comprising task corresponding time range data (clear task execution start and end time, time interval and other key time constraints, used to match sensor raw data acquisition time), task covered space range data (covering task concerned geographical area boundary, space resolution requirement and other space constraints, corresponding to sensor raw data acquisition geographical position information), and task required core observation object data (i.e. task needs to monitor or obtain data of specific target related information such as target type, target characteristic parameter, used to preliminarily screen whether the raw data contains the observation object concerned by the task), the target observation parameters comprising task time data, task space data and task angle data; performing primary screening on the raw data of the low-orbit satellite multi-source sensor based on the task target data, eliminating data not meeting the task target data range, and obtaining effective task data; calculating actual observation matching degree parameters of each effective task data based on the target observation parameters, the actual observation matching degree parameters comprising time matching degree, space matching degree and observation angle matching degree; comparing the actual observation matching degree parameters of each effective task data with corresponding observation matching degree evaluation parameters, if the actual observation matching degree parameters of any effective task data all meet the corresponding observation matching degree evaluation parameters, marking the effective task data as effective data, otherwise marking the effective task data as invalid data; and marking the sensor corresponding to the effective data as a to-be-fused sensor.

[0025] In the embodiment, the application provides standardized and quantifiable judgment basis for data screening by explicitly observing three core dimensions of matching degree evaluation parameters and presetting threshold values, replaces the fuzzy judgment mode relying on experience in traditional screening, completely solves the problem of no unified standard for whether the data meets the task requirements, makes the screening results have strong objectivity and reproducibility, and effectively avoids missing of effective data or misselection of invalid data caused by subjective judgment differences; the task requirement is taken as the starting point to anchor the target data and observation parameters, a task-oriented screening logic is constructed, the data screening is no longer limited to the performance of the sensor itself, but closely surrounds the actual needs of core tasks such as on-orbit target identification, ensures that the effective data screened is highly consistent with the task target, avoids redundant data irrelevant to the task from entering the subsequent processing flow from the source, and significantly improves the adaptation accuracy of data and task; the double screening mechanism of primary screening and accurate comparison forms an efficient hierarchical data purification path, the primary screening quickly eliminates the original data completely inconsistent with the task target data, greatly reduces the data range of subsequent processing, reduces the consumption of limited computing power and storage resources on the satellite, and the accurate comparison based on the actual observation matching degree parameters further controls the data quality in detail, through all-round checking of time synchronization, spatial coincidence degree and observation angle rationality, ensures the high-quality properties of the final effective data, and provides an excellent input basis for subsequent data fusion; by explicitly marking the effective data and the corresponding to-be-fused sensor, the accurate association of data and sensor is realized, the subsequent data compression, weight adjustment and other operations can be directly focused on the to-be-fused sensor and its data, avoiding invalid processing of non-target sensors, and improving the targeting and efficiency of the entire data processing flow; the screening step eliminates invalid data and corresponding sensors, not only reduces the calculation load of subsequent data transmission and fusion, but also avoids the interference of invalid data on the fusion result: the noise, bias and other problems that may be contained in the invalid data are completely isolated, ensuring the reliability of the fusion result from the data input level, and the explicit to-be-fused sensor also provides a clear object range for subsequent dynamic adjustment of fusion weight, so that the weight resources can be concentrated on sensors with qualified performance and effective data, further optimizing the fusion effect; the entire screening process forms a complete link from task requirements to data output, converts abstract task requirements into specific data screening indicators, realizes deep coupling of task requirements and data processing, and enables the low-orbit satellite multi-source sensor system to flexibly adjust the screening standard according to different tasks, enhances the task adaptability and environmental response ability of the system, and provides solid data guarantee for accurate target identification in complex on-orbit scenarios.

