An information correlation method and device based on a cross-subsystem causal diagram, a program product, and an information visualization presentation method for a UAV ground station

Abstract

This invention relates to an information association method, apparatus, and program product based on a cross-subsystem causal graph, as well as an information visualization method for UAV ground stations. It pertains to the field of UAV technology and addresses the technical problem that existing data presentation methods are severely disconnected from the physical characteristics of the UAV system itself, making it impossible to provide operators with an intuitive and efficient global fault view across subsystems. The method includes constructing a cross-subsystem causal graph; in response to a detected fault state, matching the detected fault state with nodes in the cross-subsystem causal graph, and traversing the cross-subsystem causal graph using a directed graph traversal algorithm starting from the matched node; and constructing a fault subgraph based on the traversal results.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to an information association method, apparatus, program product based on cross-subsystem causal graphs, and an information visualization presentation method for UAV ground stations. Background Technology

[0002] In the current field of industrial drone ground stations, the presentation of data and alarm information generally adopts a "separate system independent display" architecture. The specific implementation scheme is as follows: 1. Subsystem Isolation Layout: The ground station human-machine interface (HMI) typically uses pagination, tabs, or side-by-side independent windows to divide the UAV platform into several independent units, such as the power system, avionics system, flight control system, mission payload system, and environmental perception system. Each subsystem has its own dedicated display page for centrally displaying its detailed parameters, status indicators, and alarm information.

[0003] 2. In-depth information analysis: Within each subsystem page, the design often emphasizes data completeness and longitudinal traceability. For example, the power system page displays detailed information on all raw or processed telemetry data, such as engine speed, temperature, voltage, and current; the flight control system page focuses on information such as attitude, position, and flight path tracking errors. This design allows operators to perform in-depth, specialized monitoring of individual subsystems.

[0004] 3. Simple alarm prompts: When a subsystem parameter is abnormal, common alarm methods include, but are not limited to: Generate a text alarm message in a local area (such as the alarm list) of the corresponding subsystem page; The tabs or window labels corresponding to this subsystem will change color (e.g., turn yellow or red) or display an alarm count; There may be an audio prompt.

[0005] While existing technical solutions have achieved the classification and display of data, their inherent "information silo" model has serious flaws, especially in complex fault scenarios, which greatly increases the cognitive load and decision-making risks for operators. 1. Lack of cross-domain semantic connections, resulting in fragmented global situational awareness: Existing technologies can only reflect anomalies within subsystems and cannot reveal the propagation of faults and causal relationships between different subsystems. For example, when "avionics system cooling fan failure" leads to "engine overheating in the power system," two separate alarms will appear on two isolated pages: "Avionics System: Fan Failure" and "Power System: Temperature Exceeds Limits." The causal relationship between them is implicit and missing at the interface display level. Operators must rely on their own experience and technical manuals to manually perform fault correlation and reasoning in their minds, which is inefficient and prone to errors.

[0006] 2. Manual comparison is inefficient and results in slow emergency response: In emergency situations, operators need to manually switch between different subsystem pages or scan back and forth between multiple windows to manually cross-compare the occurrence time and parameter change trends of different alarms. This process is not only time-consuming and lengthy, but also very prone to overlooking key correlations under time pressure and stress, leading to misjudgments. The brief decision-making window may therefore be missed, resulting in a major accident.

[0007] 3. Difficulty in identifying the root cause and poor targeting of treatment measures: Because the displayed symptoms are not the root cause, operators struggle to quickly distinguish between the underlying cause and the derivative fault. For example, an "engine overheating" warning could be caused by a fault in the engine itself, or by a failure in the cooling system (a subsystem of avionics or power). Current technology cannot provide this distinction, leading operators to address the symptoms (such as reducing engine power) while neglecting the true root cause (repairing the cooling system), resulting in ineffective solutions or even exacerbating the problem.

