An unmanned aerial vehicle complex terrain extreme climate self-adaptive flight control system

Abstract

This invention discloses an adaptive flight control system for unmanned aerial vehicles (UAVs) in complex terrain and extreme weather conditions, belonging to the field of UAV adaptive flight control technology. This adaptive flight control system for UAVs in complex terrain and extreme weather conditions includes a flight data acquisition and preprocessing module, a climate prediction and risk assessment module, an optimal path and task sequence adjustment module, a real-time flight control and path optimization module, and a safety emergency response and recording module. Specifically, the flight data acquisition and preprocessing module acquires weather data; the climate prediction and risk assessment module constructs a neural network model to predict the climate; the optimal path and task sequence adjustment module performs path and task safety assessments; the real-time flight control and path optimization module converts the path and task sequence into aircraft commands; and the safety emergency response and recording module executes emergency strategies and records the data. This system solves the problem in existing technologies where constantly changing weather conditions make it difficult for UAVs to adjust their flight in real time.

Description

Technical Field

[0001] This invention relates to the field of adaptive flight control technology for unmanned aerial vehicles (UAVs), specifically to an adaptive flight control system for UAVs in complex terrain and extreme weather conditions. Background Technology

[0002] In recent years, drones have played an increasingly important role in fields such as power line inspection, geological exploration, forest fire prevention, disaster monitoring, and emergency rescue. Especially in complex terrain areas, drones offer significant advantages over traditional manual methods in terms of speed, efficiency, and cost. However, in practical applications, drones often face challenges from various uncertainties, particularly when complex terrain and extreme weather conditions combine, placing higher demands on their flight control systems.

[0003] Current technologies for climate prediction still have significant shortcomings. They rely heavily on the analysis of real-time data and rule-based judgments for path adjustments, lacking the ability to model and predict climate evolution trends in advance. This makes it difficult to accurately identify extreme weather phenomena such as strong winds, rainfall, or sudden changes in air pressure that may occur in the near future. This short-sighted response mechanism often leaves the system in a passive state when facing sudden climate changes, unable to proactively avoid risk areas or dynamically adjust mission sequences. This reduces flight safety and mission success rates, especially in complex terrain where it is more prone to flight deviations or loss of control.

[0004] Therefore, in order to address the above problems, there is an urgent need for an adaptive flight control system for UAVs in complex terrain and extreme weather conditions. Summary of the Invention

[0005] Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides an extreme weather adaptive flight control system for UAVs in complex terrain, which solves the problem that UAVs are unable to adjust their flight in real time due to constantly changing weather conditions.

[0007] Technical solution

[0008] To achieve the above objectives, this invention provides the following technical solution: an adaptive flight control system for unmanned aerial vehicles (UAVs) in complex terrain under extreme weather conditions, comprising a flight data acquisition and preprocessing module, a climate prediction and risk assessment module, an optimal path and task sequence adjustment module, a real-time flight control and path optimization module, and a safety emergency response and recording module. The flight data acquisition and preprocessing module acquires climate data, ground altitude, and positioning data through climate sensors, a DEM (Digital Elevation Model), and a GPS system, and preprocesses these data. The climate prediction and risk assessment module constructs a neural network model based on the climate data to predict the climate and performs a comprehensive risk assessment, determining whether to change the path and adjust the task sequence. The optimal path and task sequence adjustment module performs path safety and task safety assessments based on the comprehensive risk assessment results, and changes the path and adjusts the task sequence based on the assessment results. The real-time flight control and path optimization module converts the output optimal path and task sequence into aircraft commands, controls the UAV flight in real time, and dynamically updates and adjusts them. The safety emergency response and recording module executes emergency strategies in case of emergencies and records the entire process, ensuring flight safety and traceability.

[0009] Furthermore, the specific steps for acquiring climate data, ground altitude, and positioning data through climate sensors, DEM digital elevation models, and GPS systems are as follows: Climate data is collected in real time using climate sensors, including temperature, humidity, air pressure, wind speed, and wind direction; climate vectors are obtained in real time; the ground altitude of the current flight position is obtained in real time by querying the DEM; and the positioning data of the UAV and the target point location on the path are obtained in real time through the GPS system.

[0010] Further, the specific steps for preprocessing climate data, ground altitude, and positioning data are as follows: Noise removal from climate data: Using Kalman filtering, the noise signal of the climate data is gradually corrected by weighted averaging of measurement errors to remove noise; Outlier removal from climate data: The standard deviation method is used to detect and remove outliers in the climate data. When the climate data deviates from its mean by more than 3 times the standard deviation, it is considered an outlier and removed; Normalization of climate data, ground altitude, and positioning data: Through min-max standardization, climate data, ground altitude, and positioning data of different dimensions are converted into a unified standard range, and the data falls within the [0,1] interval; Calculation of default cruise speed: The flight distance is calculated using the GPS position coordinates of two adjacent time points, and then divided by the time interval between these two time points to obtain the default cruise speed of the UAV.

[0011] Furthermore, the specific steps for predicting climate by constructing a neural network model are as follows:

[0012] A neural network model is constructed using historical climate data and real-time collected temperature, humidity, air pressure, wind speed, and wind direction data. Standardized climate vectors are obtained from historical climate data and input into the model training channel. A time-step sliding window is used to label the climate vectors, generating input data vectors as training targets. The model is trained using neural network optimization algorithms and an accelerated computing platform, and cross-validation and hyperparameter tuning techniques are applied to improve the model's generalization ability and prevent overfitting. The real-time received climate vectors are input into the trained neural network model, and multi-dimensional hidden state vectors are output as climate dynamic perception vectors. A single bias term is obtained by minimizing the loss function using a gradient descent-type optimization algorithm based on historical climate data.