[0026] As Figure 2As shown, the data compression dynamic processing flowchart of the low-orbit satellite multi-source sensor data fusion processing method provided by the embodiment of the application, the step of performing data compression dynamic processing on the effective data based on the bandwidth performance dynamic evaluation result and the link bandwidth occupancy rate includes: matching the system running mode with the preset link bandwidth threshold mapping table to obtain the corresponding link bandwidth threshold parameter, the link bandwidth threshold parameter including a link bandwidth occupancy rate first threshold and a link bandwidth occupancy rate second threshold, and the link bandwidth threshold mapping table defining the link bandwidth threshold parameter corresponding to different system running modes; if the current link bandwidth occupancy rate of the low-orbit satellite does not reach the link bandwidth occupancy rate first threshold, it is determined that the link resource is abundant, the difference between the link bandwidth occupancy rate first threshold and the current link bandwidth occupancy rate of the low-orbit satellite is marked as a link bandwidth margin value, the link bandwidth margin value is matched with the preset transmission mode mapping table, and the data transmission mode of the communication link is adjusted to the target data transmission mode obtained by matching, and the transmission mode mapping table defines different data transmission modes corresponding to different link bandwidth margin value intervals; if the current link bandwidth occupancy rate of the low-orbit satellite reaches the link bandwidth occupancy rate first threshold and does not exceed the link bandwidth occupancy rate second threshold, it is determined that the link resource is in a normal load state, and the current data transmission mode and the data compression ratio are maintained without additional processing; if the current link bandwidth occupancy rate of the low-orbit satellite exceeds the link bandwidth occupancy rate second threshold, it is determined that the link resource is in a tense state, the individual bandwidth occupancy rate of the second type of sensor is coupled with the preset bandwidth margin value to obtain a corresponding bandwidth allocation value, and the bandwidth allocated to the sensor is adjusted to the bandwidth allocation value; the difference between the current link bandwidth occupancy rate of the low-orbit satellite and the link bandwidth occupancy rate second threshold is marked as a link bandwidth overload rate, the data compression ratio of the to-be-fused sensor is dynamically adjusted based on the sensor priority level and the link bandwidth overload rate, and the effective data corresponding to the to-be-fused sensor is compressed based on the adjusted data compression ratio. First, for the adjusted compression ratio of each to-be-fused sensor, an appropriate compression algorithm is selected according to the type characteristics (such as image, spectrum, and structured monitoring data) of the effective data: for image data, an algorithm supporting controllable distortion degree such as JPEG2000 can be used, and for structured data, a lossless or low-loss compression algorithm such as LZ77 / LZ78 can be used to realize accurate matching of the algorithm and the data attribute; during the compression execution process, the algorithm parameters are set based on the adjusted compression ratio, for example, for the effective data of a high-priority sensor, if the adjusted compression ratio is low, the algorithm parameters are set to a high-fidelity mode, which focuses on retaining core information such as target contour and feature parameters, and for the data of a low-priority sensor and a high compression ratio, the compression efficiency and the basic data integrity are considered in parameter setting to avoid loss of key associated information.Meanwhile, the data volume change and distortion index during compression are collected in real time, the compressed data volume is compared with the link bandwidth occupation target, if the expected target is not reached, the algorithm parameters (such as quantization step, coding depth) are fine-tuned without breaking the core constraint of the adjusted compression ratio, if the distortion exceeds the threshold corresponding to the sensor data accuracy requirement, the current compression is terminated immediately and the rationality of the compression ratio is traced back; the integrity check and feature consistency verification are performed on the compressed data to ensure that there is no packet loss and garbled code problem in the compressed data, and the key observation features (such as target coordinates, monitoring numerical fluctuation range) are consistent with the effective data before compression, finally the compressed data meeting the bandwidth requirement and satisfying the subsequent fusion quality requirement is formed, providing reliable support for efficient transmission and accurate fusion of low-orbit satellite multi-source sensor data.

[0027] In the embodiment, the application associates the system operation mode with the link bandwidth threshold mapping table, binds the link bandwidth occupancy rate determination threshold with the satellite resource state depth, avoids the problem that the traditional fixed threshold determination is out of touch with the actual resource scene, provides adaptability for the compression strategy under different resource conditions, and ensures that the determination result is more suitable for the actual operation condition of the satellite; for the scene where the link resource is abundant, the bandwidth margin value is calculated and the transmission mode mapping table is matched to adjust the data transmission mode, breaking the extensive mode of indiscriminate transmission under abundant resources, and the optimal transmission mode (such as high-fidelity transmission or efficient transmission) can be selected according to the margin size, which ensures data integrity while improving transmission efficiency and avoids waste of bandwidth resources; when the link resource is in a normal load state, the processing mode of maintaining the status quo is adopted, which avoids unnecessary power consumption caused by transmission mode and compression ratio adjustment, and also ensures the stability of the data processing process, realizing the optimal efficient operation goal without interference; in the case of link resource shortage, the bandwidth allocation value is first adjusted through the bandwidth coupling processing result of the second type of sensor, the redundant bandwidth of the compliant sensor is accurately released, and resource space is created for critical data transmission, and then the compression ratio is dynamically adjusted based on the link bandwidth overload rate and the sensor priority level, which not only quantifies the compression demand through the overload rate to ensure effective relief of bandwidth pressure, but also ensures that the compression ratio of high-level sensors does not exceed the limit through priority, avoiding the loss of features caused by excessive compression of core data, and solving the problem of damage to key data caused by one-size-fits-all in traditional compression; the entire processing process takes the real-time link state and the pre-bandwidth performance evaluation result as the dual basis, upgrades the data compression from static preset to dynamic response, and can quickly adapt to the real-time fluctuations of link bandwidth, for example, when the link changes from abundant to tight, the compression adjustment can be quickly started, and when the link changes from tight to normal, the excessive intervention is stopped in time, greatly improving the system's ability to adapt to complex link environments; through accurate response to different link states and dynamic optimization of compression ratio, the core features of effective data are maximized while the link bandwidth occupancy is effectively controlled, providing reliable protection for high-quality fusion of subsequent sensor data to be fused, avoiding the problem of decreased fusion accuracy caused by improper compression, while realizing efficient utilization of on-board link resources and dual protection of core task data transmission, significantly improving the reliability and resource adaptation ability of the low-orbit satellite multi-source sensor data processing system.

[0028] Further, the step of dynamically adjusting the data compression ratio of the to-be-fused sensors based on the sensor priority level and the link bandwidth overload rate comprises: obtaining a data precision requirement corresponding to each sensor priority level, determining a data compression ratio threshold of each to-be-fused sensor based on the data precision requirement; matching the link bandwidth overload rate with a preset compression ratio adjustment mapping table to obtain a corresponding compression ratio adjustment coefficient, the compression ratio adjustment mapping table defining the compression ratio adjustment coefficients corresponding to different link bandwidth overload rate intervals; multiplying the current data compression ratio of each to-be-fused sensor by the compression ratio adjustment coefficient to obtain a target data compression ratio of each to-be-fused sensor; for each to-be-fused sensor, judging whether the target data compression ratio exceeds the corresponding data compression ratio threshold; if yes, adjusting the data compression ratio of the sensor to the corresponding data compression ratio threshold; otherwise, adjusting the data compression ratio of the sensor to the corresponding target data compression ratio.

