Visual display method, device and system for real-time state of unmanned aerial vehicle

Abstract

The invention relates to the technical field of unmanned aerial vehicle state visualization, in particular to an unmanned aerial vehicle real-time state visualization display method, device and system, and the method comprises the steps of multi-source data dynamic collection, adaptive time window modeling, state weight calculation, hierarchical decision regulation and control, real-time interface updating and the like. Through combination of dynamic weight adjustment and a distributed architecture, a resource allocation and interface rendering strategy is optimized, and the operation efficiency in a complex task scene is improved. According to the invention, comprehensive perception and efficient display of the operation state of the unmanned aerial vehicle can be realized, user interaction experience is enhanced, key information is prevented from being ignored, and system stability and real-time performance are ensured.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) technology and data visualization technology, specifically a method, device and system for real-time status visualization of UAVs. Background Technology

[0002] Real-time UAV status visualization methods refer to a class of technical solutions that focus on the multi-dimensional data collection, integration, and dynamic display during UAV flight missions. Through intelligent analysis and optimized display strategies, they aim to achieve "comprehensive perception of UAV operational status, efficient integration of multi-source information, and enhanced human-machine interaction experience." The core logic is that in complex mission scenarios, users' rapid understanding and accurate decision-making regarding UAV status depend on the real-time nature, intuitiveness, and comprehensiveness of information. Therefore, technical means are needed to transform scattered data into clear and actionable visualizations.

[0003] Current methods and systems for real-time status visualization of unmanned aerial vehicles (UAVs) suffer from the following technical limitations: Existing technologies are insufficient in multi-dimensional data integration, failing to fully integrate key information such as flight parameters, environmental perception, and fault diagnosis; while a UAV operation and maintenance management method and system (publication number CN113658455B) achieves centralized management of flight information, it lacks in-depth analysis and dynamic visualization capabilities for real-time status data; and while a vehicle-mounted UAV ground control system (publication number CN110045663B) provides a wide operational field of view, it does not address intelligent integration of status data or interactive interface design. These limitations may affect the efficiency of information acquisition and ease of operation for users in complex task scenarios. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method, device and system for real-time status visualization of unmanned aerial vehicles.

[0005] The technical solution adopted to solve the above-mentioned technical problems is: a method for real-time status visualization of unmanned aerial vehicles (UAVs), comprising the following specific steps:

[0006] Step 1: Dynamic acquisition of multi-source data: The acquisition objects include flight parameters, environmental perception data, mission execution status, fault diagnosis information, and communication link status;

[0007] Step 2: Adaptive Time Window Modeling: A dynamic time window is used, with a default window length of 12 hours. Data within the window is divided into multiple sub-intervals according to fixed time slices, with each 3 minutes serving as a storage unit. A circular queue is used to store the window data, and when new data enters, the oldest data block that exceeds the window length is evicted.

[0008] Step 3: State weight calculation: Based on the collected multi-source data, calculate the state weight value of each UAV, and smooth the weight value using the exponential weighted moving average method;

[0009] Step 4: Hierarchical decision-making and control: Construct a state priority matrix, set different priority division rules and control strategies, and then optimize resource allocation;

[0010] Step 5: Real-time UI update: Push update commands to the display terminal via the WebSocket protocol, adjust the rendering queue order, limit the update frequency of low-priority states, reserve rendering buffers for high-priority states, and achieve elastic scaling in response to sudden data flows.

[0011] The above technical solutions combine comprehensive perception of the drone's operational status with optimization of user interaction experience. They also propose a display strategy based on dynamic weight adjustment and design a hybrid system that integrates distributed architecture and intelligent analysis algorithms to improve operational efficiency in complex task scenarios.

[0012] Furthermore, the state weight values ​​are calculated using the following formula: For each UAV, calculate its... Wherein, numerator: the sum of weights corresponding to all states within the statistical window, W is the weight coefficient of the i-th state, and State is the corresponding state value; denominator: the total number of states generated by UAV k within the window period.

[0013] Furthermore, the exponentially weighted moving average processing adopts the following formula: a smoothing factor α is introduced to smooth the time series data and suppress short-term fluctuations: y(t)=α×y(t)+(1-α)×y(t-1). After each time slice, y(t) is recalculated based on the latest window data and weighted and merged with the smoothed value y(t-1) of the previous period.

