Intelligent remote sensing unmanned aerial vehicle
By employing a multi-dimensional evaluation module and a dynamic response mechanism, the flight stability and safety issues of remote sensing UAVs in complex environments have been resolved, enabling intelligent flight control and mission execution.
Patent Information
- Application Number
- CN202511669617.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-14
- Publication Date
- 2025-12-19
AI Technical Summary
Existing remote sensing drones have shortcomings in terms of environmental adaptability and flight intelligence, especially in flight stability and safety under the interaction of multiple environmental parameters.
It employs an environmental status assessment module, a power status assessment module, a flight status assessment module, and a flight risk assessment module, combined with a flight speed control module, to adjust the flight speed in real time to cope with complex environments through multi-dimensional assessment and dynamic response mechanisms.
It enhances the autonomous decision-making ability and flight control adaptability of UAVs in complex environments, thereby improving flight safety and mission execution efficiency.
Smart Images
Figure CN121158261A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the technical field of unmanned aerial vehicle manufacturing, and particularly relates to an intelligent remote sensing unmanned aerial vehicle. BACKGROUND
[0002] As an efficient and flexible aerial data acquisition platform, remote sensing unmanned aerial vehicles have been widely used in many fields such as agricultural monitoring, geographic mapping, disaster assessment and urban planning. A typical remote sensing unmanned aerial vehicle system is usually composed of a body structure, a power system (including a motor, a propeller and a battery assembly), a flight control system and a task load (such as a camera device), and its core goal is to safely and reliably complete the predetermined flight task and data acquisition work under complex and variable actual environmental conditions. However, the remote sensing unmanned aerial vehicles under the current technology still face many challenges in environmental adaptability and flight intelligence level:
[0003] The flight control mechanism of most unmanned aerial vehicles, especially the adjustment process of propeller speed and flight speed, mainly relies on pre-set fixed programs or manual operation by operators, and lacks real-time autonomous assessment and dynamic response capability of comprehensive risks in the flight process.
[0004] Specifically, the existing system has obvious limitations in environmental factor processing, and can only respond to a single variable such as wind speed, but cannot effectively integrate the interaction effects of wind strength, GPS signal quality, external electromagnetic interference and precipitation conditions and other environmental parameters, resulting in significant reduction of flight stability and overall safety of unmanned aerial vehicles in severe weather or severe signal interference scenarios.
[0005] Therefore, it is urgent to develop a new control system that can realize comprehensive real-time assessment of flight risks and intelligently optimize flight speed parameters accordingly, so as to improve the autonomous decision-making capability, operation safety level and task execution efficiency of unmanned aerial vehicles under complex working conditions. The existing technology needs to be improved in view of the above problems. SUMMARY
[0006] The purpose of the embodiment of the application is to provide an intelligent remote sensing unmanned aerial vehicle, which aims to solve the problem that the remote sensing unmanned aerial vehicle under the current technology lacks real-time autonomous assessment and dynamic response capability of comprehensive risks in the flight process.
[0007] The application is implemented as follows: an intelligent remote sensing unmanned aerial vehicle, comprising a body, a landing gear fixedly connected to the bottom of the body, four connecting rods equally spaced connected to the body, a plurality of through air holes provided on each connecting rod, a propeller fixedly connected to the end of the connecting rod, the propellers being electrically connected with a battery, the battery being fixed in the body, and a camera provided on one side of the battery; characterized in that it further comprises a rotating speed control system, which comprises:
[0008] environment state evaluation module: it can construct environment state evaluation model according to external wind force, GPS signal quality, external electromagnetic interference intensity and precipitation intensity, and output environment state index;
[0009] power state evaluation module: it can construct power state evaluation model according to current fluctuation of battery, voltage fluctuation, maximum speed difference of each propeller and motor temperature, and output power state index;
[0010] flight state evaluation module: it can construct flight state evaluation model according to vibration intensity of fuselage, attitude control increment, power state index, and output flight state index;
[0011] flight risk evaluation module: it can construct flight risk evaluation model according to environment state index and flight state index under the influence of accuracy of gyroscope and accelerometer of unmanned aerial vehicle, and output flight risk evaluation index;
[0012] flight speed regulation module: it can construct flight speed regulation model according to current flight speed and flight risk evaluation index, and output target flight speed.
[0013] Further technical solutions, actual wind intensity, actual signal-to-noise ratio, actual electromagnetic interference measurement value and actual precipitation intensity are substituted into the maximum-minimum value normalization formula in turn for processing, and wind index, signal-to-noise ratio index, electromagnetic interference intensity index and precipitation intensity index are generated in turn.
[0014] Further technical solutions, in the environment state evaluation model: the environment state index is equal to the weighted sum of wind index, (1-signal-to-noise ratio index), electromagnetic interference intensity index and precipitation intensity index, and each weight coefficient is positive and the sum is one.