[0029] In the embodiment, the application directly links the sensor priority level with the data precision requirement and explicitly compresses the threshold value, builds a rigid constraint of the precision bottom line determined by the priority, completely breaks the disadvantages of the traditional compression strategy of paying more attention to bandwidth and less attention to precision, ensures that the high-priority sensor always has compression guarantee not lower than the threshold value, avoids the loss of key data features caused by excessive compression, and fundamentally guards the data quality bottom line of the core task; the quantitative adjustment coefficient is obtained by matching the link bandwidth overload rate with the compression ratio adjustment mapping table, the abstract bandwidth pressure is converted into a calculable adjustment basis, the adjustment of the compression ratio is no longer dependent on experience judgment, but is based on real-time bandwidth load to realize precise quantization, for example, high adjustment coefficient corresponding to high overload rate to strengthen the compression strength, low adjustment coefficient corresponding to low overload rate to weaken the intervention, to ensure the pertinence and effectiveness of the bandwidth pressure relief; the application adopts the process of target compression ratio calculated by adjustment coefficient product and threshold value checking, realizes the organic combination of elastic adjustment and rigid constraint: elastic adjustment enables it to quickly adapt to bandwidth load fluctuation, and threshold value checking sets a ceiling for the compression ratio of each sensor, avoiding the target compression ratio breaking the precision bottom line due to high bandwidth overload rate, forming a double protection for core data; the application realizes differentiated compression strategy for different priority levels and different bandwidth load scenarios, provides precision priority exclusive guarantee for high-priority sensors, and gives low-priority sensors flexible adjustment space for bandwidth adaptation, avoiding the dual problems of core data damage or insufficient bandwidth optimization caused by traditional one-size-fits-all compression; the whole adjustment process takes data precision requirement as the core and real-time bandwidth state as the guide, deeply couples sensor task value with link resource state, makes the compression ratio adjustment from passive response to active adaptation, not only effectively relieves the link bandwidth pressure, but also finds the best balance point between bandwidth optimization and data quality, provides key support for the integrity and reliability of subsequent to-be-fused data, reduces the fusion error caused by improper compression ratio, and further improves the operation efficiency and task adaptation ability of the whole low-orbit satellite multi-source sensor data processing system.

[0030] Further, the step of dynamically adjusting the fusion weight based on the screening result of the historical accuracy and adaptability of the to-be-fused sensors comprises: obtaining an observation matching degree parameter of a corresponding sensor based on an actual observation matching degree parameter of effective data, classifying and attributing each effective data to a corresponding to-be-fused sensor according to the source of the effective data, forming an effective data set in each to-be-fused sensor, and directly assigning the actual observation matching degree parameter corresponding to each effective data to the observation matching degree parameter of the to-be-fused sensor corresponding to the effective data, thereby obtaining the observation matching degree parameter of the to-be-fused sensor; obtaining a corresponding fusion weight adjustment value based on the observation matching degree parameter of each to-be-fused sensor; obtaining the historical data accuracy of each to-be-fused sensor, matching the historical data accuracy with a preset weight adjustment coefficient mapping table to obtain a corresponding fusion weight adjustment coefficient, and defining the fusion weight adjustment coefficient corresponding to different historical data accuracy intervals by the weight adjustment coefficient mapping table; obtaining a corresponding reference fusion weight based on the sensor priority level of each to-be-fused sensor; multiplying the corresponding fusion weight adjustment value by the fusion weight adjustment coefficient of each to-be-fused sensor to obtain a final adjustment value of each to-be-fused sensor, and processing the corresponding reference fusion weight by using the final adjustment value of each to-be-fused sensor to obtain the fusion weight of each to-be-fused sensor.

[0031] In the embodiment, the application obtains sensor observation matching degree parameters and derives fusion weight adjustment values based on the actual observation matching degree parameters of effective data, deeply binds weight adjustment to the task adaptability of current sensor data, solves the problem of traditional fixed weight deviating from real-time data quality, for example, giving higher adjustment values to sensors with higher time, space and angle matching degrees, ensuring that weight distribution is tilted towards data with stronger task adaptability, and improving the degree of fit between weight and real-time task requirements; the accuracy of sensor historical data is introduced and converted into a quantitative adjustment coefficient through a weight adjustment coefficient mapping table, the past performance of the sensor is included in the weight calculation system, avoiding the one-sidedness of fixing the weight based on single data performance: even if the historical accuracy of a sensor is high and the single data matching degree is slightly low, the sensor can still obtain a relatively reasonable weight through the adjustment coefficient, while the historical performance of a sensor is poor and the weight is moderately reduced, ensuring the reliability of weight distribution from the long-term performance dimension; the reference fusion weight is established based on the sensor priority level, which builds the core bottom line of weight distribution, ensures that high-priority sensors that bear key tasks can still maintain the advantage of basic weight in multi-factor adjustment, and will not lose the weight proportion due to real-time matching degree fluctuations or slight deviations in historical accuracy, effectively ensuring the dominant position of core data in fusion; the final adjustment value is obtained by multiplying the adjustment coefficient and the adjustment value, and the reference weight is corrected, which converts abstract sensor performance indicators into quantifiable weight parameters, replacing the traditional experience-dependent weight setting method, so that the weight of each sensor has a clear calculation basis, realizing differentiated and personalized weight configuration, and avoiding the problem of insufficient contribution of high-quality data and excessive interference of poor-quality data caused by one-size-fits-all weight; the entire adjustment process is quantitatively coupled with threshold constraints through multi-dimensional parameters, so that the fusion weight can dynamically respond to the comprehensive changes of sensor real-time performance, historical performance and task priority, and when the sensor data adaptability improves and the historical accuracy optimizes, the weight is reasonably improved, and vice versa, so that the fusion weight is always in a dynamic balance state that matches the actual performance of the sensor; accurate weight distribution directly optimizes the subsequent data fusion effect, so that high-adaptability, high-reliability and high-priority sensor data play a dominant role in fusion, maximally reducing the interference of low-quality data, significantly improving the overall quality and reliability of the fused data, providing more accurate data support for low-orbit satellite on-orbit target identification and other core tasks, and improving the intelligentization and self-adaptation ability of the entire multi-source sensor data processing system.