[0008] In summary, the core flaw of existing technologies lies in the severe disconnect between their data presentation methods and the physical characteristics of the UAV system itself, which are closely coupled with the system's inherent characteristics. This makes it impossible to provide operators with an intuitive and efficient global fault view across subsystems, which has become a major technical bottleneck in improving operational safety and efficiency. Summary of the Invention

[0009] In view of the above analysis, the embodiments of the present invention aim to provide an information association method, device, program product based on cross-subsystem causal graphs, and an information visualization presentation method for UAV ground stations, in order to solve the technical problem that the data presentation method of the prior art is seriously out of sync with the physical characteristics of the UAV system itself, which is closely coupled, and cannot provide operators with an intuitive and efficient cross-subsystem global fault view.

[0010] In a first aspect, embodiments of the present invention provide an information association method based on cross-subsystem causal graphs, including: Construct a cross-subsystem causal graph, wherein the nodes in the cross-subsystem causal graph are the fault states of each subsystem of the UAV, and the edges in the cross-subsystem causal graph are directed edges that characterize the causal relationship between the nodes, and the directed edges have weights that characterize the strength of the causal relationship. In response to the detected fault state, the detected fault state is matched with the nodes in the cross-subsystem cause-effect graph, and the cross-subsystem cause-effect graph is traversed using a directed graph traversal algorithm, starting from the matched node. A fault subgraph is constructed based on the traversal results, wherein the fault subgraph includes nodes and paths related to the fault state.

[0011] Based on a further improvement of the above method, the construction process of the cross-subsystem causal graph is as follows: Obtain fault data sources for each subsystem of the UAV ground station; Define the various fault states in each subsystem as nodes; If the cause of the first node being the second node is determined from the faulty data source, then a directed edge is created from the first node to the second node; and The causal relationship strength between the first node and the second node is quantified based on the fault data source, and the directed edge between the first node and the second node is weighted according to the quantization result.

[0012] Based on further improvements to the above method, the fault data source includes at least one of the following: Fault mode and impact analysis tables for each subsystem, historical fault data records for UAVs, and expert experience rule base.

[0013] Based on a further improvement of the above method, the weight of the directed edge between the first node and the second node is the statistical probability that the first node affects the second node.

[0014] Further improvements to the above method include constructing a fault subgraph based on the traversal results, such as: Determine whether the weight of the traversed directed edge is greater than a preset threshold. If so, the directed edge is included in the fault chain; otherwise, it is not included in the fault chain. The fault chain is a subgraph connected end to end by one or more directed edges.

[0015] Based on further improvements to the above method, the directed graph traversal algorithm includes any one of the following: Breadth-first search, depth-first search.

[0016] Secondly, embodiments of the present invention provide a method for visualizing information from a UAV ground station, including: A fault subgraph is constructed using the information association method based on cross-subsystem causal graphs as described in any one of the first aspects of the present invention. The fault sub-graph is visualized on the human-machine interface of the UAV ground station.

[0017] Further improvements to the above method include visualizing the fault sub-graph on the human-machine interface of the UAV ground station, including: The fault sub-graph is output in text form to a fixed global situation panel located on one side of the human-machine interaction main interface of the UAV ground station; Send instructions to the display windows of each UAV ground station subsystem involved in the fault subgraph to change their visual attributes.

[0018] Further improvements to the above method include sending instructions to the display windows of each subsystem of the UAV ground station involved in the fault subgraph to change its visual attributes, including: If the UAV ground station subsystem is identified as a potential root cause in the fault subgraph, the border of its display window will turn red and flash continuously. If the UAV ground station subsystem is identified as an affected object in the fault subgraph, the background color of its display window changes to yellow.

[0019] Thirdly, embodiments of the present invention provide an information association device based on a cross-subsystem causal graph, comprising: The first construction unit is used to construct a cross-subsystem causal graph, wherein the nodes in the cross-subsystem causal graph are the fault states of each subsystem of the UAV, and the edges in the cross-subsystem causal graph are directed edges that characterize the causal relationship between the nodes, and the directed edges have weights that characterize the strength of the causal relationship. The matching traversal unit is used to match the detected fault state with the nodes in the cross-subsystem causal graph in response to the detected fault state, and to traverse the cross-subsystem causal graph using a directed graph traversal algorithm starting from the matched node. The second construction unit is used to construct a fault subgraph based on the traversal results, wherein the fault subgraph includes nodes and paths related to the fault state. Fourthly, an embodiment of the present invention provides a computer program product, characterized in that the computer program product includes a stored computer program, which, when run by a processor, implements the information association method based on cross-subsystem causal graphs as described in any one of the first aspects of the present invention or the information visualization presentation method for UAV ground stations as described in any one of the second aspects of the present invention.