[0013] Further, the specific steps for conducting a comprehensive risk assessment to determine whether to change the path and adjust the task order are as follows: Obtain the climate dynamic perception vector; sum the squares of each hidden dimension of the climate dynamic perception vector and take the square root to obtain the Euclidean norm of the climate dynamic perception vector; sum the Euclidean norm and the single bias term; and input the summation result into the activation function to obtain the comprehensive risk assessment value; compare the comprehensive risk assessment value with the danger threshold in real time; if the comprehensive risk assessment value is less than or equal to the danger threshold, the current path does not need to be adjusted; if the comprehensive risk assessment value is greater than the danger threshold, the current path needs to be adjusted, and the comprehensive risk assessment value is output to the optimal path and task order adjustment module.

[0014] Furthermore, the specific steps for conducting path safety assessment and mission safety assessment based on the comprehensive risk assessment results are as follows: Obtain the Euclidean norm of the climate dynamic sensing vector and the comprehensive risk assessment value of the path segment; multiply the Euclidean norm of the climate dynamic sensing vector by the comprehensive risk assessment value of the path segment, and multiply by the path adjustment factor; add 1 to the result and take the reciprocal to obtain the path safety assessment value; multiply the Euclidean norm of the climate dynamic sensing vector at the current moment by the comprehensive risk assessment value of the path segment, and multiply by the mission adjustment factor; add 1 to the result and take the reciprocal to obtain the mission safety assessment value.

[0015] Furthermore, the specific steps for changing routes and adjusting task order based on the evaluation results are as follows: Calculate the path safety assessment value for all flight paths, select a benchmark value using quicksort, divide the path safety assessment value sequence into two subsequences, one less than and one greater than the benchmark value, recursively sort these two subsequences, and finally merge them to obtain an ordered result. After sorting, select the flight path with the lowest path safety score. Calculate the task safety assessment value for all flight tasks, select a benchmark value using quicksort, divide the task safety assessment value sequence into two subsequences, one less than and one greater than the benchmark value, recursively sort these two subsequences, and finally merge them to obtain an ordered result. After sorting, prioritize executing the task with the highest task safety assessment value.

[0016] Further, the specific steps for converting the output optimal path and task sequence into aircraft commands are as follows: Calculate the required heading setpoint using the heading angle formula based on the UAV's positioning data and the path target point positioning data; multiply the comprehensive risk assessment value by the wind speed adjustment factor, and multiply the difference between 1 and this product by the UAV's default cruise speed to calculate the rate detection value of the current path segment; add the ground elevation of the path point to the basic safe altitude, and perform altitude compensation based on the product of the comprehensive risk assessment value of the path segment and the altitude adjustment factor to calculate the altitude detection value of the current path segment; dynamically convert the heading setpoint, rate detection value, and altitude detection value into aircraft control commands using a PID controller method, and send the control commands to the motor controller to dynamically adjust the aircraft attitude and achieve path adjustment; automatically load the next task target after reaching each task node according to the task sequence, and enter the next segment of flight scheduling logic.

[0017] Furthermore, the specific steps for real-time control and dynamic updating of UAV flight are as follows: During flight, the system continuously collects climate data and calculates the comprehensive risk assessment value in real time; a risk jump threshold is set. If the difference between the comprehensive risk assessment value at the current moment and the comprehensive risk assessment value at the previous moment is greater than the risk jump threshold, the flight control system will send the feedback information back to the optimal path and task sequence selection module, triggering path replanning and task sequence adjustment to form a closed-loop control.

[0018] Furthermore, the specific steps for implementing emergency strategies and recording the entire process in case of emergencies to ensure flight safety and traceability are as follows: Matching corresponding emergency strategies according to the type of anomaly: Emergency return to base: Generate the fastest return path to a safe point; Temporary landing: Select a relatively stable weather area to implement controlled landing; Altitude avoidance: Temporarily increase flight altitude to avoid near-ground wind disturbances; Flight path deviation: Reselect a low-risk area route segment for detour; All emergency paths can also be recalculated to ensure that the emergency plan has minimal risk; Operator notification and log recording: The system automatically records the abnormal situation, and the recorded data includes: time, location, risk type, and emergency measures; Alarm information is fed back to the ground station and operator in voice and text form; At the same time, it is written to the log system for subsequent analysis and model training.

[0019] Beneficial effects

[0020] The present invention has the following beneficial effects:

[0021] (1) This invention integrates climate sensor, DEM and GPS data, and combines neural network to dynamically predict and assess climate conditions, thereby achieving accurate prediction of flight risks in complex terrain and extreme climate environments, thus effectively improving the flight safety and mission completion rate of UAVs in harsh environments.

[0022] (2) This invention achieves intelligent optimization and adjustment of flight paths and mission execution order by dynamically calculating and sorting path safety assessment values ​​and mission safety assessment values, thereby improving the efficiency of multi-mission execution and the rationality of path decision-making.

[0023] (3) This invention constructs a closed-loop feedback mechanism during flight by setting a risk jump threshold, and dynamically triggers path reconstruction and task sequence reordering by combining real-time updated risk assessment values, thereby realizing the adaptive control capability and dynamic response capability of the flight control system.