[0032] Further, the step of obtaining the corresponding fusion weight adjustment value based on the observation matching degree parameter of each to-be-fused sensor includes: obtaining a preset observation matching degree weight parameter, the observation matching degree weight parameter including a time matching degree weight, a space matching degree weight and an observation angle matching degree weight; respectively performing weighting coupling processing on the observation matching degree parameter of each to-be-fused sensor by using the observation matching degree weight parameter, to obtain an observation matching degree comprehensive index of each to-be-fused sensor, the observation matching degree comprehensive index being used to quantitatively evaluate a comprehensive evaluation value of the overall matching degree between the sensor data and the current task; and respectively matching the observation matching degree comprehensive index of each to-be-fused sensor with a preset weight adjustment value mapping table to obtain the corresponding fusion weight adjustment value, the weight adjustment value mapping table defining the fusion weight adjustment value corresponding to different observation matching degree comprehensive index intervals.

[0033] The observation matching degree comprehensive index of each to-be-fused sensor is obtained in the following manner: ; In the formula, represents the observation matching degree comprehensive index of the i-th to-be-fused sensor, 、 and respectively represent the time matching degree weight, the space matching degree weight and the observation angle matching degree weight, 、 and respectively represent the time matching degree, the space matching degree and the observation angle matching degree of the i-th to-be-fused sensor, wherein i is the number of each to-be-fused sensor, i = 1, 2, 3,..., N, and N is the total number of to-be-fused sensors.

[0034] In the embodiment, the present application disassembles the observation matching degree into three core dimensions of time, space and observation angle, each dimension is provided with a preset basic weight, including a time matching degree basic weight, a space matching degree basic weight and an observation angle matching degree basic weight. The generation core of the observation matching degree weight parameter in the present application is to flexibly adjust the above-mentioned preset basic weight based on the on-orbit task characteristics, so as to form the observation matching degree weight parameter which deeply matches the task demand. The specific adjustment logic is: taking the task core target as the guide, first identifying the matching dimension which plays a key role in achieving the task, and then realizing accurate adaptation by means of improving the preset basic weight proportion of the dimension and optimizing the weight distribution of other dimensions, and different tasks correspond to clear adjustment process. For example, the target tracking task takes guaranteeing the data time sequence continuity as the core demand, will dynamically process the time matching degree basic weight proportion based on the preset time sequence continuity adjustment proportion, and then normalize the time matching degree basic weight after processing and other basic weights, finally obtains the observation matching degree weight parameter to ensure the time sequence consistency; the regional monitoring task takes guaranteeing the geographical range coverage precision as the core demand, will normalize the space matching degree basic weight and other basic weights based on the preset precision adjustment proportion, and obtains the observation matching degree weight parameter to improve the geographical positioning accuracy. This differentiated weight adjustment mode based on the task characteristics solves the problem that the traditional one-size-fits-all evaluation mode cannot adapt to diversified on-orbit tasks, makes the observation matching degree evaluation logic deeply bind with different task core demands, and significantly improves the pertinence and effectiveness of the evaluation.The single-dimensional indicators are integrated into a unified observation matching degree comprehensive index by weighting and coupling the dimensional matching degree parameters through the weight parameters, the quantitative evaluation of the sensor data and task matching degree is realized, the qualitative judgment of whether the time is synchronized or the space is coincided is converted into a calculable numerical index, the deviation of relying on subjective experience in the traditional evaluation is completely avoided, the matching degree evaluation has strong objectivity and comparability; the construction of the comprehensive index effectively balances the influence of each dimension matching degree, avoids the misjudgment of the sensor with excellent single-dimensional performance but low overall matching degree, for example, when a sensor has high time matching degree but large spatial deviation, the comprehensive calculation will get a lower index, ensuring that the evaluation result can reflect the overall adaptation level of the sensor data; the comprehensive index is converted into the corresponding fusion weight adjustment value by means of the weight adjustment value mapping table, the seamless connection of the evaluation result and the weight adjustment is realized, the abstract comprehensive index is converted into specific parameters that can be directly used for weight calculation, providing accurate and explicit quantitative basis for the dynamic adjustment of the subsequent fusion weight; the whole process is standardized by the weight preset and mapping rules, ensuring that the weight adjustment value evaluation of different sensors is under the unified standard, avoiding the imbalance of weight distribution caused by different evaluation scales, at the same time, the quantitative evaluation method makes the weight adjustment value dynamically updated with the real-time change of the sensor observation matching degree, making the weight adjustment always follow the fluctuation of data adaptation, laying a solid foundation for the subsequent high-quality data fusion, finally realizing the precise guidance that the higher the data matching degree, the larger the weight adjustment value, making the fusion resources tilt to the sensor that best matches the task, significantly improving the pertinence and reliability of data fusion.