[0020] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects: 1. Fundamentally enhance overall situational awareness capabilities, achieving a leap from "information silos" to "situational fusion": This invention, for the first time at the UAV ground station level, presents the physically coupled and logically implicit fault dependencies between subsystems to the operator in an explicit and graphical manner. A fixed global situational awareness panel intuitively displays the textual fault propagation chain, and combined with dynamic visual enhancements to the original subsystem windows (such as blinking borders and changing background colors), a three-dimensional, interconnected information view is constructed. This allows operators to form a clear understanding of the global impact of a fault within a single screen and within seconds, without requiring time-consuming mental reasoning and window switching, completely breaking down information silos.

[0021] 2. Significantly shortens fault diagnosis and decision-making time, and improves emergency response efficiency: Through simulation testing and verification, this invention can significantly shorten the time from when an operator discovers an anomaly to when they locate the root cause or predict the overall impact. The system's automated correlation reasoning replaces the tedious manual comparison process, enabling the rapid identification of the core contradictions in emergency situations. This provides a valuable decision-making window for taking appropriate measures and greatly reduces the risk of accidents caused by delayed diagnosis.

[0022] 3. Effectively reduces operator cognitive load and over-reliance on professional experience: This system acts as a "tireless expert assistant," automatically performing complex causal relationship analyses. This reduces the extremely high requirements for operators' personal experience and technical skills. Even operators with relatively little experience can, under the guidance of the system, achieve a near-expert level of fault understanding, thereby reducing the risk of misjudgment due to human negligence or lack of experience and improving the overall safety and reliability of operations.

[0023] 4. A scalable and evolvable intelligent diagnostic framework is provided: The "causal graph" knowledge base constructed in this invention is a continuously optimized digital asset. Its fault data sources (FMEA, historical data, expert rules) can be continuously accumulated and updated, enabling the system's reasoning ability to be continuously enhanced with the enrichment of data and the iteration of algorithms, laying a solid technical foundation for the evolution of ground stations from "passive monitoring" to "proactive prediction and health management (PHM)".

[0024] In summary, this invention not only solves the core defect of "information silos" in existing technologies, but also brings comprehensive improvements in security, efficiency, and human factors engineering. It is a key technology for the evolution of industrial UAV ground stations towards intelligence and high reliability.

[0025] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description

[0026] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. Figure 1 A flowchart illustrating an information association method based on cross-subsystem causal graphs according to an embodiment of the present invention is shown.

[0027] Figure 2 A flowchart illustrating an information visualization method for a UAV ground station according to an embodiment of the present invention is shown.

[0028] Figure 3 A structural block diagram of an information association device based on a cross-subsystem causal graph according to an embodiment of the present invention is shown. Detailed Implementation

[0029] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0030] According to one aspect of the present invention, an information association method based on cross-subsystem causal graphs is provided. Figure 1 A flowchart illustrating an information association method based on a cross-subsystem causal graph according to an embodiment of the present invention is shown. Figure 1 As shown, this information association method based on cross-subsystem causal graphs includes: Step 101: Construct a cause-effect graph across subsystems.

[0031] In this embodiment, the nodes in the cross-subsystem causal graph represent various fault states in each subsystem of the UAV ground station, and the edges in the cross-subsystem causal graph are directed edges that characterize the causal relationship between nodes, with weights representing the strength of the causal relationship.

[0032] In this embodiment, each fault state of each UAV ground station subsystem (e.g., power system, avionics system, etc.) can be defined as a node in a cross-subsystem cause-effect graph. The attributes of a node may include node ID, the UAV ground station subsystem to which the node belongs, fault state description, risk level, etc.

[0033] In this embodiment, if the fault represented by node A is the direct or indirect cause of the fault represented by node B, then a directed edge from A to B is created.

[0034] In this embodiment, directed edges can be weighted to quantify the strength or conditional probability of causal influence. The weights can be derived from expert scores or from statistical probabilities in historical data (e.g., the probability that B also occurs when A occurs).