[0024] (4) This invention, by pre-setting multiple emergency strategies and combining them with real-time risk scoring to select the optimal emergency path, and by automatically recording the entire process information of abnormal events and notifying the operator, realizes the full traceability of flight missions and enhances emergency response capabilities, significantly improving the safety and robustness of the system.

[0025] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0026] Figure 1 A structural diagram of an adaptive flight control system for unmanned aerial vehicles (UAVs) in complex terrain and extreme weather conditions.

[0027] Figure 2 A bar chart showing the path safety assessment values;

[0028] Figure 3 A bar chart showing the mission safety assessment values. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] Please see Figures 1-3This invention provides a technical solution: an adaptive flight control system for unmanned aerial vehicles (UAVs) in complex terrain under extreme weather conditions, comprising a flight data acquisition and preprocessing module, a climate prediction and risk assessment module, an optimal path and task sequence adjustment module, a real-time flight control and path optimization module, and a safety emergency response and recording module. The flight data acquisition and preprocessing module acquires climate data, ground altitude, and positioning data through climate sensors, a DEM (Digital Elevation Model), and a GPS system, and preprocesses these data. The climate prediction and risk assessment module constructs a neural network model based on climate data to predict the climate and performs a comprehensive risk assessment, determining whether to change the path and adjust the task sequence. The optimal path and task sequence adjustment module performs path safety and task safety assessments based on the comprehensive risk assessment results, and changes the path and adjusts the task sequence accordingly. The real-time flight control and path optimization module converts the output optimal path and task sequence into aircraft commands, controls the UAV flight in real time, and dynamically updates and adjusts them. The safety emergency response and recording module executes emergency strategies in case of emergencies and records the entire process, ensuring flight safety and traceability.

[0031] Specifically, the steps for acquiring climate data, ground altitude, and positioning data using climate sensors, a Digital Elevation Model (DEM), and a GPS system are as follows: First, climate data is collected in real time using climate sensors. This data includes temperature, humidity, air pressure, wind speed, and wind direction, covering the main meteorological parameters affecting the UAV's flight status and forming a comprehensive environmental perception dimension. The system integrates and synchronously processes these climate elements through multiple channels, generating a high-dimensional, structured climate vector in real time. This provides high-precision, timely input feature data for subsequent neural network prediction models, ensuring the accuracy and dynamism of climate modeling. Second, by querying the DEM, the ground altitude information of the current flight position is obtained in real time, enabling accurate perception of complex terrain undulations. DEM data, as a static geospatial reference, provides basic terrain constraints for flight path planning and altitude safety assessment, helping to avoid areas of abrupt terrain changes and improving the terrain adaptability and overall safety of the flight trajectory. Finally, the GPS system acquires the UAV's positioning data and path target point positioning data in real time, ensuring high-precision spatial position tracking and dynamic updates during mission execution. This positioning information is not only used to assist in trajectory planning, path correction and attitude adjustment, but also provides key positional information support for the calculation and execution of flight control commands, ensuring the continuity of mission scheduling and the accuracy of the execution process.

[0032] In this implementation scheme, climate data, ground altitude, and positioning data are collected in real time by integrating climate sensors, DEM, and GPS systems. This constructs a multi-dimensional and highly continuous environmental perception data system, enhancing the flight control system's ability to accurately model the flight environment. The acquired climate vectors, terrain altitude, and spatial positioning information provide high-precision and timely data support for subsequent climate prediction, risk assessment, and path optimization, thereby enhancing the system's adaptability and safety under complex terrain and extreme weather conditions.

[0033] Specifically, the preprocessing steps for climate data, ground elevation, and positioning data are as follows:

[0034] Noise removal from climate data: Kalman filtering is used. This algorithm is based on the optimal estimation theory of linear systems. By weighted averaging of the measurement errors of meteorological sensor input data, random noise components in climate data are gradually corrected, resulting in smoother and more reliable meteorological data. This provides a stable data foundation for subsequent model training and control command generation. Outlier removal from climate data: The standard deviation method is used to detect and remove outliers in climate data. That is, the mean and standard deviation of climate data are statistically analyzed. When the climate data at a certain moment deviates from its historical average by more than 3 times the standard deviation, it is identified as an outlier and removed to eliminate the interference of extreme weather fluctuations on the system's judgment and prediction accuracy. Normalization of climate data, ground altitude, and positioning data: Through the min-max normalization method, multi-source data with different physical dimensions and numerical ranges are uniformly processed, and their values ​​are compressed to the [0,1] interval. This normalization process can effectively improve the training efficiency and convergence speed of neural network models when processing heterogeneous data, while avoiding bias problems caused by differences in data scale. Calculate the default cruise speed: Based on the position coordinate data provided by the GPS system at two consecutive time points, the displacement distance of the current flight segment is calculated using the spherical distance formula or the planar Euclidean distance, and this distance is divided by the corresponding time interval to obtain the actual average flight speed of the UAV during that time period, which is used as the default cruise speed parameter to provide basic input values ​​for path planning and rate control.

[0035] In this implementation plan, the collected climate data, ground altitude and positioning data are systematically preprocessed, including using Kalman filtering to remove noise, standard deviation method to remove outliers, min-max normalization to achieve dimensional uniformity, and calculating the default cruise speed based on GPS location information. This effectively improves the quality and consistency of multi-source data, enhances the flight control system's adaptability to environmental changes and the rationality of flight control parameter settings, and provides stable and reliable basic data support for subsequent climate prediction, path planning and flight control.