[0015] Further technical solutions, the current fluctuation index is obtained by dividing the standard deviation of the current of the battery by the average current; the voltage fluctuation index is obtained by dividing the standard deviation of the voltage of the battery by the average voltage; the speed difference index is obtained by dividing the real-time maximum speed difference of each propeller by the rated speed of the propeller; the motor temperature index is obtained by substituting the actual motor temperature into the maximum-minimum value normalization formula for processing.
[0016] Further technical solutions, in the power state evaluation model:
[0017] The dynamic state index is equal to the weighted sum of the squares of the current fluctuation index, the voltage fluctuation index, and the speed difference index, divided by (the product of the temperature protection coefficient and the temperature loss index, plus 1), and the quotient is then added to the product of the temperature risk coefficient and the square of the motor temperature index. The temperature loss index is obtained by subtracting the motor temperature index from 1, and the weighting coefficients for each fluctuation and speed difference are positive and sum to one.
[0018] A further technical solution involves substituting the actual fuselage vibration acceleration amplitude and the deviation between the attitude control output and the benchmark into the maximum-minimum normalization formula for processing, and then generating the vibration intensity index and attitude control increment index accordingly; in the flight state assessment model:
[0019] The flight state index is equal to the square root of the weighted sum of the square of the vibration intensity index, the square of the attitude control increment index, and the square of the dynamic state index, with each weighting coefficient being a positive number and their sum being one.
[0020] A further technical solution involves dividing the actual gyroscope measurement error standard deviation by the set maximum gyroscope measurement error standard deviation to obtain the gyroscope error index; dividing the actual accelerometer measurement error standard deviation by the set maximum accelerometer measurement error standard deviation to obtain the accelerometer error index; the flight risk assessment model is as follows:
[0021] The flight risk assessment index equals 1 minus the negative value of the natural exponential function, whose exponent is the weighted sum of the environmental state index, flight state index, and flight phase index, multiplied by (the product of sensor error sensitivity and comprehensive sensor error index, plus 1).
[0022] Among them, the flight phase index is calculated from the flight phase coefficient, which corresponds to the four phases of takeoff, cruise, operation, and landing; the comprehensive sensor error index is the arithmetic mean of the gyroscope error index and the accelerometer error index.
[0023] Further technical solutions, in the flight speed control model:
[0024] The target flight speed is equal to the product of the maximum flight speed, the risk reduction factor, and the current speed relative position, plus the product of the minimum flight speed and the speed relative position complement;
[0025] Among them, the risk reduction factor is obtained by subtracting the flight risk assessment index from 1; the current speed relative position is the normalized value of the current speed within the maximum and minimum speed range; the speed relative position complement is obtained by subtracting the relative position from 1.
[0026] A further technical solution is to use the maximum-minimum normalization formula as follows: ;in This is the actual value. To obtain or preset the minimum value, To obtain or preset the maximum value.
[0027] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0028] The environmental status assessment module overcomes the shortcomings of isolated analysis of environmental factors, the dynamic status assessment module solves the problem of fragmented dynamic system status monitoring, the flight status assessment module avoids the deficiency of flight status assessment being detached from internal dynamic status, the flight risk assessment module makes up for the lack of correlation between sensor accuracy and flight phase, and the flight speed control module overcomes the limitations of rigid control strategies. This forms a closed-loop control mechanism from risk perception to speed response, effectively improving the UAV's autonomous decision-making ability in complex scenarios such as wind changes, signal interference, and equipment aging.
[0029] This application effectively solves the problem of inaccurate fusion of multi-source flight state parameters due to differences in dimensions, enabling the vibration intensity index and attitude control increment index to participate in the evaluation under a unified dimension. This significantly improves the calculation accuracy of the flight state index, provides a reliable quantitative basis for subsequent flight risk assessment, and enhances the flight control adaptability and mission execution safety of UAVs in complex environments.
[0030] This application can quantify the dynamic amplification effect of sensor errors on flight risks in real time, avoiding misjudgment of risks due to decreased sensor accuracy in complex weather environments or critical takeoff / landing phases, thereby improving the accuracy of flight risk assessment and the reliability of autonomous decision-making. Attached Figure Description
[0031] Figure 1 This is a schematic diagram of the speed control system in this invention;
[0032] Figure 2 This is a schematic diagram showing the connection between the fuselage and each propeller in this invention;
[0033] Figure 3 This is a schematic diagram of the propeller structure in this invention;
[0034] Figure 4 This is a schematic diagram of the connection between the connecting rod and the propeller in this invention.
[0035] In the attached diagram: 1. Fuselage; 2. Landing gear; 3. Connecting rod; 4. Propeller; 5. Camera; 6. Battery; 7. Air vent. Detailed Implementation
[0036] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0037] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.
