Unmanned aerial vehicle non-visual flight dynamic threat assessment and control method in uninhabited area

Abstract

The application discloses an unmanned aerial vehicle non-visual flight dynamic threat assessment and control method in an unmanned area, and belongs to the technical field of air traffic control systems. The method comprises the following steps: acquiring multi-source sensing data in the flight process of an unmanned aerial vehicle in real time, and generating unmanned aerial vehicle self-state reference parameters; generating a multi-source quantitative threat data set based on the multi-source sensing data; fusing the unmanned aerial vehicle self-state reference parameters and the multi-source quantitative threat data set, and generating a comprehensive flight risk assessment result; generating an adaptive flight control strategy based on the comprehensive flight risk assessment result; generating a dynamic adjustment control instruction set based on the adaptive flight control strategy; and performing feedback adjustment based on the execution effect of the dynamic adjustment control instruction set. The technical means of quantifying multi-source threats, performing comprehensive risk assessment, generating an adaptive control strategy and performing feedback optimization based on the execution effect can significantly improve the autonomous flight safety and intelligent level of unmanned aerial vehicles in a complex dynamic environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of air traffic control systems, in particular to a method for evaluating and controlling dynamic threats of non-visual flight of unmanned aerial vehicles in unmanned areas. BACKGROUND

[0002] Currently, unmanned aerial vehicles, commonly known as drones, have been widely used in surveying, inspection, logistics and security, etc. fields, especially in vast and difficult-to-enter unmanned areas, and their non-visual long-distance flight capabilities show great application potential. Such flight tasks usually require the unmanned aerial vehicle to complete long-distance and long-time flight autonomously without direct visual observation by the operator, which puts high requirements on the autonomy and reliability of the flight control system.

[0003] The existing flight control technology of unmanned aerial vehicles mainly relies on pre-planned flight paths and integrates basic sensor systems on the aircraft for simple obstacle avoidance. During non-visual flight, the operator of the ground station remotely monitors the basic flight telemetry data of the unmanned aerial vehicle, such as position, height, speed and power, etc. When encountering unexpected situations, the operator usually needs to make manual judgments and issue remote control instructions, or the simple obstacle avoidance logic on the aircraft triggers a pre-set hover or return program.

[0004] However, the above-mentioned prior art solution has obvious defects. The pre-set flight path cannot cope with dynamic changes in the flight process, such as sudden weather conditions, unexpected equipment failures or temporary entry of other aircraft into the airspace. Remote manual intervention is not reliable in the unmanned area environment with signal delay or interruption, and cannot meet the needs of large-scale fleet autonomous operation. The simple obstacle avoidance logic on the aircraft is usually only for a single type of threat and lacks comprehensive assessment capability for multiple concurrent threats. Its risk judgment model usually uses a fixed threshold and does not consider factors such as real-time state of the unmanned aerial vehicle, such as remaining power or communication quality, resulting in inaccurate risk assessment results and lack of targetedness and flexibility in response strategies. SUMMARY

[0005] To solve the above problems, the present application provides a method for evaluating and controlling dynamic threats of non-visual flight of unmanned aerial vehicles in unmanned areas, which uses real-time quantitative multi-source threats, integrates the real-time state of the unmanned aerial vehicle for comprehensive risk assessment, generates adaptive control strategies and optimizes feedback based on execution effect, which can significantly improve the safety and intelligence level of autonomous flight of unmanned aerial vehicles in complex dynamic environments.

[0006] The above-mentioned object can be achieved by the following solution:

[0007] The unmanned area unmanned aerial vehicle non-visual flight dynamic threat assessment and control method comprises the following steps: acquiring multi-source sensing data in the flight process of an unmanned aerial vehicle in real time, and generating unmanned aerial vehicle state reference parameters; based on the multi-source sensing data, a preset threat source is quantified, the offset of the quantized preset threat source and a preset safety threshold is calculated, and the offset is combined to generate a multi-source quantized threat data set, wherein the multi-source quantized threat data set comprises meteorological threat values, equipment failure risk threat values and airspace conflict risk threat values; the unmanned aerial vehicle state reference parameters and the multi-source quantized threat data set are input into a preset weight distribution fusion model to generate a comprehensive flight risk assessment result; based on the comprehensive flight risk assessment result, a flight control target value corresponding to a threat item is calculated, and adaptive flight control strategies are generated by combining the flight control target values; based on the adaptive flight control strategies, a dynamic adjustment control instruction set is generated and sent to the unmanned aerial vehicle, and the heading angle, height value and speed value of the unmanned aerial vehicle are set; based on the execution effect of the dynamic adjustment control instruction set, the threat values of the threat items before and after execution are generated to generate strategy optimization data.

[0008] Optionally, the generation of the unmanned aerial vehicle state reference parameters comprises: extracting the three-dimensional position, speed, attitude, remaining power and communication link quality of the unmanned aerial vehicle from the multi-source sensing data; and combining the three-dimensional position, speed, attitude, remaining power and communication link quality of the unmanned aerial vehicle to generate the unmanned aerial vehicle state reference parameters.