[0014] Furthermore, the state priority matrix includes the following formula: Global statistics calculation: When p i When μ > μ + 2σ, the display priority is increased; when μ ≤ p i When p ≤ μ + 2σ, maintain the current priority; when p i When the value is ≤μ, the priority of the classification is reduced.

[0015] Furthermore, the following specific formula is used to reduce priority in a tiered manner: Furthermore, resource allocation optimization adopts the following formula: Here, λ is the control intensity coefficient. When λ>1, the resource proportion of the high-weight state is strengthened. The denominator is the sum of the powers of all drone weight values ​​to ensure the normalization of the total resource allocation ratio. Through the above technical solution, the flexibility of the resource allocation process is improved.

[0016] Furthermore, the real-time interface update employs a feedback closed-loop mechanism, recalculating the update control strategy every 15 minutes. If a drone remains in the lower priority range for four consecutive cycles, a warning is triggered, prompting manual intervention. This technical solution improves the overall update frequency and prevents critical information from being overlooked due to program misjudgments.

[0017] Furthermore, it includes a data acquisition module, a weight calculation module, a decision control module, and an interface update module. The data acquisition module is a data capture probe deployed on the UAV communication node. The weight calculation module is based on a Spark distributed computing cluster. The decision control module includes a time window cache, a Kalman filter parameter tuning module, and a priority matrix generator. The interface update module is a dynamic rendering component based on a microservice architecture.

[0018] The beneficial effects of this invention are as follows: This invention combines comprehensive perception of the drone's operating status with optimization of user interaction experience, proposes a display strategy based on dynamic weight adjustment, and designs a hybrid system that integrates distributed architecture and intelligent analysis algorithms, thereby improving operational efficiency in complex task scenarios. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the real-time status visualization method for unmanned aerial vehicles (UAVs) in an embodiment of the present invention.

[0020] Figure 2 This is a data processing logic diagram for adaptive time window modeling in an embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram illustrating the construction and control strategy of the state priority matrix in an embodiment of the present invention;

[0022] Figure 4 This is a system module architecture diagram in an embodiment of the present invention.

[0023] The attached diagram is labeled as follows: 1. Data acquisition module; 2. Weight calculation module; 3. Decision control module; 4. Interface update module; 5. Circular queue memory; 6. Kalman filter parameter tuning module; 7. Priority matrix generator; 8. Dynamic rendering component. Detailed Implementation

[0024] This invention provides a method, apparatus, and system for real-time status visualization of unmanned aerial vehicles (UAVs), the specific implementation of which is described in conjunction with the appendix. Figure 1 To be continued Figure 4This paper provides a detailed explanation. In practical applications, this method achieves comprehensive perception of the UAV's operational status and optimizes the user interaction experience through the collaborative work of multiple modules. It also proposes a display strategy based on dynamic weight adjustment and designs a hybrid system that integrates distributed architecture and intelligent analysis algorithms to improve operational efficiency in complex task scenarios.

[0025] like Figure 1 As shown, the real-time status visualization method for unmanned aerial vehicles (UAVs) in this embodiment of the invention includes a complete set of steps from data acquisition to interface update. First, in the data acquisition module 1, a data acquisition probe deployed at the UAV communication node is responsible for collecting multi-source data, including flight parameters, environmental perception data, mission execution status, fault diagnosis information, and communication link status. This data is transmitted to a ground station or cloud server via the UAV's onboard sensors and communication equipment, forming a complete data flow channel. The data acquisition module 1 is directly connected to the weight calculation module 2, which transmits the collected raw data to the subsequent processing unit.

[0026] The weight calculation module 2 processes the collected data based on a Spark distributed computing cluster. For example... Figure 2 As shown, the adaptive time window modeling adopts a dynamic time window mechanism with a default window length of 12 hours. Data within the window is divided into multiple sub-intervals according to fixed time slices, with each sub-interval lasting 3 minutes. A circular queue memory 5 is used to store the window data. When new data arrives, the circular queue memory 5 automatically discards the oldest data block exceeding the window length, ensuring the timeliness of data storage. Based on this, the weight calculation module 2 calculates the state weight value for each UAV based on the collected multi-source data. The formula for calculating the state weight value is: numerator is the sum of weights corresponding to all states within the statistical window, W is the weight coefficient of the i-th state, State is the corresponding state value, and the denominator is the total number of states generated by UAV k within the window period (Total). After calculation, the weight calculation module 2 smooths the weight values ​​using an exponentially weighted moving average method, introducing a smoothing factor α to smooth the time series data and suppress short-term fluctuations. After each time slice, y(t) is recalculated based on the latest window data and weighted and merged with the smoothed value y(t-1) of the previous period. The smoothed weight value is then passed to the decision control module 3.