[0038] like Figures 1-4 As shown, an intelligent remote sensing drone according to an embodiment of the present invention includes a fuselage 1, a landing gear 2 fixedly connected to the bottom of the fuselage 1, four connecting rods 3 evenly spaced on the fuselage 1, each connecting rod 3 having multiple through air holes 7, a propeller 4 fixedly connected to the end of each connecting rod 3, the propellers 4 being electrically connected to a battery 6, the battery 6 being fixed in the fuselage 1, and a camera 5 disposed on one side of the battery 6; characterized in that it further includes a rotation speed control system, which includes:
[0039] Environmental status assessment module: It can build an environmental status assessment model based on external wind force, GPS signal quality, external electromagnetic interference intensity and precipitation intensity, and output environmental status index;
[0040] Power status assessment module: It can construct a power status assessment model based on the current fluctuation and voltage fluctuation of battery 6, the maximum speed difference of each propeller 4, and the motor temperature, and output the power status index.
[0041] Flight status assessment module: It can construct a flight status assessment model based on the vibration intensity, attitude control increment, and dynamic state index of fuselage 1, and output the flight status index;
[0042] Flight Risk Assessment Module: It can construct a flight risk assessment model and output a flight risk assessment index based on environmental state index and flight state index, combined with the current flight stage, under the influence of the accuracy of the UAV's gyroscope and accelerometer.
[0043] Flight speed control module: It can construct a flight speed control model based on the current flight speed and flight risk assessment index, and output the target flight speed.
[0044] In this embodiment, the environmental status assessment module refers to the unit that constructs an environmental status assessment model and outputs an environmental status index based on external wind force, GPS signal quality, external electromagnetic interference intensity, and precipitation intensity. It can be implemented using a fuzzy logic-based reasoning system, for example, by classifying and judging each environmental parameter through a membership function, or by using a support vector machine model to train historical environmental data to generate assessment results. Its main purpose is to achieve a quantitative representation of the coupled influence of multiple environmental factors.
[0045] Furthermore, the power status assessment module refers to the unit that constructs a power status assessment model and outputs a power status index based on the current fluctuations and voltage fluctuations of the battery 6, the maximum speed difference of each propeller 4, and the motor temperature. Specifically, the fluctuation index can be calculated using a sliding window statistical method, such as using the moving standard deviation to analyze the instantaneous changes in current and voltage, or using anomaly detection algorithms to identify the deviation trend of power system parameters. Its main purpose is to achieve real-time monitoring of the internal power health status of the UAV.
[0046] The flight status assessment module refers to the unit that constructs a flight status assessment model and outputs a flight status index based on the vibration intensity, attitude control increment, and dynamic state index of the fuselage 1. As a preferred implementation, a Kalman filter can be used to fuse multi-source sensor data for state estimation, such as combining vibration accelerometer and attitude angular rate information, or analyzing the characteristic frequencies of the vibration signal of the fuselage 1 based on spectrum analysis technology. Its main purpose is to achieve a comprehensive judgment on the stability of the flight process.
[0047] The flight risk assessment module refers to the unit that constructs a flight risk assessment model and outputs a flight risk assessment index based on the environmental state index and flight state index, combined with the current flight stage, under the influence of the accuracy of the UAV's gyroscope and accelerometer. Furthermore, it can use a Bayesian network model to integrate risk factors, such as using sensor accuracy as a conditional probability parameter, or using time series prediction methods to assess the dynamic evolution of the risk index. Its main purpose is to achieve real-time quantitative assessment of flight risks.
[0048] The flight speed control module is a unit that constructs a flight speed control model based on the current flight speed and flight risk assessment index and outputs the target flight speed. In practical applications, a proportional-integral-derivative control strategy can be used to dynamically adjust the speed. For example, the speed adjustment range can be linearly mapped according to the risk index, or the boundary value of the speed range can be calculated through optimization algorithms. Its main purpose is to achieve adaptive adjustment of flight speed.
[0049] Therefore, by operating the above modules independently and transferring data, this application avoids the limitations of existing technologies that rely solely on a single environmental factor or preset program for regulation, and achieves multi-dimensional risk assessment of the external environment, internal dynamic state and flight process, thereby dynamically balancing flight safety and mission efficiency under complex conditions.
[0050] In a preferred embodiment of the present invention, the actual wind force intensity, actual signal-to-noise ratio, actual electromagnetic interference measurement value, and actual precipitation intensity are sequentially substituted into the maximum-minimum normalization formula for processing, and the wind force index, signal-to-noise ratio index, electromagnetic interference intensity index, and precipitation intensity index are generated sequentially; the environmental state assessment model is as follows:
[0051] ;
[0052] in As for wind power weight, For signal quality weights, For electromagnetic interference intensity weighting, As a weight for precipitation intensity, ,and , , as well as All greater than , , , as well as It can be set by fitting expert experience or simulation experimental data; Wind force index The signal-to-noise ratio is denoted as . Electromagnetic interference intensity index, The precipitation intensity index. This is an environmental status index.