[0009] Optionally, the generation of the multi-source quantized threat data set comprises: classifying the multi-source sensing data to generate environment data, body data and airspace data; respectively performing risk assessment on the environment data, body data and airspace data to generate meteorological threat values, equipment failure risk threat values and airspace conflict risk threat values; and combining the meteorological threat values, equipment failure risk threat values and airspace conflict risk threat values to generate the multi-source quantized threat data set.

[0010] Optionally, the generation of the meteorological threat values, equipment failure risk threat values and airspace conflict risk threat values comprises: comparing real-time parameters in the environment data, body data and airspace data with preset environment data safety thresholds, body data safety thresholds and airspace data safety thresholds respectively; calculating the deviation degree of the real-time parameters relative to the safety thresholds; and quantizing the deviation degree to generate the meteorological threat values, equipment failure risk threat values and airspace conflict risk threat values.

[0011] Optionally, the generating the comprehensive flight risk assessment result comprises: inputting the unmanned aerial vehicle self-state reference parameter and the multi-source quantitative threat data set as input items into a preset weight allocation fusion model to dynamically allocate different weight coefficients to the meteorological threat threat value, the equipment failure risk threat value and the airspace conflict risk threat value after weighting fusion to generate a comprehensive flight risk assessment result.

[0012] Optionally, the generating the adaptive flight management and control strategy specifically comprises: comparing the comprehensive flight risk assessment result with a preset risk level threshold to generate a risk disposal level; when the risk disposal level indicates that the flight state needs to be adjusted, identifying the specific threat source and the specific threat source quantitative parameter from the original data of the comprehensive flight risk assessment result that cause the risk assessment result to exceed the corresponding threshold; based on the identified specific threat source and the quantitative parameter thereof, calculating and generating a corresponding flight control target value; combining the flight control target values to form the adaptive flight management and control strategy.

[0013] Optionally, the generating the dynamic adjustment control instruction set comprises: analyzing the adaptive flight management and control strategy to extract the flight control target value; based on the flight control target value, generating a to-be-executed basic control instruction, wherein the basic control instruction comprises the setting of a target heading angle, a target height value and a target speed value; combining the basic control instruction to form the dynamic adjustment control instruction set.

[0014] Optionally, the generating the strategy optimization data comprises: analyzing the dynamic adjustment control instruction set to obtain the specific threat item to be coped with and the threat value of the threat item before execution of the dynamic adjustment control instruction set; after execution of the dynamic adjustment control instruction set, based on the real-time multi-source sensing data, generating a new threat value; calculating the change amount of the threat value of the threat item before execution of the dynamic adjustment control instruction set and the new threat value to generate the strategy optimization data.

[0015] Optionally, the method further comprises: based on the strategy optimization data, adjusting the weight coefficient allocated by the preset weight allocation fusion model.

[0016] Based on the same inventive concept, the present application also provides an unmanned area unmanned aerial vehicle non-visual flight management and control system based on dynamic threat assessment, comprising:

[0017] A state monitoring module for real-time acquisition of multi-source sensing data in the flight process of the unmanned aerial vehicle to generate unmanned aerial vehicle self-state reference parameters;

[0018] a threat quantification module, configured to quantize preset threat sources based on the multi-source sensing data, calculate offset amounts of the quantized preset threat sources from preset safety thresholds, and merge the offset amounts to generate a multi-source quantized threat data set, wherein the multi-source quantized threat data set includes a meteorological threat threat value, a device failure risk threat value, and an airspace conflict risk threat value;

[0019] a risk fusion module, configured to input the unmanned aerial vehicle self-state reference parameters and the multi-source quantized threat data set into a preset weight allocation fusion model to generate a comprehensive flight risk assessment result;

[0020] a strategy decision module, configured to calculate flight control target values corresponding to threat items based on the comprehensive flight risk assessment result, and combine the flight control target values to generate an adaptive flight management and control strategy;

[0021] an instruction issuing module, configured to generate a dynamic adjustment control instruction set based on the adaptive flight management and control strategy, and issue the dynamic adjustment control instruction set to the unmanned aerial vehicle to set a heading angle, an altitude value, and a speed value of the unmanned aerial vehicle;

[0022] an execution feedback module, configured to generate strategy optimization data by comparing threat values of threat items before and after execution of the dynamic adjustment control instruction set;

[0023] an adaptive updating module, configured to generate strategy optimization data by comparing threat values of threat items before and after execution of the dynamic adjustment control instruction set.

[0024] Compared with the prior art, the present application has the following advantages:

[0025] 1. The present application can generate a risk profile most relevant to the current flight situation in real time by constructing a comprehensive evaluation model dynamically fusing the unmanned aerial vehicle self-state and external multi-source threats, greatly improving the accuracy and authenticity of risk assessment compared with the traditional method of isolated evaluation of various threats or the use of static risk thresholds, thereby providing a solid and reliable foundation for subsequent decision-making and comprehensively enhancing the safety of flight.

[0026] 2. The present application realizes adaptive and precise management and control strategy, and the system can backtrack and identify specific threat sources that cause risk to rise based on the comprehensive risk assessment result, and then generate flight control targets directly targeting the threat sources. This point-to-point disposal method avoids broad or ineffective flight adjustments, improves the efficiency and success rate of risk avoidance operations, and ensures that the unmanned aerial vehicle can effectively resolve risks at the minimum cost.