[0035] Further, the step of self-optimizing the fusion weight based on the fusion effect comprises: calculating average deviation values between each to-be-fused sensor data and the fused data; matching the average deviation values with a preset deviation threshold value, if the average deviation values exceed the deviation threshold value, monitoring fluctuation conditions of the fusion weights of the sensors, if the fluctuation frequency of the fusion weight of any sensor exceeds a preset first fluctuation frequency threshold value, marking a difference value between the weight fluctuation frequency and the first fluctuation frequency threshold value as a first deviation fluctuation frequency, matching the first deviation fluctuation frequency with a preset first adjustment proportion mapping table to obtain a corresponding adjustment coefficient adjustment proportion, and performing product processing on the fusion weight adjustment coefficient of the sensor based on the adjustment coefficient adjustment proportion, the first adjustment proportion mapping table defining adjustment coefficient adjustment proportions corresponding to different first deviation fluctuation frequency intervals; if the fluctuation frequencies of the fusion weights of the sensors all do not exceed the preset fluctuation frequency threshold value, monitoring fluctuation conditions of the sensor data, if the fluctuation frequency of any sensor data exceeds a preset second fluctuation frequency threshold value, marking a difference value between the data fluctuation frequency and the second fluctuation frequency threshold value as a second deviation fluctuation frequency, matching the second deviation fluctuation frequency with a preset second adjustment proportion mapping table to obtain a corresponding adjustment value adjustment proportion, and performing product processing on the fusion weight adjustment value of the sensor based on the adjustment value adjustment proportion, the second adjustment proportion mapping table defining adjustment value adjustment proportions corresponding to different second deviation fluctuation frequency intervals, otherwise issuing a data fusion abnormality prompt to notify preset personnel to check the sensor equipment; if the average deviation values do not exceed the deviation threshold value, no additional processing is performed.

[0036] In the embodiment, the application takes the average deviation value of each sensor data and fused data as the core quantitative index of fusion effect, converts the abstract fusion quality into calculable and comparable numerical basis, replaces the traditional extensive mode of relying on artificial judgment of fusion effect, makes the fusion effect evaluation have strong objectivity and precision, provides clear starting trigger condition for weight optimization, avoids the problem of no basis optimization or missing judgment of poor fusion; the application establishes a layered troubleshooting logic of deviation threshold verification and fluctuation tracing, when the average deviation value exceeds the threshold, first monitors the weight fluctuation and then troubleshoots the data fluctuation, accurately locates the root cause of poor fusion effect, whether it is caused by dynamic imbalance of weight distribution or insufficient data stability of the sensor itself, solves the drawbacks of blind adjustment of weight in traditional optimization, ensures that the optimization measures hit the problem core; for the case that the weight fluctuation exceeds the standard, the first deviation fluctuation frequency is calculated and matched with the first adjustment proportion mapping table, the weight fluctuation degree is converted into a quantitative adjustment coefficient proportion, the fusion weight adjustment coefficient is accurately corrected, which effectively suppresses the instability caused by excessive weight fluctuation, avoids the rigid imbalance of weight caused by one-size-fits-all correction, and makes the weight adjustment more suitable for the actual situation of fluctuation; when the data fluctuation exceeds the threshold, the second adjustment proportion mapping table is matched with the second deviation fluctuation frequency to obtain the adjustment value proportion, the fusion weight adjustment value is optimized, the indirect avoidance of the interference of unstable data on the fusion result is realized through weight adjustment, and the effect of hedging data fluctuation through weight optimization is realized; in the scene where the weight and data fluctuation are both compliant but the deviation exceeds the standard, an abnormal prompt is given, the organic combination of system self-processing and manual intervention is realized, the continuous deterioration of fusion anomaly caused by hidden problems such as sensor hardware failure is avoided, and the fault response ability and reliability of the system are improved; when the deviation value is compliant, the present situation is maintained, unnecessary weight adjustment is avoided, the on-demand optimization energy-saving and efficient goal is realized; the whole self-optimization process upgrades the fusion weight from static configuration to dynamic self-adaptation, can continuously respond to the changes of fusion effect and sensor state, continuously optimizes the weight distribution scheme, significantly improves the stability, accuracy and reliability of fused data, provides adaptive data quality guarantee for low-orbit satellite on-orbit target recognition and other core tasks, and further strengthens the intelligent level and environmental adaptability of the whole multi-source sensor data processing system.

[0037] As Figure 3As shown, the structural schematic diagram of the low-orbit satellite multi-source sensor data fusion processing system provided by the embodiment of the application includes an efficiency evaluation module, a data screening module, a data compression module, and a data fusion module. The efficiency evaluation module is configured to perform dynamic bandwidth efficiency evaluation on the basis of the residual resource amount of the low-orbit satellite and the sensor characteristic parameters and individual bandwidth occupancy rate of the multi-source sensor, the dynamic bandwidth efficiency evaluation representing the dynamic evaluation of the bandwidth occupancy of each sensor under the satellite resource condition, and providing the efficiency basis for the subsequent data compression strategy. The data screening module is configured to perform adaptability screening on the original data of the low-orbit satellite multi-source sensor on the basis of the task demand, and obtain the effective data and the corresponding to-be-fused sensor on the basis of the adaptability screening result, the adaptability screening representing the screening of the effective data and the corresponding to-be-fused sensor meeting the task target, and establishing the high-quality input basis for the subsequent data compression strategy. The data compression module is configured to perform dynamic data compression processing on the effective data on the basis of the dynamic bandwidth efficiency evaluation result and the link bandwidth occupancy rate, the dynamic data compression processing representing the dynamic adjustment of the data compression strategy to match the real-time processing mechanism of the current low-orbit satellite communication channel condition and the task demand. The data fusion module is configured to perform dynamic adjustment of the fusion weight on the basis of the adaptability parameters and the adaptability screening result of the to-be-fused sensor, perform data fusion on the basis of the adjusted sensor fusion weight and the effective data, and self-optimize the fusion weight on the basis of the fusion effect, the dynamic adjustment of the fusion weight representing the dynamic adjustment of the fusion weight of the to-be-fused sensor to optimize the overall quality and reliability of the fused data.