[0035] The cross-subsystem causal graph constructed in this embodiment is a structured directed weighted graph containing all nodes and weighted directed edges. This cross-subsystem causal graph can serve as a knowledge base for reasoning.

[0036] In this embodiment, the fault data source for constructing the cross-subsystem cause-effect graph may include at least one of the following: Fault mode and impact analysis table for each subsystem of the UAV ground station, historical fault data records of UAVs, and expert experience rule base.

[0037] The Failure Mode and Effects Analysis (FMEA) table clearly defines each failure state and its consequences, providing authoritative, engineering-verified initial causal knowledge for this invention. Historical failure data records refer to failure-event chain data extracted from past flight missions, used to correct and enrich the theoretical impact relationships in the FMEA, making them more consistent with actual statistical patterns. The expert experience rule base, existing in the form of a rule base, provides domain expert knowledge to supplement complex or implicit relationships not covered in the FMEA and historical data.

[0038] Step 102: In response to the detected fault state, the detected fault state is matched with the nodes in the cross-subsystem causal graph, and the cross-subsystem causal graph is traversed using a directed graph traversal algorithm starting from the matched node.

[0039] In this embodiment, telemetry data streams transmitted from UAVs can be received online in real time and parsed to generate standardized fault states. Once a new fault state is captured, it is immediately matched with nodes in the cross-subsystem causal graph. If a node is matched, a directed graph traversal algorithm can be used to traverse the cross-subsystem causal graph, starting from the matched node.

[0040] In this embodiment, breadth-first search or depth-first search can be used to traverse along the outgoing edge direction (forward tracing) and the incoming edge direction (backward deduction).

[0041] Step 103: Construct a fault subgraph based on the traversal results.

[0042] In this embodiment, during the traversal of the cross-subsystem causal graph, all nodes and paths strongly correlated with the current fault state can be dynamically identified based on the weights of the directed edges, forming one or more connected subgraphs, called "fault subgraphs". These fault subgraphs clearly demonstrate the potential "root causes" and possible "impacts" of the current fault state.

[0043] In some embodiments, constructing a fault subgraph based on the traversal results includes: Determine if the weight of each directed edge encountered during traversal is greater than a preset threshold. If it is, add the directed edge to the fault chain; otherwise, do not add it. A fault chain is a subgraph connected end-to-end by one or more directed edges. If the weight of the directed edge is derived from statistical probability, the preset threshold can be 50%.

[0044] According to another aspect of the present invention, an information visualization presentation method for unmanned aerial vehicle (UAV) ground stations is provided. Figure 1 A flowchart illustrating an information visualization method for an unmanned aerial vehicle (UAV) ground station according to an embodiment of the present invention is shown. Figure 1 As shown, the information visualization method for UAV ground stations includes: Step 201: Construct a cause-effect graph across subsystems.

[0045] Step 202: In response to the detected fault state, the detected fault state is matched with the nodes in the cross-subsystem causal graph, and the cross-subsystem causal graph is traversed using a directed graph traversal algorithm starting from the matched node.

[0046] Step 203: Construct a fault subgraph based on the traversal results.

[0047] Steps 201 to 203 are the same as steps 101 to 103, and will not be repeated here.

[0048] Step 204: Visualize the fault sub-graph on the human-machine interface of the UAV ground station.

[0049] In this embodiment, the fault subgraph can be integrated into the human-machine interface of the existing UAV ground station in an intuitive and non-intrusive manner.

[0050] In some embodiments, step 104 includes: The fault sub-map is output in text form to the fixed global situation panel set on one side of the human-machine interaction main interface of the UAV ground station; Send commands to the display windows of each UAV ground station subsystem involved in the fault subgraph to change their visual attributes.

[0051] In this embodiment, a vertical panel with a fixed width and adaptive height can be created on one side of the main human-machine interface of the UAV ground station. Its position and size are predefined in the layout and will not change dynamically. This vertical panel is the fixed global situation panel in this embodiment. The panel displays the output fault sub-graphs in a clear list or thumbnail format, for example: [Avionics System - Cooling Fan] - [May cause (0.82)] -> [Power System - Engine Overheating]. The panel supports vertical scrolling to accommodate multiple related pieces of information.