[0027] The decision control module 3 includes a time window buffer, a Kalman filter parameter tuning module 6, and a priority matrix generator 7. For example... Figure 3 As shown, the priority matrix generator 7 constructs the state priority matrix and sets the rules for different priority levels and control strategies. Global statistics calculations include the system average weight and standard deviation. The system average weight is the mean of all UAVs, and the standard deviation is calculated to measure the dispersion of the weight distribution. When p... iWhen μ > μ + 2σ, the display priority is increased; when μ ≤ p i When p ≤ μ + 2σ, maintain the current priority; when p i When μ is less than or equal to μ, the priority is reduced in stages. The specific formula for reducing priority in stages is as follows: Resource allocation optimization uses formulas Where λ is the control intensity coefficient, when λ>1, the resource proportion of the high-weight state is strengthened. The denominator is the sum of the powers of the weight values ​​of all UAVs, ensuring that the total resource allocation ratio is normalized. The decision control module 3 generates control instructions based on the above calculation results and sends the instructions to the interface update module 4.

[0028] The interface update module 4, based on the microservice architecture's dynamic rendering component 8, pushes update commands to the display terminal via the WebSocket protocol. For example... Figure 4 As shown, the interface update module 4 adjusts the rendering queue order, limits the update frequency of low-priority states, and reserves rendering buffers for high-priority states to achieve elastic scaling in the event of sudden data surges. The interface update module 4 also employs a feedback loop mechanism, recalculating the update control strategy every 15 minutes. If a drone remains in the lower priority range for four consecutive cycles, a warning is triggered, prompting manual intervention. Through these mechanisms, the interface update module 4 effectively improves the overall update frequency and prevents critical information from being overlooked due to program misjudgments.

[0029] In this embodiment of the invention, the modules work closely together through data flow and control flow to jointly complete the task of real-time visualization of the UAV's status. Data acquisition module 1 transmits multi-source data to weight calculation module 2. Weight calculation module 2 processes the data using a distributed computing cluster and generates smoothed weight values, which are then transmitted to decision control module 3. Decision control module 3 uses Kalman filter parameter tuning module 6 and priority matrix generator 7 to generate control commands and sends these commands to interface update module 4. Interface update module 4 displays the final visualization result on the user terminal using dynamic rendering component 8. Throughout the process, data transmission and command delivery between modules are achieved through a high-speed network, ensuring the system's real-time performance and stability.

[0030] In practical applications, such as when a swarm of drones performs forest fire monitoring tasks, the data acquisition module 1 collects multi-source data in real time, including drone flight altitude, speed, battery level, ambient temperature, and wind speed, and transmits this data to the weight calculation module 2. The weight calculation module 2 calculates the state weight value of each drone based on a dynamic time window mechanism and a smoothing algorithm, and then transmits the results to the decision control module 3. The decision control module 3 generates control instructions based on a priority matrix, such as prioritizing the display of drone status information near the fire source area while reducing the status update frequency of drones far from the fire source area. The interface update module 4 adjusts the display interface according to the control instructions, ensuring that users can quickly obtain key information and make decisions.

[0031] Through the above implementation methods, this invention achieves comprehensive perception of the UAV's operational status and optimizes the user interaction experience. It also proposes a display strategy based on dynamic weight adjustment and designs a hybrid system that integrates distributed architecture and intelligent analysis algorithms, thereby improving operational efficiency in complex task scenarios.

[0032] To enable those skilled in the art to fully understand and implement this invention, the specific implementation principles of this invention are further supplemented below with a specific application scenario.