[0036] Specifically, the steps for predicting climate using a neural network model are as follows: First, a neural network model is constructed using historical climate data and real-time collected temperature, humidity, air pressure, wind speed, and wind direction data, establishing a data-driven prediction framework based on multi-dimensional temporal features. Standardized climate vectors are obtained from historical climate data and input into the model training channel. A sliding window mechanism with a certain time span is used to label the climate vectors, generating input data vectors with temporal features as training targets, thus improving the model's sensitivity to dynamic changes over time. Second, neural network optimization algorithms and accelerated computing platforms are used to further refine the model. The model is trained efficiently, and cross-validation and hyperparameter tuning techniques are applied to improve its generalization ability and enhance its stability under different environmental conditions, effectively preventing overfitting. Then, real-time received climate vectors are input into the trained neural network model, and a multi-dimensional hidden state vector is output through a deep network structure as a high-dimensional, non-linear expression of the current environmental state, i.e., a climate dynamic perception vector. Finally, the loss function is minimized using a gradient descent-type optimization algorithm based on historical climate data, thereby obtaining a single bias term that reflects the overall climate risk trend, further enhancing the accuracy of risk assessment and the interpretability of the model.

[0037] In this implementation plan, a neural network prediction model integrating historical and real-time climate data is constructed. The input data is labeled using a time-step sliding window mechanism, and efficient training is performed using optimization algorithms and accelerated computing platforms. At the same time, cross-validation and hyperparameter tuning strategies are introduced to improve the model's generalization ability. The model can output climate perception vectors with dynamic environmental characteristics and a single bias term in real time, achieving accurate modeling and efficient prediction of climate change trends. This significantly improves the flight control system's sensitivity to extreme weather and its forward-looking response, and enhances the overall intelligence level of flight path planning and risk control.

[0038] Specifically, the steps for conducting a comprehensive risk assessment to determine whether to change the path and adjust the task order are as follows: First, obtain a climate dynamic perception vector as a high-dimensional abstract expression of the current environmental state, where the vector consists of multi-dimensional hidden states output by a neural network model; sum the squares of each hidden dimension of the climate dynamic perception vector and take the square root to calculate its Euclidean norm, which is used to measure the overall intensity of climate change and the degree of environmental disturbance; then, sum the calculated Euclidean norm with a single bias term obtained through optimization using historical data to form a risk baseline indicator that integrates the current state and long-term trends; finally, input the summation result into an activation function. The system completes nonlinear mapping and outputs a comprehensive risk assessment value to enhance the model's ability to express complex weather conditions. Next, it uses Monte Carlo simulation to comprehensively analyze historical flight data and set a hazard threshold. The system compares the comprehensive risk assessment value with the set hazard threshold in real time. When the comprehensive risk assessment value is less than or equal to the hazard threshold, the current path is considered safe and does not require adjustment. When the comprehensive risk assessment value is greater than the hazard threshold, the system determines that the current path has potential risks and triggers a path reconstruction mechanism. The comprehensive risk assessment value is then used as the decision-making basis and output to the optimal path and task order adjustment module, initiating the path and task rearrangement process.

[0039] The specific formula for calculating the comprehensive risk assessment value is as follows:

[0040]

[0041] In the formula, This represents the overall risk assessment value. Represents a dynamic climate sensing vector. Represents the dynamic sensing vector of climate The Euclidean norm, This indicates a single bias term.

[0042] In this implementation scheme, the Euclidean norm of the climate dynamic perception vector output by the neural network is calculated, and a risk assessment index system is constructed by combining it with a single bias term obtained from historical training. Furthermore, a nonlinear activation function is introduced to enhance the risk assessment model's ability to express complex climate conditions, enabling real-time calculation and determination of the comprehensive risk assessment value. By comparing this assessment value with a preset danger threshold, it is possible to dynamically determine whether there is a climate risk on the current path, thereby triggering path adjustment and task rescheduling mechanisms. This effectively improves the intelligent response capability and flight decision-making flexibility of the flight control system under complex terrain and extreme weather conditions.

[0043] Specifically, the steps for conducting path safety assessment and mission safety assessment based on the comprehensive risk assessment results are as follows: First, obtain the Euclidean norm of the climate dynamic sensing vector and the comprehensive risk assessment value corresponding to the path segment. The former reflects the overall intensity of the current environmental disturbance, while the latter reflects the potential risk level of the path area. Then, multiply the Euclidean norm of the climate dynamic sensing vector by the comprehensive risk assessment value of the path segment to form a risk intensity product index, and further multiply it by a path adjustment factor set by the system. The path adjustment factor is obtained through offline evaluation of historical flight data and automatic parameter tuning using a Bayesian optimization algorithm. The value range of the path adjustment factor is 0-1, and this factor is used to adjust the weights of different path terrain complexity, passage importance, etc. The above multiplication... The product result is multiplied by 1 and its reciprocal is taken to eliminate the influence of extreme values, thus obtaining the path safety assessment value. The higher this value, the safer the path, which facilitates subsequent path ranking and optimization. Next, the Euclidean norm of the current climate dynamic perception vector is multiplied again by the comprehensive risk assessment value of the path segment to reflect the mission execution risk level under the current time window. This is then multiplied by a mission adjustment factor, which is obtained through offline evaluation of historical flight data and automatic parameter tuning using a Bayesian optimization algorithm. The path adjustment factor has a value range of 0-1 and is dynamically assigned based on mission type, priority, and completion conditions. The result is also multiplied by 1 and its reciprocal is taken to obtain the mission safety assessment value, providing a reliable risk basis and quantitative support for subsequent mission execution sequence optimization.