[0027] 3.The application establishes a complete closed-loop feedback and self-optimization mechanism, which quantitatively evaluates the actual implementation effect of each control strategy, and the system can learn from it and continuously iterate its internal threat assessment and decision-making model.This makes the intelligent level of the flight control system continuously improve with the accumulation of flight experience, enhancing its robustness and environmental adaptability when facing unknown and unexpected conditions.

[0028] Other features and advantages of the present application will be set forth in the following description, and in part will be apparent from the description, or can be learned by practice of the application. The objects and other advantages of the present application will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0030] Figure 1 is a structural schematic diagram of the unmanned aerial vehicle non-visual flight dynamic threat assessment and control method in the unmanned area of the embodiment of the present application.

[0031] Figure 2 is a multi-source threat quantification and dynamic weight distribution relationship curve schematic diagram of the embodiment of the present application.

[0032] Figure 3 is a self-adaptive control strategy execution effect curve schematic diagram of the embodiment of the present application.

[0033] Figure 4 is a structural schematic diagram of the unmanned aerial vehicle non-visual flight dynamic threat assessment and control method in the unmanned area of the embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0035] REFERENCE Figure 1An embodiment of the present application provides a method for evaluating and controlling non-visual flight dynamic threats of unmanned aerial vehicles in uninhabited areas, which adopts technical means of real-time quantitative multi-source threats, comprehensive risk assessment by fusing the state of the unmanned aerial vehicle itself, generation of adaptive control strategies and feedback optimization based on the execution effect, and can significantly improve the autonomous flight safety and intelligent level of the unmanned aerial vehicle in a complex dynamic environment.

[0036] The method of the embodiment specifically comprises:

[0037] Real-time acquisition of multi-source sensing data in the flight process of the unmanned aerial vehicle, generation of the state reference parameter of the unmanned aerial vehicle itself;

[0038] Based on the multi-source sensing data, the preset threat sources are quantified, the offset of the quantized preset threat sources and the preset safety threshold is calculated, and the offset is combined to generate a multi-source quantitative threat data set, wherein the multi-source quantitative threat data set includes a meteorological threat threat value, a device failure risk threat value and an airspace conflict risk threat value;

[0039] The state reference parameter of the unmanned aerial vehicle itself and the multi-source quantitative threat data set are input into a preset weight distribution fusion model to generate a comprehensive flight risk assessment result;

[0040] Based on the comprehensive flight risk assessment result, a flight control target value corresponding to the threat item is calculated, and the flight control target value is combined to generate an adaptive flight control strategy;

[0041] Based on the adaptive flight control strategy, a dynamic adjustment control instruction set is generated and delivered to the unmanned aerial vehicle, and the heading angle, height value and speed value of the unmanned aerial vehicle are set;

[0042] Based on the execution effect of the dynamic adjustment control instruction set, the threat values of the threat items before and after execution are generated to generate strategy optimization data.

[0043] Specifically, a complete and self-optimizing closed-loop flight control process is constructed. The core is to dynamically fuse the self-state cognition of the unmanned aerial vehicle with the external threat perception to achieve intelligent decision-making. The method first establishes the unmanned aerial vehicle's own state benchmark and the quantitative threat model of the external environment, equipment and airspace through multi-source sensor data synchronization; the two are fused to generate a comprehensive flight risk assessment result, which not only reflects the severity of the threat, but also considers the bearing capacity of the unmanned aerial vehicle under the current state; based on the comprehensive risk, the system autonomously generates a customized and adaptive flight control strategy, and converts it into specific control instructions for execution; a feedback mechanism is introduced, which generates strategy optimization data by evaluating the actual mitigation effect of the threat after the execution of the control instructions, and uses the data to adjust and optimize the initial threat quantification evaluation process, thereby forming an intelligent control closed loop that continuously learns and iteratively evolves. Significantly improve the autonomy, safety and environmental adaptability of the unmanned aerial vehicle when performing non-visual flight tasks in complex, dynamic and unpredictable environments such as unmanned areas. Through dynamic threat assessment and adaptive control, the method enables the unmanned aerial vehicle to actively and in real time identify and respond to multi-dimensional potential risks in the flight process, rather than passively executing pre-set routes or relying on remote human intervention. More importantly, the built-in feedback optimization closed-loop mechanism gives the flight control system the ability to learn and continuously evolve, enabling the risk disposal strategy of the unmanned aerial vehicle to be continuously improved through experience accumulation, thereby exhibiting higher reliability and intelligence level in the long-term operation, and ultimately ensuring that the unmanned aerial vehicle can safely and efficiently complete the flight task.

[0044] Optionally, the generating the unmanned aerial vehicle's own state benchmark parameter comprises:

[0045] extracting the three-dimensional position, speed, attitude, remaining power and communication link quality of the unmanned aerial vehicle from the multi-source sensor data;

[0046] combining the three-dimensional position, speed, attitude, remaining power and communication link quality of the unmanned aerial vehicle to generate the unmanned aerial vehicle's own state benchmark parameter.