[0038] The various features and processes described above can be used independently of one another or can be combined in various ways. All possible combinations and subcombinations are intended to fall within the scope of the present disclosure. In addition, some of the method or process blocks described herein can be omitted in some implementations. The methods and processes described herein are also not limited to any particular order or sequence, and the blocks or states relating thereto can be performed in other suitable orders or sequences. For example, blocks or states described in succession can be performed at least partially at the same time or can be performed in reverse order, depending upon the particular needs of the implementation. Examples of blocks or states can be performed in serial, in parallel, or in some other manner. Blocks or states can be added to or removed from the disclosed example embodiments. The example systems and components described herein can be configured differently than described. For example, elements can be added to, removed from, or rearranged in the disclosed example embodiments.

[0039] The various operations of example methods described herein can be performed, at least partially, by an algorithm. The algorithm can be embodied in program code or instructions stored in a memory (e.g., the non-transitory computer-readable storage medium described above). Such an algorithm can include a machine learning algorithm. In some embodiments, a machine learning algorithm can not explicitly program a computer to perform a function, but can learn from training data to make a predictive model that performs the function.

[0040] The various operations of example methods described herein can be performed, at least partially, by one or more processors that are temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors can constitute processor- implemented engines that operate to perform one or more operations or functions described herein.

[0041] Similarly, the methods described herein can be at least partially processor- implemented, with a particular processor or processors being an example of hardware. For example, at least some of the operations of a method can be performed by one or more processors or processor-implemented engines. Moreover, a processor or processors can also operate to support performance of the relevant operations in a "cloud computing" environment or as a "software as a service" (SaaS). For example, at least some of the operations can be performed by a group of computers (as examples of machines including processors), with these operations being accessible via a network (e.g., the Internet) and via one or more appropriate interfaces (e.g., an application program interface (API)).

[0042] The performance of certain of the operations can be distributed among the processors, not only residing within a single machine, but deployed across a number of machines. In some example embodiments, the processors or processor-implemented engines can be located in a single geographic location (e.g., within a home environment, an office environment, or a server farm). In other example embodiments, the processors or processor-implemented engines can be distributed across a number of geographic locations.

[0043] In this specification, a plurality of instances can implement a component, operation, or structure described as a single instance. Although individual operations of one or more methods are illustrated and described as separate operations, one or more of the separate operations can be performed concurrently, and nothing requires that the operations be performed in the order illustrated. Structures and functionality presented as separate components in example configurations can be implemented as a combined structure or component. Similarly, structures and functionality presented as a single component can be implemented as separate components. These and other variations, modifications, additions, and improvements fall within the scope of the subject matter herein.

[0044] While the overview of the subject matter has been described with reference to particular examples embodiments, various modifications and changes can be made to the embodiments without departing from the broader scope of the embodiments of the present disclosure. Such embodiments of the inventive subject matter can be referred to herein individually or collectively as the “application” for convenience and are not intended to limit the scope of the application to a single disclosure or otherwise, unless the context clearly indicates otherwise, with the entirety of each individual embodiment being within the scope of the present disclosure.

[0045] The embodiments illustrated herein are described in sufficient detail to enable those skilled in the art to practice the disclosed teachings. Other embodiments can be utilized and derived therefrom, without departing from the scope of the disclosure, such that structural and logical substitutions and changes can be made without departing from the scope of the present disclosure. Accordingly, the detailed description is to be regarded as illustrative only and not limiting, with the scope of various embodiments being indicated by the appended claims along with the full range of equivalents to which such claims are entitled.

Claims

1. A method for processing multi-source sensor data fusion of low earth orbit satellites, characterized in that, The method comprises the following steps: performing bandwidth performance dynamic evaluation on the sensor characteristic parameters and individual bandwidth occupancy of the multi-source sensors based on the residual resource amount of the low-orbit satellite, wherein the bandwidth performance dynamic evaluation represents the bandwidth occupancy of each sensor under the dynamic evaluation of the satellite resource conditions, and provides performance basis for the subsequent data compression strategy; performing adaptability screening on the original data of the multi-source sensors of the low-orbit satellite based on the task demand, and obtaining effective data and corresponding to-be-fused sensors based on the adaptability screening result, wherein the adaptability screening represents screening of the effective data and corresponding to-be-fused sensors that meet the task target, and establishes a high-quality input basis for the subsequent data compression strategy; performing data compression dynamic processing on the effective data based on the bandwidth performance dynamic evaluation result and the link bandwidth occupancy, wherein the data compression dynamic processing represents a real-time processing mechanism of dynamically adjusting the data compression strategy to match the current low-orbit satellite communication channel conditions and the task demand; performing fusion weight dynamic adjustment based on the adaptability parameters and the adaptability screening result of the to-be-fused sensors, performing data fusion based on the adjusted sensor fusion weight and the effective data, and self-optimizing the fusion weight based on the fusion effect, wherein the fusion weight dynamic adjustment represents dynamically adjusting the fusion weight of the to-be-fused sensors to optimize the overall quality and reliability of the fused data.