[0033] In forest fire monitoring missions, drone swarms need to collect multi-source data in real time, including flight altitude, speed, battery level, ambient temperature, and wind speed, and transmit this data to ground stations or cloud servers for processing. First, data acquisition module 1 acquires multi-source data through data acquisition probes deployed at the drone's communication nodes. This data includes, but is not limited to, flight parameters, environmental perception information, mission execution status, fault diagnosis results, and communication link status. Data acquisition module 1 then transmits the collected raw data to weight calculation module 2 via a high-speed network, forming a complete data stream. This step ensures that all critical information is captured promptly and enters the subsequent processing flow.

[0034] The weight calculation module 2 processes the received multi-source data based on a Spark distributed computing cluster. For example... Figure 2As shown, the adaptive time window modeling adopts a dynamic time window mechanism with a default window length of 12 hours. Data within the window is divided into multiple sub-intervals according to fixed time slices, with each sub-interval lasting 3 minutes. A circular queue memory 5 is used to store the window data. When new data arrives, the circular queue memory 5 automatically discards the oldest data block exceeding the window length, ensuring the timeliness of data storage. Based on this, the weight calculation module 2 calculates the state weight value for each UAV based on the collected multi-source data. The formula for calculating the state weight value is: numerator is the sum of weights corresponding to all states within the statistical window, W is the weight coefficient of the i-th state, State is the corresponding state value, and the denominator is the total number of states generated by UAV k within the window period (Total). After calculation, the weight calculation module 2 smooths the weight values ​​using an exponentially weighted moving average method, introducing a smoothing factor α to smooth the time series data and suppress short-term fluctuations. After each time slice, y(t) is recalculated based on the latest window data and weighted and merged with the smoothed value y(t-1) of the previous period. The smoothed weight value is then passed to the decision control module 3. This step effectively improves the real-time performance and stability of data processing through a dynamic time window mechanism and an exponentially weighted moving average method, while avoiding misjudgments caused by short-term fluctuations.

[0035] The decision control module 3 receives smoothed weight values ​​from the weight calculation module 2 and generates control commands using the Kalman filter parameter tuning module 6 and the priority matrix generator 7. For example... Figure 3 As shown, the priority matrix generator 7 constructs the state priority matrix and sets the rules for dividing different priorities and the control strategies. Global statistics calculations include the system average weight and standard deviation. The system's average weight is calculated as the mean of all drones, and the standard deviation is calculated to measure the dispersion of the weight distribution. When p i When μ > μ + 2σ, the display priority is increased; when μ ≤ p i When p ≤ μ + 2σ, maintain the current priority; when p i When μ is less than or equal to μ, the priority is reduced in stages. The specific formula for reducing priority in stages is as follows: Resource allocation optimization uses formulas Where λ is the control intensity coefficient, λ represents the resource proportion of the high-weight state, and the denominator is the sum of the powers of all UAV weight values ​​to ensure the normalization of the total resource allocation ratio. The decision control module 3 generates control instructions based on the above calculation results and sends the instructions to the interface update module 4. This step, through the priority matrix and resource allocation optimization algorithm, realizes intelligent hierarchical and dynamic control of UAV status information, ensuring that key information can be displayed first.

[0036] The interface update module 4, based on the microservice architecture's dynamic rendering component 8, pushes update commands to the display terminal via the WebSocket protocol. For example... Figure 4 As shown, the interface update module 4 adjusts the rendering queue order, limits the update frequency of low-priority states, and reserves rendering buffers for high-priority states to achieve elastic scaling in the event of sudden data surges. The interface update module 4 also employs a feedback loop mechanism, recalculating the update control strategy every 15 minutes. If a drone remains in the lower priority range for four consecutive cycles, a warning is triggered, prompting manual intervention. Through this mechanism, the interface update module 4 effectively improves the overall update frequency, preventing program misjudgments that could lead to the neglect of critical information. This step, through dynamic rendering components and a feedback loop mechanism, ensures that users can quickly obtain key information and make accurate decisions in complex task scenarios.

[0037] Throughout the process, the modules work closely together through data and control flows to jointly complete the task of real-time visualization of the UAV's status. Data acquisition module 1 transmits multi-source data to weight calculation module 2. Weight calculation module 2 processes the data using a distributed computing cluster and generates smoothed weight values, which are then transmitted to decision-making and control module 3. Decision-making and control module 3 uses Kalman filter parameter tuning module 6 and priority matrix generator 7 to generate control commands and sends them to interface update module 4. Interface update module 4 displays the final visualization results on the user terminal using dynamic rendering component 8. Throughout the entire process, data transmission and command delivery between modules are achieved through a high-speed network, ensuring the system's real-time performance and stability.