[0047] Specifically, through the multi-source sensors and the airborne data link carried by the unmanned aerial vehicle, multi-source sensing data streams are collected and summarized in real time. From the multi-source sensing data streams, several key parameters representing the core operating state of the unmanned aerial vehicle are accurately analyzed. This includes three-dimensional position information and three-dimensional velocity information, roll angle, pitch angle and yaw angle of the body, current remaining percentage of power, and communication link quality between the ground station and the unmanned aerial vehicle, which is usually represented in the form of received signal strength indication or signal-to-noise ratio. The three-dimensional position information and three-dimensional velocity information, the roll angle, the pitch angle and the yaw angle of the body, the current remaining percentage of power, and the communication link quality between the ground station and the unmanned aerial vehicle are integrated into a structured data set, i.e. the unmanned aerial vehicle state reference parameter, which is used to determine the unmanned aerial vehicle state reference parameter

[0048]

[0049] wherein P represents a three-dimensional position vector of the unmanned aerial vehicle in a preset coordinate system; V represents a three-dimensional velocity vector thereof; A represents an attitude vector composed of three Euler angles of roll, pitch and yaw; B is a scalar representing the remaining power; and C is a scalar representing the communication link quality. The unmanned aerial vehicle state reference parameter completely describes the instantaneous physical state and communication state of the unmanned aerial vehicle at any time.

[0050] Optionally, the generating of the multi-source quantitative threat data set comprises:

[0051] classifying the multi-source sensing data to generate environment data, body data and airspace data;

[0052] respectively performing risk assessment on the environment data, the body data and the airspace data to generate meteorological threat threat values, equipment failure risk threat values and airspace conflict risk threat values;

[0053] merging the meteorological threat threat values, the equipment failure risk threat values and the airspace conflict risk threat values to generate the multi-source quantitative threat data set.

[0054] ​​Specifically, the multi-source sensing data is classified into three independent data subsets according to its source and physical meaning, to generate environment data, body data and airspace data. The environment data mainly includes information such as wind speed, wind direction, temperature, humidity and atmospheric pressure obtained through meteorological sensors and other devices. The body data mainly covers information from sensors of various systems inside the unmanned aerial vehicle, such as vibration data, voltage and current feedback from the battery management system, and the rotating speed and temperature of each motor of the power system. The airspace data includes the position and speed information of other aircraft in the surrounding area, as well as airspace restriction information such as electronic fences and no-fly zones obtained from ground stations or air traffic control systems. Based on the environment data, body data and airspace data, independent threat quantitative evaluation is carried out. For meteorological threats, the real-time environment data is compared with the preset environment data safety threshold to generate a meteorological threat threat value. For equipment failure risk, the real-time body data is compared with the preset body data safety threshold to generate an equipment failure risk threat value. For airspace conflict risk, the real-time airspace data is compared with the preset airspace data safety threshold to generate an airspace conflict risk threat value. The meteorological threat threat value, equipment failure risk threat value and airspace conflict risk threat value are combined into a structured data set to generate a multi-source quantitative threat data set. For the multi-source quantitative threat data set ,

[0055] ,

[0056] wherein, represents the meteorological threat threat value, represents the equipment failure risk threat value, represents the airspace conflict risk threat value. The size of the threat value directly reflects the severity of the corresponding threat source.

[0057] Optionally, the generation of the meteorological threat threat value, the equipment failure risk threat value and the airspace conflict risk threat value comprises:

[0058] Comparing the real-time parameters in the environment data, body data and airspace data with the preset environment data safety threshold, body data safety threshold and airspace data safety threshold, respectively;

[0059] Calculating the deviation of the real-time parameters from the safety threshold, and quantifying the deviation to generate the meteorological threat threat value, the equipment failure risk threat value and the airspace conflict risk threat value.

[0060] Specifically, a pre-defined safety threshold system is established, including environmental data safety thresholds, such as maximum permissible flight wind speed and upper and lower limits of operating temperature; airframe data safety thresholds, such as maximum permissible motor temperature, minimum safe battery voltage, and maximum airframe vibration amplitude; and airspace data safety thresholds, such as minimum safe separation distance from other aircraft and minimum buffer distance from the boundary of a no-fly zone. Each real-time parameter from the acquired environmental, airframe, and airspace data is compared one by one with its corresponding pre-defined safety threshold. A normalization function is used to calculate the degree of deviation of the real-time parameter from its safety threshold.

[0061] ,

[0062] in, A numerical value representing a specific threat. These are the real-time parameter values ​​corresponding to the threat item, such as real-time wind speed or motor temperature, which are obtained through multi-source sensor data. This is the preset safety threshold for this parameter, that is, the upper or lower limit for normal operation. This is a preset critical threshold, representing the state where the system will be in an extremely dangerous condition once this value is reached or exceeded. Function f is a clamping function that ensures that when... Less than or equal to hour, =0; when Greater than or equal to hour, =1; when When it is in between, It grows between 0 and 1. This process is repeated on all key parameters in environmental, physical, and airspace data to generate meteorological threat values, equipment failure risk values, and airspace conflict risk values.

[0063] Optionally, the generation of the comprehensive flight risk assessment results includes:

[0064] The UAV's own state baseline parameters and the multi-source quantified threat dataset are used as input items. A preset weight allocation fusion model is used to dynamically assign different weight coefficients to the meteorological threat value, the equipment failure risk threat value, and the airspace conflict risk threat value, and then the weighted fusion is performed to generate a comprehensive flight risk assessment result.