[0044] The specific formula for calculating the path security assessment value is as follows:

[0045]

[0046] In the formula, Representing a path The path security assessment value, Represents a dynamic climate sensing vector. Represents the dynamic sensing vector of climate The Euclidean norm, Indicates the path adjustment factor. Representing a path The comprehensive risk assessment value;

[0047] Among them, the Euclidean norm of the climate dynamic perception vector, the comprehensive risk assessment value of the path, and the path adjustment factor represent the numerical values ​​of the impact of climate disturbance, geographical environmental risk, and path weight on the path safety assessment value, respectively, and can be directly called from the above modules when used.

[0048] In this embodiment, the disturbance intensity of path segment A is set to 2.0, the comprehensive risk value is set to 0.3, and the path adjustment factor is set to 0.8; the disturbance intensity of path segment B is set to 1.5, the risk value is set to 0.5, and the adjustment factor is set to 0.6; the disturbance intensity of path segment C is set to 3.0, the comprehensive risk value is set to 0.4, and the path adjustment factor is set to 0.9; the disturbance intensity of path segment D is set to 2.5, the comprehensive risk value is set to 0.6, and the path adjustment factor is set to 0.7; and the disturbance intensity of path segment E is set to 1.0, the comprehensive risk value is set to 0.2, and the path adjustment factor is set to 0.5, as shown in Table 1.

[0049] Table 1. Path Security Assessment Values ​​Data Table

[0050]

[0051] like Figure 2 As shown, this is a bar chart of path security assessment values ​​provided in an embodiment of this application, based on Table 1 and... Figure 2 It can be seen that path segment E has the highest safety assessment value, indicating that its climate disturbance is the weakest, the risk is the lowest, and the overall flight safety is the best. Path segment C has the lowest safety assessment value, mainly because its corresponding disturbance intensity is large, the risk value is high, and the adjustment factor is large, which makes it a high-risk path at the current moment. The system will dynamically sort the flight path priorities according to the score map and give priority to the path segments with high score values ​​and low flight risks for mission scheduling and flight control command generation.

[0052] The specific formula for calculating the mission safety assessment value is as follows:

[0053]

[0054] In the formula, Indicates task Mission safety assessment value, Represents a dynamic climate sensing vector. Represents the dynamic sensing vector of climate The Euclidean norm, Indicates the task adjustment factor. Indicates task The comprehensive risk assessment value.

[0055] Among them, the Euclidean norm of the climate dynamic perception vector, the comprehensive risk assessment value of the task, and the task adjustment factor represent the numerical values ​​of the influence of climate disturbance, geographical environmental risk, and task weight on the task safety assessment value, respectively, and can be directly called from the above modules when used.

[0056] In this embodiment, the disturbance intensity of task number T6 is set to 2.8, the comprehensive risk value of the task is set to 0.35, and the task adjustment factor is set to 0.85; the disturbance intensity of task number T7 is set to 1.1, the risk value is set to 0.53, and the adjustment factor is set to 0.75; the disturbance intensity of task number T8 is set to 2.2, the comprehensive risk value is set to 0.45, and the task adjustment factor is set to 0.65; the disturbance intensity of task number T9 is set to 1.6, the comprehensive risk value is set to 0.25, and the task adjustment factor is set to 0.90; and the disturbance intensity of task number T10 is set to 3.1, the comprehensive risk value is set to 0.50, and the task adjustment factor is set to 0.60, as shown in Table 2.

[0057] Table 2. Task Safety Assessment Value Data Table

[0058]

[0059] like Figure 3 As shown, this is a bar chart of task safety assessment values ​​provided in an embodiment of this application, based on Table 2 and... Figure 3 It can be seen that mission T9 has the highest safety score, mainly because its disturbance intensity and risk value are relatively low. Despite the large adjustment factor, it still obtains the best score. Mission T10 has the lowest score because its climate disturbance intensity is the largest and its risk value is high, resulting in a low safety assessment value. The system will dynamically sort the flight mission priorities according to the score map and prioritize the path segments with high scores and low flight risks for mission scheduling and flight control command generation.

[0060] In this implementation scheme, a safety assessment model is constructed by combining the Euclidean norm of the climate dynamic sensing vector with the comprehensive risk assessment value of the path segment. A path adjustment factor and a task adjustment factor are introduced to weight and correct the assessment results. Furthermore, the interference of extreme values ​​on the assessment results is avoided by adding 1 and taking the reciprocal, thus achieving a quantitative safety assessment of the execution priority of each flight path and task. This method can accurately determine the path risk level and task feasibility based on real-time risk status, providing a quantitative basis for path reconstruction and task reordering, thereby significantly improving the system's flight strategy optimization capability and mission execution safety in highly dynamic weather environments.