[0053] In this embodiment, the environmental state assessment model refers to a mathematical framework that quantifies and integrates multi-dimensional environmental parameters into a single assessment index. This model can be implemented using linear weighted combinations or nonlinear mapping functions, aiming to establish a quantifiable correlation between environmental factors and flight risks. Wind weight refers to a coefficient characterizing the degree of influence of wind factors on the environmental state. It can be dynamically adjusted based on historical wind speed-accident rate data using machine learning algorithms, aiming to flexibly allocate risk contribution according to actual meteorological conditions. Signal quality weight refers to a coefficient reflecting the importance of GPS signal stability. It can be adaptively calculated by combining the number of visible satellites and the signal attenuation rate, aiming to avoid erroneous underestimation of risk when signal quality improves. Electromagnetic interference intensity weight refers to a coefficient quantifying the priority of electromagnetic interference risks. It can be determined based on the real-time frequency band occupancy rate of spectrum monitoring equipment, aiming to ensure the independent identifiability of electromagnetic interference factors in the comprehensive assessment. Precipitation intensity weight refers to a coefficient characterizing the impact of precipitation. The severity coefficient can be used to classify and set the risk level by linking meteorological forecast data with the waterproof rating of equipment, aiming to compensate for the deficiency that precipitation factors are often ignored in traditional schemes; the wind force index refers to the standardized actual wind force value, which can be generated by smoothing wind speed sensor data using the sliding window statistical method, aiming to eliminate the interference of instantaneous fluctuations on the assessment results; the signal-to-noise ratio index refers to the normalized GPS signal quality value, which can be calculated based on the carrier phase noise power spectral density, aiming to directly convert the degree of poor signal quality into risk contribution; the electromagnetic interference intensity index refers to the quantitative characterization value of external electromagnetic interference, which can be determined by integrating the energy of a spectrum analyzer in a specific frequency band, aiming to achieve a linear mapping of electromagnetic interference risk; the precipitation intensity index refers to the standardized value of the actual precipitation severity, which can be comprehensively judged by combining rain sensor and visibility detection data, aiming to explicitly incorporate precipitation risk into the model.
[0054] Specifically, the scheme in this application generates the first contribution by multiplying the wind force index by the wind force weight. Since the wind force index directly quantifies the actual wind intensity, the stronger the wind, the more significant its negative impact on flight stability. This design ensures that the contribution of the wind force factor is linearly amplified; by converting the signal-to-noise ratio index to... The second contribution is generated because the signal-to-noise ratio (SNR) index represents the GPS signal quality; a larger G value indicates a more stable signal, but the risk is inversely proportional to the signal quality. This inverse vectorization method avoids the risk being incorrectly underestimated when the signal quality improves. The third contribution is generated by multiplying the electromagnetic interference intensity index by the electromagnetic interference intensity weight. Stronger electromagnetic interference is more likely to cause control system failure, and this design achieves a linear mapping of electromagnetic interference risk. The fourth contribution is generated by multiplying the precipitation intensity index by the precipitation intensity weight. Stronger precipitation is more likely to cause equipment failure, and this design explicitly incorporates precipitation risk into the model. At the same time, the constraint that all weights are greater than 0 and sum to 1 ensures that the contributions of multiple environmental factors are normalized. The adjustability of the weights allows for dynamic allocation of relative importance according to the actual scenario. For example, in a strong wind environment, the wind weight can be increased to highlight the wind-dominated risk. The environmental state index, as the comprehensive output of the model, integrates multi-dimensional environmental parameters into a single quantitative indicator, providing comparable and traceable input for subsequent flight risk assessment.
[0055] As a specific implementation method, the environmental status assessment module of this application is implemented as follows: The environmental status assessment module can specifically use a microcontroller based on the ARM Cortex-M7 core as the processing unit. This processing unit is connected to a wind speed sensor, a GPS module, an electromagnetic spectrum analyzer, and a rain sensor via an I²C bus. The wind speed sensor collects three-dimensional wind speed data in real time, and after sliding window filtering, it is input into a maximum-minimum value normalization module to generate a wind force index. The carrier-to-noise ratio data output by the GPS module is logarithmically transformed to generate a signal-to-noise ratio index. The integrated energy of the electromagnetic spectrum analyzer in the 2.4GHz band is compared with a threshold to generate an electromagnetic interference intensity index. The pulse count of the rain sensor is linearly interpolated to generate a precipitation intensity index. The processing unit calls a pre-stored weight configuration table, which dynamically loads different weight combinations according to the meteorological characteristics of the flight area. For example, for urban areas, different weight combinations are used. =0.35、 =0.25、 =0.25、 =0.15, while the coastal area was adjusted to =0.45、 =0.20、 =0.15、 =0.20; The processing unit executes the environmental status assessment model calculation, multiplies each index by its corresponding weight, sums them, and outputs the environmental status index in the range of 0-1 to the flight risk assessment module.
[0056] Through the above technical solution, this application realizes a quantitative and comprehensive assessment of multiple environmental factors such as wind force, GPS signal quality, electromagnetic interference intensity, and precipitation intensity. It solves the problem of considering only one environmental factor, so that flight risk assessment no longer relies on isolated parameters or experience judgment, but is based on the overall interaction of environmental factors. This significantly improves the accuracy and adaptability of environmental status assessment, thus providing a reliable quantitative basis for the safe flight of UAVs in complex weather or poor signal environments.