[0065] Specifically, the unmanned aerial vehicle state reference parameter and the multi-source quantitative threat data set are standardized, and the maximum and minimum normalization method is used to map them to the dimensionless interval of 0 to 1, forming a standardized state vector and a threat matrix. The standardized unmanned aerial vehicle state reference parameter is constructed into a query vector, which represents the current situation and state of the unmanned aerial vehicle. The standardized multi-source quantitative threat data set is constructed into a key matrix, which represents the identity of each threat. The quantitative values of each threat are constructed into a value matrix, which represents the danger level of each threat. The model calculates the relevance of the query vector and each threat identity in the key matrix to generate an attention score:

[0066] ,

[0067] wherein, is the attention score, is the query vector, K is the key matrix, and T represents the transpose operation of the matrix. The attention score is input into the Softmax function for normalization to generate the assigned weight coefficient. After obtaining the dynamically assigned weight coefficient, weighted summation is performed,

[0068] ,

[0069] wherein, is the final generated comprehensive flight risk assessment result, which is a dimensionless scalar value directly reflecting the overall risk level currently faced by the unmanned aerial vehicle. , and represent the meteorological threat value, equipment failure risk threat value and airspace conflict risk threat value obtained from the multi-source quantitative threat data set. , and are the weight coefficients corresponding to the above three types of threats generated based on the unmanned aerial vehicle state reference parameter, and the sum is set to 1. As shown in Figure 2 is a multi-source threat quantification and dynamic weight allocation relationship curve diagram.

[0070] Optionally, the generation of the adaptive flight control strategy specifically includes:

[0071] comparing the comprehensive flight risk assessment result with a preset risk level threshold to generate a risk disposal level;

[0072] when the risk disposal level indicates that the flight state needs to be adjusted, identifying the specific threat source and specific threat source quantitative parameter that causes the risk assessment result to exceed the corresponding threshold from the original data of the comprehensive flight risk assessment result;

[0073] Based on the identified specific threat source and its quantitative parameters, the corresponding flight control target value is calculated; the flight control target values are combined to form an adaptive flight control strategy.

[0074] Specifically, the comprehensive flight risk assessment result is compared with a set of preset risk level thresholds. The threshold system divides the continuous risk value into several discrete levels, such as safe, attention, warning, and danger. Through comparison, the system can quickly determine the severity of the current risk and generate a corresponding risk disposal level. When the risk disposal level indicates that the flight state needs to be adjusted, that is, when the comprehensive flight risk assessment result exceeds the safe threshold, the multi-source quantitative threat data set is checked, such as the meteorological threat value, the equipment failure risk threat value, and the airspace conflict risk threat value, and their corresponding specific parameters, to identify the specific threat source and its quantitative parameters that cause the overall risk level to rise. The corresponding flight control target value for avoiding or mitigating the threat is calculated. For example, if the identified threat is excessive crosswind, the system will calculate a new target heading angle to adjust the attitude of the UAV to offset the crosswind effect, and calculate a new target speed value to ensure the stability of the flight and the accuracy of the flight path. If the threat is airspace conflict, the system will calculate a target height value or a target heading angle that can increase the safety interval. These calculated target values, such as target heading angle, target height value, and target speed value, are combined to generate an adaptive flight control strategy.

[0075] Optionally, the generation of the dynamic adjustment control instruction set comprises:

[0076] Parsing the adaptive flight control strategy to extract flight control target values;

[0077] Based on the flight control target values, generating the basic control instructions to be executed, wherein the basic control instructions include the setting of target heading angle, target height value, and target speed value;

[0078] Combining the basic control instructions to form a dynamic adjustment control instruction set.

[0079] Specifically, the adaptive flight control strategy is parsed. The target heading angle, target height value, and target speed value are extracted one by one from the data packet. Based on the target heading angle, target height value, and target speed value, the corresponding basic control instructions to be executed are generated in accordance with the communication protocol of the UAV flight control system. For example, if the extracted target heading angle is 90 degrees, the system generates an instruction with the content of setting the target heading angle of the UAV to 90 degrees. All the multiple basic control instructions calculated for the current risk are combined to generate a complete and time-sequenced dynamic adjustment control instruction set.

[0080] Optionally, the generation of the strategy optimization data comprises:

[0081] parsing the dynamic adjustment control instruction set to obtain a specific threat item to be coped with and a threat value of the threat item before execution of the dynamic adjustment control instruction set;

[0082] generating a new threat value based on the real-time acquired multi-source sensing data after execution of the dynamic adjustment control instruction set;

[0083] calculating a change amount of the threat value of the threat item before execution of the dynamic adjustment control instruction set and the new threat value, and generating strategy optimization data.