[0061] Specifically, the steps for changing routes and adjusting task order based on assessment results are as follows: First, calculate the path safety assessment value for all flight paths to form a path safety value sequence, which is used to measure the relative safety of each path under the current environmental conditions. Then, select a benchmark value using quicksort, and divide the path safety assessment value sequence into two subsequences: one less than the benchmark value and the other greater than it. Recursively perform the quicksort process on each of the two subsequences to ensure sorting efficiency and stability. Finally, merge the sorted subsequences to obtain a complete ordered path safety sequence. Based on the sorting results, select the flight path with the lowest path safety score, indicating the lowest risk level. The path scheme with the lowest risk and highest airworthiness is selected as the preferred target for the current flight path. Next, the mission safety assessment value of all flight missions is calculated to evaluate the feasibility and risk level of each mission in the current environment. Similarly, a benchmark value is selected by quicksort, and the sequence of mission safety assessment values ​​is divided into two subsequences, one less than the benchmark value and the other greater than the benchmark value. These two subsequences are then sorted using a recursive sorting algorithm. After sorting, all results are merged to generate an ordered list of mission safety assessment values. The system prioritizes the mission with the highest mission safety assessment value based on this sorting result, thereby ensuring the safety priority principle and the stability of mission completion during mission scheduling.

[0062] In this implementation scheme, the safety assessment values ​​of all flight paths and missions are rapidly sorted. An efficient recursive partitioning and merging strategy is used to quickly and orderly arrange the assessment sequence. Then, the optimal flight path is selected based on the lowest path safety score, and the priority of missions is determined based on the highest mission safety score. This ensures that path and mission scheduling decisions are dynamically optimized while minimizing risk. This method not only improves the response efficiency of path and mission switching but also enhances the system's global scheduling capabilities and the safety and reliability of flight decisions in multi-path, multi-mission concurrent environments.

[0063] Specifically, the steps for converting the output optimal path and task sequence into aircraft commands are as follows: 1. Calculate the required heading setpoint using the heading angle formula based on the UAV's positioning data and the target point positioning data. 2. Multiply the comprehensive risk assessment value by the wind speed adjustment factor. The wind speed adjustment factor is obtained by regression analysis of the relationship between task effectiveness and risk level in multiple flight simulations, and fitting using a minimum error optimization algorithm. The wind speed adjustment factor ranges from 0 to 1. Multiply the difference between 1 and the product by the UAV's default cruise speed to calculate the rate detection value for the current path segment. 3. Add the ground elevation of the path point to the basic safe altitude, and then... Altitude compensation is achieved by multiplying the comprehensive risk assessment value of the path segment by the altitude adjustment factor. The altitude adjustment factor is obtained by simulating the minimum safe altitude gain required for flight under different risk levels, using regression fitting and fault tolerance analysis algorithms. The altitude adjustment factor ranges from 0 to 1, thereby calculating the altitude detection value of the current path segment. The PID controller method is used to dynamically convert the heading setpoint, rate detection value, and altitude detection value into aircraft control commands. The control commands are sent to the motor controller to dynamically adjust the aircraft attitude and achieve path adjustment. According to the mission sequence, the next mission objective is automatically loaded after each mission node is reached, and the next segment of flight scheduling logic is entered.

[0064] The specific formula for calculating the rate detection value is as follows:

[0065]

[0066] In the formula, Representing a path The rate detection value, This indicates the drone's default cruising speed. Indicates the speed adjustment factor. Representing a path The comprehensive risk assessment value;

[0067] The specific formula for calculating the height detection value is as follows:

[0068]

[0069] In the formula, Representing a path The height detection value, Indicates the ground altitude of the current flight position. Indicates the basic safety height. Indicates a high-level adjustment factor. Representing a path The comprehensive risk assessment value.

[0070] In this implementation scheme, the optimal path and task sequence are transformed into control commands executable by the aircraft. The heading setpoint is calculated using a heading angle formula, and based on a comprehensive risk assessment value combined with wind speed and altitude adjustment factors, the speed and altitude detection values ​​for each path segment are calculated. A PID controller then dynamically converts these parameters into control commands, enabling real-time adjustments to the aircraft's attitude, speed, and altitude. Simultaneously, the system automatically loads subsequent task objectives according to the task execution sequence, ensuring the continuity and autonomy of the flight process. This method effectively enhances the flight control accuracy and task scheduling efficiency of UAVs under complex weather and terrain conditions, improving the overall intelligence level and operational stability of the flight system.

[0071] Specifically, the real-time control and dynamic adjustment of UAV flight are achieved through the following steps: During flight, the system continuously collects climate data, including multi-dimensional environmental parameters such as temperature, humidity, air pressure, wind speed, and wind direction. Combined with positioning and altitude information, a flight environment perception model is constructed in real time, and a comprehensive risk assessment value is calculated based on this data to dynamically reflect the current safety status of the flight environment. The system sets a risk jump threshold as a risk sensitivity criterion. When the difference between the current comprehensive risk assessment value and the previous comprehensive risk assessment value exceeds this risk jump threshold, it is determined that a significant change in the environmental state has occurred. At this time, the flight control system immediately transmits the change information back to the optimal path and task sequence selection module, initiating a path replanning and task sequence adjustment mechanism to achieve information linkage between the flight control layer and the decision-making and scheduling layer. Through the above mechanism, the system forms a closed-loop control logic based on dynamic risk changes, ensuring the continuity and safety of flight missions under complex or sudden weather conditions, and achieving intelligent and autonomous flight strategy adjustment capabilities.

[0072] In this implementation scheme, multi-dimensional climate data is continuously collected during flight, and a comprehensive risk assessment value is calculated in real time. Combined with a set risk abrupt change threshold, a sensitive response mechanism to sudden environmental changes is achieved. When a drastic change in risk is detected, the system can immediately transmit feedback information back to the path and task scheduling module, triggering path replanning and task sequence adjustment, thereby constructing a risk-driven flight control closed loop. This mechanism effectively enhances the flight system's adaptability to dynamic climate changes and its risk response speed, ensuring the safety and stability of flight missions in complex environments.