2. The method of claim 1, wherein: The sensor characteristic parameters comprise data dimension, data transmission amount and sensor priority level. The step of performing bandwidth performance dynamic evaluation on the sensor characteristic parameters and individual bandwidth occupancy of the multi-source sensors based on the residual resource amount of the low-orbit satellite comprises the following steps: matching the sensor priority level of each sensor with a preset standard occupancy rate mapping table to obtain a corresponding reference bandwidth occupancy rate threshold value; matching the residual resource amount of the low-orbit satellite with a preset operation mode mapping table to obtain a corresponding system operation mode, and obtaining a corresponding preset bandwidth adjustment proportion based on the system operation mode; matching the data dimension and data transmission amount of each sensor with a preset bandwidth first adjustment mapping table and a bandwidth second adjustment mapping table respectively to obtain a corresponding bandwidth occupancy rate first adjustment value and a bandwidth occupancy rate second adjustment value; performing weighting processing on the coupling processing result of the bandwidth adjustment proportion on the bandwidth occupancy rate first adjustment value and the bandwidth occupancy rate second adjustment value to obtain the bandwidth occupancy rate adjustment value of each sensor; processing the corresponding reference bandwidth occupancy rate threshold value of each sensor by using the bandwidth occupancy rate adjustment value of each sensor to obtain the individual occupancy rate threshold value of each sensor; performing bandwidth performance dynamic evaluation based on the bandwidth occupancy rate adjustment value of each sensor and the individual occupancy rate threshold value.

3. The method of claim 2, wherein: The step of performing bandwidth performance dynamic evaluation based on the bandwidth occupancy rate adjustment value of each sensor and the individual occupancy rate threshold value comprises the following steps: if the individual bandwidth occupancy of any sensor exceeds the individual occupancy rate threshold value, marking it as a first type sensor; if the individual bandwidth occupancy of any sensor does not exceed the individual occupancy rate threshold value, marking it as a second type sensor.

4. The method of claim 1, wherein: The step of adaptively screening the raw data of the low-orbit satellite multi-source sensor based on the task requirement and obtaining effective data and corresponding to-be-fused sensors based on the adaptively screened results comprises: acquiring preset observation matching degree evaluation parameters, the observation matching degree evaluation parameters comprising a time matching degree threshold, a space matching degree threshold and an observation angle matching degree threshold; determining task target data and corresponding target observation parameters based on the task requirement, the target observation parameters comprising task time data, task space data and task angle data; performing primary screening on the raw data of the low-orbit satellite multi-source sensor based on the task target data to obtain effective task data; calculating actual observation matching degree parameters of each effective task data based on the target observation parameters, the actual observation matching degree parameters comprising a time matching degree, a space matching degree and an observation angle matching degree; comparing the actual observation matching degree parameters of each effective task data with corresponding observation matching degree evaluation parameters, if the actual observation matching degree parameters of any effective task data all reach the corresponding observation matching degree evaluation parameters, marking the effective task data as effective data, otherwise marking the effective task data as invalid data; marking the sensors corresponding to the effective data as to-be-fused sensors.

5. The method of claim 4, wherein: The step of dynamically processing the effective data based on the bandwidth performance dynamic evaluation results and the link bandwidth occupancy rate comprises: matching the system running mode with a preset link bandwidth threshold mapping table to obtain corresponding link bandwidth threshold parameters, the link bandwidth threshold parameters comprising a link bandwidth occupancy rate first threshold and a link bandwidth occupancy rate second threshold; if the current link bandwidth occupancy rate of the low-orbit satellite does not reach the link bandwidth occupancy rate first threshold, marking the difference between the link bandwidth occupancy rate first threshold and the current link bandwidth occupancy rate of the low-orbit satellite as a link bandwidth margin value, matching the link bandwidth margin value with a preset transmission mode mapping table, and adjusting the data transmission mode of the communication link to a target data transmission mode obtained by the matching; if the current link bandwidth occupancy rate of the low-orbit satellite reaches the link bandwidth occupancy rate first threshold and does not exceed the link bandwidth occupancy rate second threshold, no additional processing is performed; if the current link bandwidth occupancy rate of the low-orbit satellite exceeds the link bandwidth occupancy rate second threshold, marking the coupling processing result of the individual bandwidth occupancy rate of the second type of sensor and a preset bandwidth margin value as a corresponding bandwidth allocation value, and adjusting the bandwidth allocated to the sensor to the bandwidth allocation value; marking the difference between the current link bandwidth occupancy rate of the low-orbit satellite and the link bandwidth occupancy rate second threshold as a link bandwidth overload rate, dynamically adjusting the data compression ratio of the to-be-fused sensors based on the sensor priority level and the link bandwidth overload rate, and performing compression processing on the effective data corresponding to the to-be-fused sensors based on the adjusted data compression ratio.