[0057] In a preferred embodiment of the present invention, the current fluctuation index is obtained by dividing the standard deviation of the current of the statistical battery 6 by the average current; the voltage fluctuation index is obtained by dividing the standard deviation of the voltage of the statistical battery 6 by the average voltage; the speed difference index is obtained by dividing the maximum speed difference of each propeller 4 in real time by the rated speed of the propeller 4; the actual temperature of the motor is substituted into the maximum-minimum normalization formula for processing, and the motor temperature index is generated; the power state evaluation model is as follows:
[0058] ;
[0059] in Weighting for current fluctuations, As voltage fluctuation weight, Weighting for the propeller speed difference 4 , as well as All greater than , ,, , as well as It can be set by fitting expert experience or simulation experimental data. For current fluctuation index, This is the voltage fluctuation index. This is the speed difference index. This refers to the motor temperature index. For temperature protection coefficient, Temperature risk factor, This is the dynamic state index.
[0060] In this embodiment, current fluctuation weighting This refers to the parameter used to adjust the contribution ratio of the current fluctuation index in the dynamic state assessment. It can be implemented using the output value of a neural network trained on historical flight data or a fixed threshold set by an expert system. Its purpose is to ensure that the impact of current fluctuations on system stability is reasonably quantified; voltage fluctuation weight. This refers to the parameter used to adjust the contribution ratio of the voltage fluctuation index. It can adopt a dynamic adjustment mechanism similar to the current fluctuation weight or a preset empirical coefficient. Its purpose is to highlight the key role of voltage stability in the health of the power system; propeller speed difference weight. This refers to the parameter used to adjust the contribution ratio of the speed difference between the four propellers. It can adopt a numerical value based on adaptive adjustment during flight phases or a fixed weight allocation strategy. Its purpose is to reflect the impact of power output consistency on flight safety; current fluctuation index. This refers to a normalized index characterizing the degree of current fluctuation in battery 6. It can be a normalized value of the standard deviation calculated based on a sliding time window or a fluctuation feature extracted through wavelet transform. Its purpose is to transform raw current data into a comparable quantitative indicator; voltage fluctuation index. This refers to a normalized index characterizing the voltage fluctuation of battery 6. It can be obtained from voltage signal frequency domain analysis or a normalized value based on the standard deviation of a moving average. Its purpose is to quantify voltage stability; Speed difference index. This refers to a normalized index characterizing the maximum speed difference among the four propellers. It can be a normalized value representing the deviation between the real-time speed difference and the historical average, or a speed consistency assessment value based on multi-sensor fusion. Its purpose is to evaluate the balance of power output; motor temperature index. This refers to a normalized index characterizing the temperature state of a motor. It can be a standardized value output from a temperature sensor or a normalized temperature value predicted based on a thermodynamic model. Its purpose is to reflect the impact of thermal management on system reliability; temperature protection coefficient. This refers to a parameter used to adjust the degree of influence of temperature on the basic risk component. It can be an empirical coefficient dynamically adjusted according to the ambient temperature or a preset constant based on the thermal failure threshold. Its purpose is to amplify the risk of fluctuating parameters at high temperatures; temperature risk coefficient. It refers to a parameter used to quantify the independent risk of temperature itself. It can be a fixed value or a dynamically adjusted value based on the characteristics of motor materials. Its purpose is to directly assess the threat of the cumulative effect of temperature to the power system.
[0061] Specifically, the proposed solution generates a basic risk component by weighted sum of squares of current fluctuation index, voltage fluctuation index, and speed difference index. The weighting coefficients ensure a reasonable distribution of the contribution of each parameter, while the squaring operation highlights the impact of significant fluctuations on system stability. The denominator is dynamically adjusted based on the motor temperature index and temperature protection coefficient. When the motor temperature rises, the temperature index increases, leading to a decrease in the denominator value, thereby amplifying the basic risk component. This reflects the physical characteristic that the danger of current, voltage, and speed fluctuations is significantly enhanced under high-temperature conditions. Simultaneously, the model generates additional risk terms based on the motor temperature index, directly quantifying the independent risk of high temperature to the power system through temperature risk coefficients and squaring operations. This design not only achieves deep integration of multi-source power parameters but also enables the model to adaptively reflect subtle changes in equipment aging or the early stages of failure through a nonlinear temperature feedback mechanism.
[0062] As a specific implementation method, the solution of this application is implemented as follows: The dynamic state assessment model is executed by the central processing unit of the UAV. The central processing unit specifically adopts a microcontroller based on the ARM Cortex-M7 core. It collects data from the battery 6 current, voltage, propeller 4 speed and motor temperature sensors in real time through embedded software. It calculates the current fluctuation index and voltage fluctuation index using the sliding window algorithm, obtains the speed difference index using multi-sensor fusion technology, and generates the motor temperature index based on the thermodynamic model. Finally, it outputs the dynamic state index according to the preset weight coefficients and model parameters.