[0052] In this embodiment, when a fault sub-graph is listed in the fixed global situation panel, the original display windows corresponding to all subsystems involved in that fault sub-graph immediately receive visual enhancement instructions. For example, if a subsystem is identified as a potential root cause in the fault sub-graph, the border of its display window turns red and flashes continuously; if a subsystem is identified as an affected object in the fault sub-graph, the background color of its display window turns yellow. This design allows the operator's gaze to quickly establish a connection between the fixed global situation panel and the existing, familiar subsystem windows without changing their existing operating habits.

[0053] To enable those skilled in the art to clearly understand the embodiments of the present invention, the following describes in detail the specific embodiments of the present invention in conjunction with a typical engine overheating fault scenario.

[0054] Example scenario: During a power line inspection mission, a certain type of industrial drone experienced an overheating of its engine due to the failure of the avionics system's cooling fan.

[0055] I. Initialization and Knowledge Base Construction (Offline Phase) The knowledge base building module begins working during the system startup or maintenance phase.

[0056] 1.1 Data Loading: The module reads the pre-stored FMEA table and extracts two key pieces of information from it: Avionics system FMEA: Failure Mode = Cooling Fan Stuck; Local Impact = Poor Heat Dissipation of Onboard Computer; Final Impact = May Cause Degradation of Computing Performance or Warp Reboot; Powertrain FMEA: Failure Mode = Insufficient engine cooling; Cause = Cooling system malfunction (including external cooling fan); End Effect = Engine overheating, which may cause permanent damage.

[0057] 1.2 Constructing the Cause-Effect Graph: Based on the above information, the module creates two nodes and establishes directed edges in the cross-subsystem cause-effect graph: Node A: ID: AV_FAN_STALL Subsystem: Avionics Description: Cooling fan stuck and malfunctioning. Node B: ID: PU_ENGINE_OVERHEAT Subsystem: Power Unit Description: Engine overheating Directed edge: AV_FAN_STALL ->PU_ENGINE_OVERHEAT Edge weight: According to historical data, when the fan malfunctions, there is an 82% probability that it will cause the engine to overheat. Therefore, this side is assigned a weight value of 0.82.

[0058] II. Real-time monitoring and fault triggering (online phase) Data reception: While the UAV is in flight, the real-time data processing module continuously receives and parses the telemetry data stream.

[0059] Alarm Generation: At a certain moment, the powertrain sensor reports an engine oil temperature of 135℃, exceeding a preset threshold (e.g., 130℃). Based on this, the data processing module generates a standardized alarm message: {Alarm_ID: 1037, SubSystem: PowerUnit, Component: Engine, Message: "Engine Oil Overheat!", Value: 135, Timestamp: 12:05:03.256} III. The Process of Associative Reasoning Alarm Matching: The correlation inference engine listened for and captured alarm 1037. The engine matched it with nodes in the cause-effect graph and successfully matched the node PU_ENGINE_OVERHEAT.

[0060] Graph traversal search: The engine immediately starts from the node and traverses backward (in the opposite direction of the directed edges) to perform the traversal (e.g., using the breadth-first search algorithm - BFS).

[0061] Constructing the fault subgraph: After traversing one layer, the engine finds that the node AV_FAN_STALL points to the current alarm node through an edge with a weight of 0.82. The engine determines that the association strength (0.82) exceeds a preset threshold (e.g., 0.5), and therefore includes it in the fault chain. At this point, the "minimum cross-domain fault forest" is completed, containing one fault chain: AV_FAN_STALL -> (0.82) -> PU_ENGINE_OVERHEAT.

[0062] Output: The engine sends the following structured inference results to the GUI rendering module: RootCause: {Node: AV_FAN_STALL, Confidence: 0.82} CurrentAlarm: {Node: PU_ENGINE_OVERHEAT} Path: [AV_FAN_STALL, PU_ENGINE_OVERHEAT] IV. Visual Presentation and Interaction After receiving the inference results, the GUI rendering module synchronously performs the following two rendering operations: 1. Update the fixed global situation panel: In the fixed panel on the right side of the screen, add a new text record: "[Avionics System] Cooling Fan Stuck -- [May Cause (0.82)] --> [Power System] Engine Overheating". This record is highlighted (e.g., on a yellow background) to draw the operator's attention. If the panel is full, activate the vertical scroll bar to allow the operator to scroll through all historical related information.