[0038] Through the specific implementation methods described above, this invention achieves comprehensive perception of UAV operational status and optimizes user interaction experience. It also proposes a display strategy based on dynamic weight adjustment and designs a hybrid system integrating distributed architecture and intelligent analysis algorithms, significantly improving operational efficiency in complex task scenarios. For example, in forest fire monitoring tasks, the system can prioritize displaying UAV status information near the fire source area while reducing the status update frequency of UAVs far from the fire source area, thereby ensuring that users can quickly obtain key information and make decisions. This dynamic weight adjustment and priority allocation mechanism enables the system to exhibit higher flexibility and reliability when facing complex task scenarios.

Claims

1. A method for real-time status visualization of a drone, characterized in that, The specific steps include the following: Step 1: Dynamic acquisition of multi-source data: The acquisition objects include flight parameters, environmental perception data, mission execution status, fault diagnosis information, and communication link status; Step 2: Adaptive time window modeling: A dynamic time window is adopted, with a default window length of 12 hours. The data within the window is divided into multiple sub-intervals according to fixed time slices, with each 3 minutes serving as a storage unit. A circular queue memory (5) is used to store the window data. When new data enters, the earliest data block that exceeds the window length is eliminated. Step 3: State weight calculation: Based on the collected multi-source data, calculate the state weight value of each UAV, and smooth the weight value using the exponential weighted moving average method; Step 4: Hierarchical decision-making and control: Construct a state priority matrix, set different priority division rules and control strategies, and then optimize resource allocation; Step 5: Real-time UI update: Push update commands to the display terminal via the WebSocket protocol, adjust the rendering queue order, limit the update frequency of low-priority states, reserve rendering buffers for high-priority states, and achieve elastic scaling in response to sudden data flows.

2. The method for real-time status visualization of a UAV according to claim 1, characterized in that, The state weight value is calculated using the following formula: For each UAV, calculate its state weight value = 1 / (W×State)Total, where the numerator is the sum of the weights corresponding to all states within the statistical window, W is the weight coefficient of the i-th state, and State is the corresponding state value; the denominator is the total number of states generated by UAV k within the window period, Total.

3. The method for real-time status visualization of a UAV according to claim 2, characterized in that, The exponentially weighted moving average processing adopts the following formula: a smoothing factor α is introduced to smooth the time series data and suppress short-term fluctuations: y(t)=α×y(t)+(1-α)×y(t-1). After each time slice, y(t) is recalculated based on the latest window data and weighted and merged with the smoothed value y(t-1) of the previous period.

4. The method for real-time status visualization of a UAV according to claim 3, characterized in that, The state priority matrix includes the following formula: Global statistics calculation: When p i When μ > μ + 2σ, the display priority is increased; when μ ≤ p i When p ≤ μ + 2σ, maintain the current priority; when p i When the value is ≤μ, the priority of the classification is reduced.

5. The method for real-time status visualization of a UAV according to claim 4, characterized in that, The specific formula for reducing priority in the hierarchical manner is as follows:

6. The method for real-time status visualization of a UAV according to claim 5, characterized in that, The resource allocation optimization adopts the following formula: Where λ is the control intensity coefficient, when λ>1, the resource proportion of the high-weight state is strengthened, and the denominator is the sum of the powers of the weight values ​​of all drones to ensure the normalization of the total resource allocation ratio.

7. The method for real-time status visualization of a UAV according to claim 6, characterized in that, The real-time interface update adopts a feedback closed-loop mechanism, recalculating and updating the control strategy every 15 minutes. If a drone is in the reduced priority range for four consecutive cycles, a warning is triggered and manual intervention is prompted.

8. A real-time status visualization system for unmanned aerial vehicles (UAVs), characterized in that, It includes a data acquisition module (1), a weight calculation module (2), a decision control module (3), and an interface update module (4). The data acquisition module (1) is a data capture probe deployed on the UAV communication node. The weight calculation module (2) is based on a Spark distributed computing cluster. The decision control module (3) includes a time window buffer, a Kalman filter parameter tuning module (6), and a priority matrix generator (7). The interface update module (4) is based on a dynamic rendering component (8) of a microservice architecture.

Images