[0063] Through the above scheme, this application realizes a comprehensive quantitative evaluation of multi-source parameters of the power system, which can effectively identify the coupled effects of abnormal fluctuations and temperature changes in the early stage of equipment aging or failure, thereby significantly improving the ability to predict potential risks of the power system and providing reliable protection for flight safety.
[0064] In a preferred embodiment of the present invention, the vibration acceleration amplitude of the actual fuselage 1 and the deviation between the attitude control output and the reference are sequentially substituted into the maximum-minimum normalization formula for processing, and the vibration intensity index and attitude control increment index are generated sequentially; the flight state evaluation model is as follows:
[0065] ;
[0066] in For vibration weights, Incremental weights for attitude control For dynamic state weights, , , as well as All greater than , , as well as It can be set by fitting expert experience or simulation experimental data. The vibration intensity index, For attitude control incremental index. The dynamic state index, This is the flight status index.
[0067] In this embodiment, the maximum-minimum normalization formula refers to converting the original measured values into a standardized index with uniform dimensions. This can be achieved through linear transformation or piecewise function mapping, aiming to eliminate the input scale imbalance caused by unit differences in different physical quantities. The vibration intensity index can be understood as a quantitative indicator characterizing the severity of structural vibration of fuselage 1. It can be generated by extracting feature values from the time-domain signal collected by the accelerometer through spectral analysis, aiming to convert the vibration acceleration amplitude into a dimensionless evaluation parameter. The attitude control increment index is a quantitative indicator reflecting the stability of the flight control system. It can be obtained by statistical processing based on the deviation sequence between the PID controller output and the theoretical reference value, aiming to quantify the dynamic response characteristics during attitude adjustment. The flight state evaluation model can be understood as a multi-parameter fusion weighted evaluation mechanism, which is constructed using the square root of the sum of squares, aiming to enhance the influence of high-risk states and constrain the output range through nonlinear calculations.
[0068] Specifically, the proposed solution first normalizes the vibration acceleration amplitude and attitude control increment of the fuselage 1, mapping the original measured values to the [0,1] interval to form the vibration intensity index and attitude control increment index. This process eliminates the dimensional difference between the vibration acceleration amplitude (unit: m / s²) and the attitude control deviation (unit: angle or torque). Subsequently, the normalized vibration intensity index, attitude control increment index, and dynamic state index are input into the flight state assessment model. The three information are fused through a weighted sum of squares operation, where the weight coefficients dynamically allocate the contribution of each parameter according to the flight conditions. Since the model uses a square term to amplify the impact of abnormal states, when the vibration intensity index or attitude control increment index increases significantly, its square value will accelerate the increase of the flight state index, making the assessment results more sensitive to high-risk conditions such as severe vibration or attitude loss of control. The square root operation ensures that the output value remains within a reasonable range, avoiding assessment distortion caused by excessive amplification of the index, thereby forming a precise quantification of the flight state.
[0069] As a specific implementation method, the solution of this application is implemented as follows: The vibration acceleration amplitude of the fuselage 1 is collected in real time by MEMS accelerometers installed at key nodes of the frame, and the attitude control increment is obtained by comparing the deviation between the control surface deflection command output by the flight control system and the theoretical reference value; the normalization process is executed in the UAV main control microprocessor, and the vibration acceleration amplitude and attitude control increment are respectively substituted into the preset normalization formula, wherein the minimum and maximum values are determined according to the statistics of historical flight data; the generated vibration intensity index and attitude control increment index are input together with the dynamic state index into the flight state evaluation model, and the weight coefficients are dynamically adjusted by the flight phase. For example, the vibration weight is increased in the cruise phase to cope with airflow disturbance, and the attitude control increment weight is increased in the operation phase to ensure camera stability.
[0070] In a preferred embodiment of the present invention, the gyroscope error index is obtained by dividing the actual gyroscope measurement error standard deviation by the set maximum gyroscope measurement error standard deviation; the accelerometer error index is obtained by dividing the actual accelerometer measurement error standard deviation by the set maximum accelerometer measurement error standard deviation; the flight risk assessment model is as follows:
[0071] ;
[0072] in For environmental state weights, For flight state weights, Weights for each flight phase. , as well as All greater than , , , as well as The settings can be configured based on expert experience or simulation data. This is an environmental status index. For flight status index, For flight phase index, , For flight phase coefficients, , The values, from smallest to largest, correspond to the four phases: takeoff, cruise, operation, and landing. To comprehensively measure the sensor error index, , This refers to the gyroscope error index. For accelerometer error index, For sensor error sensitivity, This is a flight risk assessment index.