[0073] Specifically, the steps for implementing emergency strategies and recording the entire process in case of emergencies to ensure flight safety and traceability are as follows: First, match the corresponding emergency strategy according to the anomaly type. The system presets multiple emergency response scenarios and automatically identifies fault or high-risk event types through the anomaly identification module, and calls on the corresponding measures in the emergency strategy library. These include: Emergency return to base: Based on the spatial topology relationship between the current flight position and the safe point, the system generates the fastest return path in real time to ensure that the UAV leaves the danger zone as soon as possible and returns safely; Temporary landing: When the flight risk continues to rise and a return to base is not feasible, the system will select an area with relatively stable weather conditions and minimal terrain undulations to carry out controlled landing, reducing the safety hazards caused by continued flight. Altitude avoidance: When low-altitude wind disturbances or obstacle threats occur, the system instructs the UAV to temporarily increase its flight altitude and detour away from the danger zone, achieving rapid avoidance of near-ground risks; Flight path deviation: If the current path segment is in a high-risk state, the system will reselect a path segment with a lower risk value for detour, achieving dynamic flight path adjustment; All emergency paths can be evaluated by recalculating the path safety score to ensure that the selected emergency plan still has the characteristics of the minimum risk path under emergency response, ensuring the effectiveness and safety of emergency measures; Secondly, in terms of operator notification and log recording: The system automatically records abnormal situations, including the time of the event, the UAV's location, the type of risk detected, and the corresponding emergency response measures;

[0074] Alarm information is promptly fed back to the ground station and remote operators through voice broadcasts and text messages, realizing a human-machine collaborative decision-making mechanism. At the same time, relevant data is completely written into the log recording system, providing high-quality data support for subsequent flight data analysis, system optimization and model retraining, and realizing traceability and closed-loop optimization of the entire flight process.

[0075] This implementation plan establishes a matching mechanism between anomaly types and various emergency strategies. Combined with real-time risk assessment and path reconstruction capabilities, it enables diverse emergency response schemes such as emergency return to base, temporary landing, altitude avoidance, and route deviation. Furthermore, it ensures the minimum risk characteristics of emergency paths by recalculating path safety scores. Simultaneously, the system features automatic recording and feedback of anomalies, simultaneously transmitting the time, location, type, and response measures of risk events to the operator via voice and text, and fully recording them in the log system for subsequent model training and flight behavior retrospection. These mechanisms significantly enhance the safety assurance capabilities of flight missions under sudden risks and the traceability of the system, thereby strengthening the intelligence and reliability of the entire flight control system.

[0076] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0077] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. An adaptive flight control system for unmanned aerial vehicles (UAVs) in complex terrain and extreme weather conditions, characterized in that: It includes modules for flight data acquisition and preprocessing, climate prediction and risk assessment, optimal path and mission sequence adjustment, real-time flight control and path optimization, and safety emergency response and recording. The flight data acquisition and preprocessing module acquires climate data, ground altitude and positioning data through climate sensors, DEM digital elevation model and GPS system, and preprocesses the climate data, ground altitude and positioning data. The climate prediction and risk assessment module uses climate data to build a neural network model to predict the climate and conduct a comprehensive risk assessment to determine whether to change the route and adjust the task order. The specific steps for conducting a comprehensive risk assessment to determine whether to change the path and adjust the task order are as follows: Obtain the climate dynamic perception vector, sum the squares of each hidden dimension of the climate dynamic perception vector and take the square root to obtain the Euclidean norm of the climate dynamic perception vector, sum the Euclidean norm and the single bias term, and then input the summation result into the activation function to obtain the comprehensive risk assessment value. The system compares the comprehensive risk assessment value with the danger threshold in real time. If the comprehensive risk assessment value is less than or equal to the danger threshold, the current path does not need to be adjusted. If the comprehensive risk assessment value is greater than the danger threshold, the current path needs to be adjusted, and the comprehensive risk assessment value is output to the optimal path and task order adjustment module. The optimal path and task order adjustment module performs path safety assessment and task safety assessment based on the comprehensive risk assessment results, and changes the path and adjusts the task order based on the assessment results. The specific steps for changing the path and adjusting the task order based on the evaluation results are as follows: Calculate the path safety assessment value for all flight paths. Select a benchmark value using the quicksort method. Divide the path safety assessment value sequence into two subsequences: one less than the benchmark value and the other greater than the benchmark value. Recursively sort these two subsequences and finally merge them to obtain an ordered result. After sorting, select the flight path with the lowest path safety score. Calculate the mission safety assessment value for all flight missions, select a benchmark value using quicksort, divide the mission safety assessment value sequence into two subsequences that are less than and greater than the benchmark value, recursively sort these two subsequences, and finally merge them to obtain an ordered result. After sorting, the mission with the highest mission safety assessment value is executed first. The real-time flight control and path optimization module converts the output optimal path and task sequence into aircraft commands, controls the UAV flight in real time, and dynamically updates and adjusts them. The safety emergency response and recording module executes emergency strategies and records the entire process in case of emergencies, ensuring flight safety and traceability.