[0063] 2. Enhance the display of the atomic system window: Based on the RootCause information, locate the display window corresponding to the avionics system, send a command to set its window border to red and make it flash continuously.

[0064] Based on the CurrentAlarm information, locate the display window corresponding to the power system, send a command to it, and temporarily change its window background color to yellow.

[0065] 3. Final Human-Computer Interaction Interface: The operator saw a change in the interface: the power system window turned yellow, while the avionics system window flashed with a red border.

[0066] The operator shifts their gaze to the fixed panel on the right and immediately sees a clear text description of the fault chain.

[0067] Without switching windows or manually recalling manuals, the operator can conclude within seconds that the root cause of the engine overheating is a malfunction of the avionics cooling fan, and that the fan problem should be checked or addressed immediately, rather than simply reducing the engine's power output.

[0068] According to another aspect of the present invention, an information association device based on a cross-subsystem causal graph is provided. For example... Figure 3As shown, the device 300 includes: a first construction unit 301, configured to construct a cross-subsystem causal graph, wherein the nodes in the cross-subsystem causal graph are fault states in various subsystems of the UAV, and the edges in the cross-subsystem causal graph are directed edges representing causal relationships between nodes, with weights representing the strength of the causal relationships; a matching and traversal unit 302, configured to, in response to a detected fault state, match the detected fault state with the nodes in the cross-subsystem causal graph, and traverse the cross-subsystem causal graph using a directed graph traversal algorithm starting from the matched node; and a second construction unit 303, configured to construct a fault subgraph based on the traversal results, wherein the fault subgraph includes nodes and paths related to the fault state. It is understood that unit 301 in the device 300... Operation of Unit 303 and Figure 1 Step 101 Step 103 is similar and will not be described in detail here.

[0069] Compared with the prior art, the embodiments of the present invention can achieve at least one of the following beneficial effects: 1. Fundamentally enhance overall situational awareness capabilities, achieving a leap from "information silos" to "situational fusion": This invention, for the first time at the UAV ground station level, presents the physically coupled and logically implicit fault dependencies between subsystems to the operator in an explicit and graphical manner. A fixed global situational awareness panel intuitively displays the textual fault propagation chain, and combined with dynamic visual enhancements to the original subsystem windows (such as blinking borders and changing background colors), a three-dimensional, interconnected information view is constructed. This allows operators to form a clear understanding of the global impact of a fault within a single screen and within seconds, without requiring time-consuming mental reasoning and window switching, completely breaking down information silos.

[0070] 2. Significantly shortens fault diagnosis and decision-making time, and improves emergency response efficiency: Through simulation testing and verification, this invention can significantly shorten the time from when an operator discovers an anomaly to when they locate the root cause or predict the overall impact. The system's automated correlation reasoning replaces the tedious manual comparison process, enabling the rapid identification of the core contradictions in emergency situations. This provides a valuable decision-making window for taking appropriate measures and greatly reduces the risk of accidents caused by delayed diagnosis.

[0071] 3. Effectively reduces operator cognitive load and over-reliance on professional experience: This system acts as a "tireless expert assistant," automatically performing complex causal relationship analyses. This reduces the extremely high requirements for operators' personal experience and technical skills. Even operators with relatively little experience can, under the guidance of the system, achieve a near-expert level of fault understanding, thereby reducing the risk of misjudgment due to human negligence or lack of experience and improving the overall safety and reliability of operations.

[0072] 4. A scalable and evolvable intelligent diagnostic framework is provided: The "causal graph" knowledge base constructed in this invention is a continuously optimized digital asset. Its fault data sources (FMEA, historical data, expert rules) can be continuously accumulated and updated, enabling the system's reasoning ability to be continuously enhanced with the enrichment of data and the iteration of algorithms, laying a solid technical foundation for the evolution of ground stations from "passive monitoring" to "proactive prediction and health management (PHM)".