[0073] In this embodiment, the gyroscope error index refers to an indicator that quantifies the loss of gyroscope measurement accuracy through a normalization mechanism. It can be achieved by calculating the ratio of the real-time measurement error standard deviation to a preset maximum error threshold. The purpose is to transform discrete error data into a standardized continuous index, facilitating the fusion assessment of multiple risk factors. The accelerometer error index can be understood as a normalized parameter characterizing the accelerometer's accuracy state. It can be achieved by dividing the actual measurement error standard deviation by a set maximum measurement error standard deviation. The purpose is to ensure the comparability of error assessments for different sensors and avoid assessments caused by differences in sensor type. The overall sensor error index is a unified index that integrates the errors of gyroscopes and accelerometers. It can be calculated by averaging the two error indices using an arithmetic mean method. The purpose is to balance the interference of individual sensor anomalies on the overall assessment, thereby comprehensively reflecting the collaborative accuracy level of the sensor system. The flight phase index can be understood as a parameter that maps discrete flight phases to continuous risk weights. It can be achieved by linearly transforming the flight phase coefficient by subtracting one and then dividing by three. The purpose is to achieve a gradual transition of risk weights in each phase of takeoff, cruise, operation, and landing, accurately adapting to the differences in sensor accuracy sensitivity in different phases.
[0074] Specifically, the proposed solution normalizes the standard deviations of the real-time measurement errors of the gyroscope and accelerometer into error indices, generating a comprehensive sensor error index as a dynamic amplification factor. This amplification factor is multiplied by the basic risk component (composed of a weighted fusion of the environmental state index, flight state index, and flight phase index) and then input into an exponential function to calculate the flight risk assessment index. Since the basic risk component reflects the combined influence of the external environment and flight state, and the amplification factor is adjusted in real time based on the sensor error index, when the sensor accuracy decreases, the amplification factor increases, leading to a non-linear increase in the risk index, thereby ensuring that the risk assessment results can dynamically respond to changes in sensor performance. At the same time, the flight phase index is processed continuously to ensure a smooth transition of risk weights between different flight phases, avoiding abrupt changes in risk assessment during phase switching.
[0075] As a preferred embodiment, the solution of this application is specifically implemented as follows: The flight risk assessment module can be implemented by a microcontroller with an ARM Cortex-M7 architecture. The microcontroller receives the raw output signals of the gyroscope and accelerometer in real time, calculates the standard deviation of the measurement error through software algorithms, and performs normalization processing based on the pre-stored maximum error threshold to obtain the error index; the comprehensive sensor error index is generated by the microcontroller's internal arithmetic unit performing an average calculation; the flight risk assessment model is implemented by the microcontroller calling the index calculation program, wherein the environmental state index, flight state index, and flight stage coefficient are synchronously input by other modules of the speed control system, and the sensor error sensitivity coefficient is dynamically configured according to the current meteorological conditions.
[0076] As a preferred embodiment of the present invention, the flight speed control model is as follows:
[0077] ;
[0078] in The maximum permissible flight speed for the drone. The minimum permissible flight speed for a drone. This represents the relative position of the current speed within the speed range. , At the current flight speed, The target flight speed.
[0079] In this embodiment, the target flight speed This refers to the desired flight speed value calculated by the flight control system, which can be calculated in real time using a digital signal processor. Its purpose is to provide a dynamic speed control reference for the flight control system; maximum flight speed. This refers to the maximum speed limit that a drone can achieve under safe conditions. It can be preset based on the drone's aerodynamic performance and propulsion system parameters, with the aim of setting a speed upper limit to prevent structural overload; minimum flight speed. This refers to the minimum speed threshold required for a drone to maintain stable flight. It can be determined based on aerodynamic characteristics and aims to avoid the risk of stall. Current speed relative to position. This refers to the normalized position of the current flight speed within the range of minimum and maximum speeds. It can be calculated from real-time speed measurements and aims to characterize the proportion of the current speed state within the feasible range; current flight speed This refers to the actual flight speed of the drone, which can be obtained through airspeed monitoring or GPS data, and its purpose is to provide speed feedback; Flight Risk Assessment Index It refers to the numerical value that quantifies the comprehensive risk of flight. It comes from the output of the flight risk assessment module and aims to reflect the comprehensive risk level of the environment and flight status.
[0080] Specifically, the solution proposed in this application uses a flight risk assessment index. Position relative to current speed Combined, the target's flight speed is dynamically adjusted. .because Directly quantify the comprehensive risk level of the external environment and flight status, based on Constructed This feature establishes an inverse relationship between target flight speed and risk level, automatically lowering the speed limit when risk increases; simultaneously... Precisely characterizing the proportion of the current velocity within the feasible range, and its relationship with... Synergistic effect, making Item weights It significantly enhances performance when risk is low and current speed is high, prioritizing maintaining an efficient cruise state. Item weights It automatically increases when the current speed is low or the risk is high, ensuring a safety margin. Furthermore, the model... and The continuous adjustment mechanism avoids sudden drops or increases in speed caused by sudden changes in risk, thereby ensuring flight stability while achieving a smooth speed transition.
[0081] As a specific implementation method, the solution of this application is implemented as follows: the flight speed control module is implemented using an ARM Cortex-M7 microcontroller, which periodically receives flight risk assessment indices from the flight risk assessment module. and current flight speed from airspeed sensor Calculate the current velocity relative to position Based on the above model, the target flight speed is output in real time. The information is sent to the flight control system, which adjusts the motor speed accordingly to achieve the target flight speed.