2. The extreme weather adaptive flight control system for unmanned aerial vehicles (UAVs) in complex terrain according to claim 1, characterized in that: The specific steps for acquiring climate data, ground elevation, and positioning data using climate sensors, DEM digital elevation models, and GPS systems are as follows: Climate data is collected in real time through climate sensors, including temperature, humidity, air pressure, wind speed and wind direction, and climate vectors are obtained in real time. By querying the DEM, the ground altitude of the current flight position can be obtained in real time; The system uses GPS to acquire real-time location data of the drone and the location data of the target points along its path.

3. The extreme weather adaptive flight control system for unmanned aerial vehicles (UAVs) in complex terrain according to claim 1, characterized in that: The specific steps for preprocessing climate data, ground elevation, and positioning data are as follows: Noise removal from climate data: Using Kalman filtering technology, the noise signal of the climate data is gradually corrected by weighted averaging of measurement errors, thereby removing noise; Outlier removal from climate data: The standard deviation method is used to detect and remove outliers in climate data. When climate data deviates from its mean by more than 3 times the standard deviation, it is considered an outlier and removed. Climate data, ground elevation and positioning data normalization: Through min-max normalization, climate data, ground elevation and positioning data of different dimensions are converted into a unified standard range and the data fall within the [0,1] interval; Calculate the default cruise speed: Calculate the flight distance using the GPS position coordinates of two adjacent time points, and then divide it by the time interval between these two time points to obtain the drone's default cruise speed.

4. The extreme weather adaptive flight control system for unmanned aerial vehicles (UAVs) in complex terrain according to claim 1, characterized in that: The specific steps for constructing a neural network model to predict climate are as follows: A neural network model is constructed by combining historical climate data with real-time collected temperature, humidity, air pressure, wind speed and wind direction data. Standardized climate vectors are obtained from historical climate data and input into the model training channel. The climate vectors are labeled using a time-step sliding window to generate input data vectors as training targets. The model is trained using neural network optimization algorithms and accelerated computing platforms, and cross-validation and hyperparameter tuning techniques are applied to improve the model's generalization ability and prevent overfitting. The real-time received climate vector is input into the trained neural network model, and the output multi-dimensional hidden state vector is used as the climate dynamic perception vector. A single bias term is obtained by minimizing the loss function using a gradient descent-type optimization algorithm based on historical climate data.

5. The extreme weather adaptive flight control system for unmanned aerial vehicles (UAVs) in complex terrain according to claim 1, characterized in that: The specific steps for conducting path safety assessment and mission safety assessment based on the comprehensive risk assessment results are as follows: Obtain the Euclidean norm of the climate dynamic sensing vector and the comprehensive risk assessment value of the path segment. Multiply the Euclidean norm of the climate dynamic sensing vector by the comprehensive risk assessment value of the path segment, multiply by the path adjustment factor, add 1 to the calculation result and take the reciprocal to obtain the path safety assessment value. Multiply the Euclidean norm of the current climate dynamic perception vector by the comprehensive risk assessment value of the path segment, multiply by the mission adjustment factor, add 1 to the calculation result and take the reciprocal to obtain the mission safety assessment value.

6. The extreme weather adaptive flight control system for unmanned aerial vehicles (UAVs) in complex terrain according to claim 1, characterized in that: The specific steps for converting the output optimal path and task sequence into aircraft commands are as follows: The required heading setting value is calculated using the heading angle formula based on the drone's positioning data and the target point positioning data along the path. Multiply the comprehensive risk assessment value by the wind speed adjustment factor, and multiply the difference between 1 and the product by the drone's default cruise speed to calculate the rate detection value for the current path segment. The ground elevation of the route point is added to the basic safe elevation, and the elevation is compensated based on the product of the comprehensive risk assessment value of the route segment and the elevation adjustment factor, thereby calculating the elevation detection value of the current route segment. The PID controller method is used to dynamically convert the heading setpoint, rate detection value and altitude detection value into aircraft control commands. The control commands are sent to the motor controller to dynamically adjust the aircraft attitude and realize path adjustment. Based on the mission sequence, the next mission objective is automatically loaded after each mission node is reached, and the next segment of flight scheduling logic begins.

7. The extreme weather adaptive flight control system for unmanned aerial vehicles in complex terrain according to claim 1, characterized in that: The specific steps for real-time control and dynamic updating of the drone's flight are as follows: During flight, the system continuously collects climate data and calculates the comprehensive risk assessment value in real time; A risk jump threshold is set. If the difference between the comprehensive risk assessment value at the current moment and the comprehensive risk assessment value at the previous moment is greater than the risk jump threshold, the flight control system will send feedback information back to the optimal path and mission sequence selection module, triggering path replanning and mission sequence adjustment to form a closed-loop control.

8. The extreme weather adaptive flight control system for unmanned aerial vehicles (UAVs) in complex terrain according to claim 1, characterized in that: The specific steps for implementing emergency strategies and recording the entire process in case of emergencies to ensure flight safety and traceability are as follows: Match the appropriate emergency response strategy based on the anomaly type: Emergency Return: Generates the fastest return route to a safe point; Temporary landing: Select a relatively stable climate area to carry out a controlled landing; Altitude avoidance: Temporarily increase flight altitude to avoid near-ground wind disturbance; Flight path deviation: Reselect a low-risk area route to detour; All emergency routes can also be recalculated to ensure that the emergency plan has minimal risk; Operator notifications and log recording: The system will automatically record abnormal situations, and the recorded data will include: time, location, risk type, and emergency measures. Alarm information is fed back to the ground station and operators via voice and text. It is also written to the log system for subsequent analysis and model training.

Images