[0073] In summary, this invention not only solves the core defect of "information silos" in existing technologies, but also brings comprehensive improvements in security, efficiency, and human factors engineering. It is a key technology for the evolution of industrial UAV ground stations towards intelligence and high reliability.

[0074] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. An information association method based on cross-subsystem causal graphs, characterized in that, include: Construct a cross-subsystem causal graph, wherein the nodes in the cross-subsystem causal graph are the fault states of each subsystem of the UAV, and the edges in the cross-subsystem causal graph are directed edges that characterize the causal relationship between the nodes, and the directed edges have weights that characterize the strength of the causal relationship. In response to the detected fault state, the detected fault state is matched with the nodes in the cross-subsystem cause-effect graph, and the cross-subsystem cause-effect graph is traversed using a directed graph traversal algorithm, starting from the matched node. A fault subgraph is constructed based on the traversal results, wherein the fault subgraph includes nodes and paths related to the fault state.

2. The information association method according to claim 1, characterized in that, The process of constructing the cross-subsystem cause-effect graph is as follows: Obtain fault data sources for each subsystem of the UAV ground station; Define various fault states as nodes; If the cause of the first node being the second node is determined based on the fault data source, then a directed edge is created from the first node to the second node. as well as The causal relationship strength between the first node and the second node is quantified based on the fault data source, and the directed edge between the first node and the second node is weighted according to the quantization result.

3. The information association method according to claim 2, characterized in that, The fault data source includes at least one of the following: Fault mode and impact analysis tables for each subsystem, historical fault data records for UAVs, and expert experience rule base.

4. The information association method according to claim 2, characterized in that, The weight of the directed edge between the first node and the second node is the statistical probability that the first node affects the second node.

5. The information association method according to claim 1, characterized in that, Constructing a fault subgraph based on the traversal results includes: Determine whether the weight of the traversed directed edge is greater than a preset threshold. If so, the directed edge is included in the fault chain; otherwise, it is not included in the fault chain. The fault chain is a subgraph connected end to end by one or more directed edges.

6. The information association method according to claim 1, characterized in that, The directed graph traversal algorithm includes any one of the following: Breadth-first search, depth-first search.

7. A method for visualizing information from a UAV ground station, characterized in that, include: A fault subgraph is constructed using the information association method based on cross-subsystem causal graphs as described in any one of claims 1 to 6; The fault sub-graph is visualized on the human-machine interface of the UAV ground station.

8. The information visualization presentation method according to claim 7, characterized in that, The visualization of the fault sub-graph in the human-computer interaction interface of the UAV ground station includes: The fault sub-graph is output in text form to a fixed global situation panel located on one side of the human-machine interaction main interface of the UAV ground station; Send instructions to the display windows of each UAV ground station subsystem involved in the fault subgraph to change their visual attributes.

9. The information visualization presentation method according to claim 8, characterized in that, Sending instructions to the display windows of each subsystem of the UAV ground station involved in the fault subgraph to change its visual attributes includes: If the UAV ground station subsystem is identified as a potential root cause in the fault subgraph, the border of its display window will turn red and flash continuously. If the UAV ground station subsystem is identified as an affected object in the fault subgraph, the background color of its display window changes to yellow.

10. An information association device based on cross-subsystem causal graphs, characterized in that, include: The first construction unit is used to construct a cross-subsystem causal graph, wherein the nodes in the cross-subsystem causal graph are the fault states of each subsystem of the UAV, and the edges in the cross-subsystem causal graph are directed edges that characterize the causal relationship between the nodes, and the directed edges have weights that characterize the strength of the causal relationship. The matching traversal unit is used to match the detected fault state with the nodes in the cross-subsystem causal graph in response to the detected fault state, and to traverse the cross-subsystem causal graph using a directed graph traversal algorithm starting from the matched node. The second construction unit is used to construct a fault subgraph based on the traversal results, wherein the fault subgraph includes nodes and paths related to the fault state.

11. A computer program product, characterized in that, The computer program product includes a stored computer program that, when executed by a processor, implements the information association method based on cross-subsystem causal graphs as described in any one of claims 1 to 6, or the information visualization presentation method for UAV ground stations as described in any one of claims 7 to 9.

Images