6. The low earth orbit satellite multi-source sensor data fusion processing method of claim 5, wherein: The step of dynamically adjusting the data compression ratio of the to-be-fused sensors based on the sensor priority level and the link bandwidth overload rate comprises: acquiring data precision requirements corresponding to each sensor priority level, and determining data compression ratio thresholds of each to-be-fused sensor based on the data precision requirements; The link bandwidth overload rate is matched with a preset compression ratio adjustment mapping table to obtain a corresponding compression ratio adjustment coefficient; The current data compression ratio of each sensor to be fused is processed by using the compression ratio adjustment coefficient to obtain a target data compression ratio of each sensor to be fused; For each sensor to be fused, it is judged whether the target data compression ratio exceeds a corresponding data compression ratio threshold value; If yes, the data compression ratio of the sensor is adjusted to the corresponding data compression ratio threshold value; Otherwise, the data compression ratio of the sensor is adjusted to the corresponding target data compression ratio.

7. The method of claim 1, wherein: The step of dynamically adjusting the fusion weight based on the historical accuracy and adaptability screening result of the sensor to be fused comprises: An observation matching degree parameter of a corresponding sensor is obtained based on an actual observation matching degree parameter of effective data; A fusion weight adjustment value of the corresponding sensor is evaluated based on the observation matching degree parameter of each sensor to be fused; A historical data accuracy rate of each sensor to be fused is obtained, and the historical data accuracy rate is matched with a preset weight adjustment coefficient mapping table to obtain a corresponding fusion weight adjustment coefficient; A reference fusion weight of the corresponding sensor is obtained based on a sensor priority level of each sensor to be fused; The fusion weight adjustment value of the corresponding sensor is processed by using the fusion weight adjustment coefficient of each sensor to be fused to obtain a final adjustment value of each sensor to be fused, and the reference fusion weight of the corresponding sensor is processed by using the final adjustment value of each sensor to be fused to obtain a fusion weight of each sensor to be fused.

8. The method of claim 7, wherein: The step of evaluating the fusion weight adjustment value of the corresponding sensor based on the observation matching degree parameter of the sensor to be fused comprises: A preset observation matching degree weight parameter is obtained, and the observation matching degree weight parameter comprises a time matching degree weight, a space matching degree weight and an observation angle matching degree weight; The observation matching degree parameter of each sensor to be fused is weighted and coupled by using the observation matching degree weight parameter respectively to obtain an observation matching degree comprehensive index of each sensor to be fused, and the observation matching degree comprehensive index is used to quantitatively evaluate a comprehensive evaluation value of the overall fitness of the sensor data and the current task; The observation matching degree comprehensive index of each sensor to be fused is matched with a preset weight adjustment value mapping table respectively to obtain a corresponding fusion weight adjustment value.

9. The method of claim 1, wherein: The step of self-optimizing the fusion weight based on the fusion effect comprises: An average deviation value between the data of each sensor to be fused and the fused data is calculated; The average deviation value is matched with a preset deviation threshold value, if the average deviation value exceeds the deviation threshold value, the fluctuation of the fusion weight of each sensor is monitored, if the fluctuation frequency of the fusion weight of any sensor exceeds a preset first fluctuation frequency threshold value, the difference between the weight fluctuation frequency and the first fluctuation frequency threshold value is marked as a first deviation fluctuation frequency, the first deviation fluctuation frequency is matched with a preset first adjustment ratio mapping table to obtain a corresponding adjustment coefficient adjustment ratio, and the fusion weight adjustment coefficient of the sensor is processed based on the adjustment coefficient adjustment ratio. If the fluctuation frequency of each sensor fusion weight does not exceed the preset fluctuation frequency threshold, the fluctuation of each sensor data is monitored, if the fluctuation frequency of any sensor data exceeds the preset second fluctuation frequency threshold, the difference between the data fluctuation frequency and the second fluctuation frequency threshold is marked as a second deviation fluctuation frequency, the second deviation fluctuation frequency is matched with the preset second adjustment proportion mapping table to obtain a corresponding adjustment value adjustment proportion, the fusion weight adjustment value of the sensor is processed based on the adjustment value adjustment proportion, otherwise an abnormal data fusion prompt is issued; If the average deviation value does not exceed the deviation threshold, no additional processing is performed.

10. A low earth orbit satellite multi-source sensor data fusion processing system, applying the low earth orbit satellite multi-source sensor data fusion processing method according to any one of claims 1-9, characterized in that, Comprise: Efficiency evaluation module, data screening module, data compression module, data fusion module; Wherein, the efficiency evaluation module is used for dynamic evaluation of bandwidth efficiency based on the residual resource amount of low-orbit satellite, according to the sensor characteristic parameters and individual bandwidth occupation rate of multi-source sensor, the dynamic evaluation of bandwidth efficiency represents the bandwidth occupation of each sensor under the condition of dynamic evaluation of satellite resources, which provides the efficiency basis for subsequent data compression strategy; The data screening module is used for adaptive screening of the original data of low-orbit satellite multi-source sensor based on task demand, and effective data and its corresponding to-be-fused sensor are obtained based on the adaptive screening result, the adaptive screening represents screening of effective data and corresponding to-be-fused sensor meeting task target, which establishes high-quality input basis for subsequent data compression strategy; The data compression module is used for dynamic processing of data compression of effective data based on the bandwidth efficiency dynamic evaluation result and link bandwidth occupation rate, the dynamic processing of data compression represents dynamic adjustment of data compression strategy to match the real-time processing mechanism of current low-orbit satellite communication channel condition and task demand; The data fusion module is used for dynamic adjustment of fusion weight based on the adaptability parameters and adaptability screening results of to-be-fused sensor, data fusion based on adjusted sensor fusion weight and effective data, and self-optimization of fusion weight based on fusion effect, the dynamic adjustment of fusion weight represents dynamic adjustment of fusion weight of to-be-fused sensor to optimize the overall quality and reliability of fused data.

Images