[0084] Specifically, by parsing the dynamic adjustment control instruction set, the historical data archive is accessed and retrieved to obtain the multi-source quantitative threat data set before decision making, and the threat value corresponding to the specific threat item is extracted from the multi-source quantitative threat data set before decision making. After the unmanned aerial vehicle executes the dynamic adjustment control instruction set and reaches a new stable flight state, the same specific threat item is re-quantitatively evaluated in terms of risk by using the latest multi-source sensing data acquired in real time. A new threat value is generated by using the same safety threshold as before. The new value reflects the severity of the threat item after implementation of the control strategy. The threat value of the threat item before execution of the instruction is compared with the new threat value generated after execution to calculate the change amount therebetween. The change amount is defined as the strategy optimization data of the current control action. The calculation formula can be expressed as

[0085]

[0086] wherein, is the generated strategy optimization data, which is a dimensionless scalar representing the effectiveness of the control strategy. is the threat value of the threat item before execution of the instruction, which is obtained from the historical multi-source quantitative threat data set. is the new threat value of the same threat item calculated based on the new multi-source sensing data after execution of the instruction. Since and are both dimensionless values normalized by the same method, subtraction operation can be directly performed. A positive value indicates that the executed strategy effectively reduces the threat, and the larger the value, the better the effect; on the contrary, a negative value or zero indicates that the strategy is ineffective or even has a counter effect.

[0087] Optionally, the method further comprises:

[0088] adjusting the weight coefficient allocated by the preset weight allocation fusion model based on the strategy optimization data.

[0089] ​Specifically, when a set of dynamic adjustment control instructions is executed and corresponding strategy optimization data is generated, the specific threat type targeted by the adjustment is identified, for example, a meteorological threat. According to the value of the strategy optimization data, the generation rule of the weight coefficient related to the threat type is adjusted:

[0090] ,

[0091] wherein, is a new weight coefficient value to be used when encountering a similar unmanned aerial vehicle state in the future. is an old weight coefficient actually used in this decision. is a preset learning rate, which is a small normal number, and it determines the magnitude of each adjustment to ensure the stability of the learning process. is the strategy optimization data, i.e., the change in the threat value of the threat item before and after the execution of the instructions, which directly quantifies the effectiveness of this strategy. After updating a weight coefficient, all weight coefficients need to be normalized to ensure that their sum is still 1, maintaining the logical consistency of the weighted fusion model.

[0092] Based on the same inventive concept, as shown in Figure 4 , the present application also provides a dynamic threat assessment-based unmanned area unmanned aerial vehicle non-visual flight management and control, which comprises:

[0093] a state monitoring module for real-time acquisition of multi-source sensing data during unmanned aerial vehicle flight to generate unmanned aerial vehicle state reference parameters;

[0094] a threat quantification module for quantifying preset threat sources based on the multi-source sensing data, calculating the offset of the quantized preset threat sources from the preset safety threshold, and merging the offset to generate a multi-source quantized threat data set, wherein the multi-source quantized threat data set includes meteorological threat values, device failure risk threat values, and airspace conflict risk threat values;

[0095] a risk fusion module for inputting the unmanned aerial vehicle state reference parameters and the multi-source quantized threat data set into a preset weight distribution fusion model to generate a comprehensive flight risk assessment result;

[0096] a strategy decision module for calculating the flight control target value of the corresponding threat item based on the comprehensive flight risk assessment result, and combining the flight control target value to generate an adaptive flight management and control strategy;

[0097] an instruction issuing module for generating a set of dynamic adjustment control instructions based on the adaptive flight management and control strategy and issuing them to the unmanned aerial vehicle, setting the unmanned aerial vehicle heading angle, unmanned aerial vehicle height value, and unmanned aerial vehicle speed value;

[0098] The feedback module is configured to execute the threat values of the threat items before and after execution of the dynamic adjustment control instruction set to generate strategy optimization data.

[0099] The adaptive updating module is configured to execute the threat values of the threat items before and after execution of the dynamic adjustment control instruction set to generate strategy optimization data.

[0100] Embodiment 1

[0101] To verify the feasibility of the application in implementation, the application is applied to the unmanned aerial vehicle inspection task of the unmanned area high-voltage power transmission corridor. The task requires the unmanned aerial vehicle to autonomously fly 50 kilometers along the predetermined route under non-visual conditions without manual remote control, and to perform inspection work on the power facilities along the route. The inspection area is located in a mountainous area with variable weather conditions, and there may be other unreported general aviation activities, which poses a serious challenge to the autonomous flight safety of the unmanned aerial vehicle.

[0102] In this embodiment, a quadcopter unmanned aerial vehicle with the code "Inspection-01" is equipped with the flight control system described in the application. The system has completed multiple complete inspection tasks from August to October 2023. During the flight, the system continuously implements the control method described in the application, including: acquiring multi-source sensor data in real time; generating unmanned aerial vehicle state reference parameters and multi-source quantitative threat data sets; fusing the two to generate comprehensive flight risk assessment results; generating adaptive flight control strategies and issuing them for execution; finally, generating strategy optimization data based on the execution effect to adjust the threat assessment model.

[0103] To verify the beneficial effects of the application, the disposal process of several typical risk events is recorded and analyzed. For example, in a flight task on September 5, 2023, "Inspection-01" flew to the middle of the task, and the unmanned aerial vehicle state reference parameter showed that the remaining power was 45%. The communication link quality was good. At this time, the airborne weather sensor detected that the lateral wind speed increased sharply to 15 meters per second, which exceeded the preset environmental data safety threshold of 10 meters per second.

[0104] The system then performs a threat quantification assessment, quantifies the wind speed parameter, and generates a high meteorological threat value. At the same time, since the UAV's own state benchmark parameter shows that the remaining power is low, the system dynamically assigns a higher weight coefficient to the meteorological threat value when generating the comprehensive flight risk assessment result. The comprehensive flight risk assessment result obtained after fusion calculation exceeds the risk threshold of the "warning" level. The system identifies "too large crosswind" as the specific threat source that causes the risk to rise, and calculates and generates new flight control target values, including deflecting the target heading angle by 12 degrees towards the wind direction to offset the crosswind effect, and reducing the target speed value from 10 meters / second to 8 meters / second to ensure flight stability. The flight control target values constitute the adaptive flight management and control strategy, and are parsed into dynamic adjustment control instructions and issued to the UAV.

[0105] After the UAV executes the instructions, its flight path stabilizes. Based on the new sensor data, the system re-calculates the meteorological threat value, which is significantly reduced. The system records the change in threat value before and after the execution of the instructions as strategy optimization data. This data is used to adjust the weight allocation model in subsequent tasks, so that when the system encounters the situation of "low power" and "strong wind" again, it can respond earlier and more decisively.

[0106] Similarly, in another task on September 28, 2023, the UAV's ADS-B receiver detected a fast-approaching recreational helicopter, and the safety separation distance in the airspace data was less than the pre-set airspace data safety threshold. The system immediately generated a high airspace conflict risk threat value, and fused to generate a comprehensive flight risk assessment result reaching the "danger" level. After identifying the specific threat source, the system generated an adaptive flight management and control strategy containing a change in target height value (emergency descent 30 meters) and target heading angle (right turn 90 degrees), and successfully commanded the UAV to execute an avoidance maneuver, avoiding a potential mid-air collision.

[0107] Table 1 Dynamic threat assessment and risk level determination data table

[0108]

[0109] Table 2 Adaptive control strategy generation and execution effect data table

[0110]

[0111] Table 3 System performance comparison data table before and after optimization

[0112]

[0113] From the above Tables 1 to 3, it can be seen that the method described in the present application has excellent autonomous risk management and control capability in practical application.

[0114] Table 1 shows the system's ability to accurately quantify and dynamically assess different types of threats. Whether facing meteorological threats or airspace conflicts, the system can generate reasonable threat values based on real-time data, and through a dynamic weight distribution mechanism, combined with the UAV's own state (such as power), generate a more contextually relevant comprehensive risk assessment result, thereby triggering the corresponding level of disposal plan.

[0115] Table 2 clearly records the decision-making loop from risk identification to strategy generation. The system can "treat the disease", targeting specific threat sources such as "too large crosswind" and "airspace conflict", and calculate and generate adaptive control strategies containing explicit flight control target values. After executing the strategy, the threat values are significantly reduced, and the generated strategy optimization data objectively quantifies the effectiveness of each intervention, providing a basis for subsequent optimization.

[0116] The data in Table 3 powerfully demonstrates the system evolution effect brought by the feedback adjustment mechanism of the invention. After two months of operation and self-learning, the system's response speed and disposal effect in dealing with similar risk events have been greatly improved, with the threat reduction rate increasing from 55% to 80%, and the accuracy of comprehensive risk assessment also significantly improved. These data show that the invention not only achieves effective risk avoidance in a single instance, but also continuously improves the UAV's autonomous survival and task execution capabilities in complex unmanned areas through continuous learning, such as Figure 3 Figure 1 shows the adaptive control strategy execution effect curve.

[0117] It should be noted that the electrical connection between the above-mentioned units does not necessarily represent the direct connection of the line, and the indirect connection mode can also be applied to the embodiments of the present application as long as the purpose of the present application is achieved. The above-mentioned is only an exemplary embodiment of the present application, and cannot limit the scope of the present application.

[0118] That is, any equivalent changes and modifications made in accordance with the teachings of the present invention are still within the scope of the present invention. Other embodiments of the present invention will be readily apparent to those skilled in the art upon considering the specification and practice of the true disclosure. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or conventional techniques in the art not disclosed by the present invention.

Claims

1. A method for evaluating and controlling non-visual flight dynamic threats of unmanned aerial vehicles in uninhabited areas, characterized in that, The method comprises the following steps: Real-time acquisition of multi-source sensing data during unmanned aerial vehicle flight, generation of unmanned aerial vehicle self-state reference parameters, the self-state reference parameters including three-dimensional position, speed, attitude, remaining power and communication link quality of the unmanned aerial vehicle; Based on the multi-source sensing data, the preset threat sources are quantified, the offset of the quantized preset threat sources and the preset safety threshold is calculated, and the offset is merged to generate a multi-source quantized threat data set, wherein the multi-source quantized threat data set includes meteorological threat values, equipment failure risk threat values and airspace conflict risk threat values, and the generation of the multi-source quantized threat data set comprises: Classify the multi-source sensing data to generate environment data, body data and airspace data; risk assessment is performed on the environment data, body data and airspace data respectively to generate meteorological threat values, equipment failure risk threat values and airspace conflict risk threat values, and the threat values are obtained by calculating the deviation of real-time parameters relative to their safety threshold through a normalization function; the meteorological threat values, equipment failure risk threat values and airspace conflict risk threat values are merged to generate a multi-source quantized threat data set; The self-state reference parameters of the unmanned aerial vehicle and the multi-source quantized threat data set are input into a preset weight allocation fusion model to generate a comprehensive flight risk assessment result, which comprises: The self-state reference parameters of the unmanned aerial vehicle and the multi-source quantized threat data set are input into a preset weight allocation fusion model to generate a comprehensive flight risk assessment result, which comprises: The self-state reference parameters of the unmanned aerial vehicle and the multi-source quantized threat data set are input into a preset weight allocation fusion model to generate a comprehensive flight risk assessment result, which comprises: Based on the comprehensive flight risk assessment result, the flight control target value corresponding to the threat item is calculated, the flight control target value is combined to generate an adaptive flight management and control strategy; Based on the adaptive flight management and control strategy, a dynamic adjustment control instruction set is generated and sent to the unmanned aerial vehicle, and the unmanned aerial vehicle heading angle, unmanned aerial vehicle height value and unmanned aerial vehicle speed value are set; Based on the execution effect of the dynamic adjustment control instruction set, the threat values of the threat items before and after execution are generated to generate strategy optimization data, which comprises: The dynamic adjustment control instruction set is parsed to obtain a specific threat item to be coped with and a threat value of the threat item before execution of the dynamic adjustment control instruction set; after execution of the dynamic adjustment control instruction set, a new threat value is generated based on real-time acquisition of multi-source sensing data; and a change amount of the threat value of the threat item before execution of the dynamic adjustment control instruction set and the new threat value is calculated to generate strategy optimization data, wherein the strategy optimization data is obtained by calculating the change amount of the threat value of the threat item before execution of the instruction and the new threat value generated after execution.

2. The method of claim 1, wherein, The generation of the unmanned aerial vehicle self-state reference parameter includes: extracting the three-dimensional position, speed, attitude, remaining power and communication link quality of the unmanned aerial vehicle from the multi-source sensing data; combining the three-dimensional position, speed, attitude, remaining power and communication link quality of the unmanned aerial vehicle to generate the unmanned aerial vehicle self-state reference parameter.

3. The method of claim 1, wherein, The generation of the meteorological threat threat value, equipment failure risk threat value and airspace conflict risk threat value includes: comparing the real-time parameters in the environment data, body data and airspace data with preset environment data safety threshold, body data safety threshold and airspace data safety threshold respectively; calculating the deviation of the real-time parameters from the safety threshold to generate the meteorological threat threat value, equipment failure risk threat value and airspace conflict risk threat value.

4. The method of claim 1, wherein, The generation of the adaptive flight control strategy specifically includes: comparing the comprehensive flight risk assessment result with a preset risk level threshold to generate a risk disposal level; when the risk disposal level indicates that the flight state needs to be adjusted, identifying the specific threat source and specific threat source quantitative parameter from the original data of the comprehensive flight risk assessment result which cause the risk assessment result to exceed the corresponding threshold; based on the identified specific threat source and its quantitative parameter, calculating and generating a corresponding flight control target value; combining the flight control target value to form the adaptive flight control strategy.

5. The method of claim 1, wherein, The generation of the dynamic adjustment control instruction set includes: parsing the adaptive flight control strategy to extract the flight control target value; based on the flight control target value, generating a to-be-executed basic control instruction, wherein the basic control instruction includes the setting of a target heading angle, a target height value and a target speed value; combining the basic control instruction to form the dynamic adjustment control instruction set; calculating the change amount of the threat value of the threat item before execution of the dynamic adjustment control instruction set and the new threat value to generate strategy optimization data.

6. The method of claim 1, wherein, The method further includes feedback adjustment on the generation process of the comprehensive flight risk assessment result: based on the strategy optimization data, adjusting the weight coefficient allocated by the preset weight distribution fusion model.

7. The system for assessing and controlling the non-visual flight dynamic threat of the unmanned aerial vehicle in the unmanned area, applied to the method for assessing and controlling the non-visual flight dynamic threat of the unmanned aerial vehicle in the unmanned area according to any one of claims 1-6, characterized in that, The system includes: a state monitoring module for real-time acquisition of multi-source sensing data in the flight process of the unmanned aerial vehicle and generation of unmanned aerial vehicle self-state reference parameters; a threat quantification module for quantifying preset threat sources based on the multi-source sensing data, calculating the offset of the quantified preset threat sources from preset safety thresholds, and merging the offsets to generate a multi-source quantified threat data set, wherein the multi-source quantified threat data set includes a meteorological threat threat value, an equipment failure risk threat value and an airspace conflict risk threat value; a risk fusion module configured to input the unmanned aerial vehicle self-state reference parameters and the multi-source quantitative threat data set into a preset weight distribution fusion model to generate a comprehensive flight risk assessment result; a strategy decision module configured to calculate flight control target values corresponding to threat items based on the comprehensive flight risk assessment result, and to combine the flight control target values to generate an adaptive flight management and control strategy; an instruction issuing module configured to generate a dynamic adjustment control instruction set based on the adaptive flight management and control strategy, and to issue the dynamic adjustment control instruction set to the unmanned aerial vehicle to set an unmanned aerial vehicle heading angle, an unmanned aerial vehicle height value, and an unmanned aerial vehicle speed value; an execution feedback module configured to generate strategy optimization data based on threat values of threat items before and after execution of the dynamic adjustment control instruction set; an adaptive updating module configured to adjust weight coefficients distributed by the preset weight distribution fusion model based on the strategy optimization data.

Images