[0082] Through the above scheme, this application can maintain a high flight speed in low-risk environments to improve the efficiency of remote sensing data acquisition, while automatically reducing the flight speed in high-risk environments to enhance flight safety, effectively solving the problem of efficiency and safety imbalance caused by rigid control strategies.
[0083] As a preferred embodiment of the present invention, the maximum-minimum value normalization formula is: ;in This is the actual value. To obtain or preset the minimum value, To obtain or preset the maximum value.
[0084] In this embodiment, the maximum-minimum normalization formula is a mathematical processing method that maps actual measured parameters to a standardized interval. It can be implemented using linear transformation, aiming to eliminate dimensional differences between different physical quantities and ensure the comparability of multi-source parameters in the risk assessment model. It refers to the dimensionless exponent value after normalization, which can be understood as the relative position of the actual parameter within a preset range, with the purpose of quantifying the sensitivity of parameter changes. It refers to the real-time measurement parameters of the external environment or system status. These can be actual measured values of physical quantities such as wind intensity, current fluctuations, and motor temperature, with the aim of providing normalized raw input. This refers to the lower boundary value of the normalization process, which can be determined through historical data statistics or preset safety thresholds. The purpose is to define the minimum reasonable range of parameters. This refers to the upper boundary value of the normalization process, which can also be determined through historical data statistics or preset safety thresholds. The purpose is to define the maximum reasonable range of the parameters.
[0085] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. An intelligent remote sensing drone, comprising a fuselage, a landing gear fixedly connected to the bottom of the fuselage, four connecting rods equally spaced along the fuselage, each connecting rod having multiple through-holes, a propeller fixedly connected to the end of each connecting rod, the propellers being electrically connected to a battery, the battery being fixed within the fuselage, and a camera disposed on one side of the battery; characterized in that, It also includes a speed control system, which comprises: Environmental status assessment module: It can build an environmental status assessment model based on external wind force, GPS signal quality, external electromagnetic interference intensity and precipitation intensity, and output environmental status index; Power condition assessment module: It can construct a power condition assessment model based on the battery current fluctuation, voltage fluctuation, maximum speed difference of each propeller and motor temperature, and output the power condition index. Flight status assessment module: It can construct a flight status assessment model based on the vibration intensity of the fuselage, attitude control increment, and dynamic state index, and output the flight status index; Flight Risk Assessment Module: It can construct a flight risk assessment model and output a flight risk assessment index based on environmental state index and flight state index, combined with the current flight stage, under the influence of the accuracy of the UAV's gyroscope and accelerometer. Flight speed control module: It can construct a flight speed control model based on the current flight speed and flight risk assessment index, and output the target flight speed.
2. The intelligent remote sensing UAV according to claim 1, characterized in that, The actual wind force intensity, actual signal-to-noise ratio, actual electromagnetic interference measurement value, and actual precipitation intensity are successively substituted into the maximum-minimum normalization formula for processing, and the wind force index, signal-to-noise ratio index, electromagnetic interference intensity index, and precipitation intensity index are generated in sequence.
3. The intelligent remote sensing UAV according to claim 2, characterized in that, In the environmental status assessment model: the environmental status index is equal to the weighted sum of the wind index, (1 - signal-to-noise ratio index), electromagnetic interference intensity index, and precipitation intensity index, with each weight coefficient being a positive number and their sum being one.
4. The intelligent remote sensing UAV according to claim 1, characterized in that, The current fluctuation index is obtained by dividing the standard deviation of the current of the statistical battery by the average current; the voltage fluctuation index is obtained by dividing the standard deviation of the voltage of the statistical battery by the average voltage. The speed difference index is obtained by dividing the real-time maximum speed difference of each propeller by the rated speed of the propeller. The actual motor temperature is substituted into the maximum-minimum normalization formula for processing, and a motor temperature index is generated.
5. The intelligent remote sensing UAV according to claim 4, characterized in that, In the dynamic state assessment model: The dynamic state index is equal to the weighted sum of the squares of the current fluctuation index, the voltage fluctuation index, and the speed difference index, divided by (the product of the temperature protection coefficient and the temperature loss index, plus 1), and the quotient is then added to the product of the temperature risk coefficient and the square of the motor temperature index. The temperature loss index is obtained by subtracting the motor temperature index from 1, and the weighting coefficients for each fluctuation and speed difference are positive and sum to one.
6. The intelligent remote sensing UAV according to claim 1, characterized in that, The actual vibration acceleration amplitude of the fuselage and the deviation between the attitude control output and the benchmark are successively substituted into the maximum-minimum normalization formula for processing, and the vibration intensity index and attitude control increment index are generated accordingly; in the flight state assessment model: The flight state index is equal to the square root of the weighted sum of the square of the vibration intensity index, the square of the attitude control increment index, and the square of the dynamic state index, with each weighting coefficient being a positive number and their sum being one.
7. The intelligent remote sensing UAV according to claim 1, characterized in that, The gyroscope error index is obtained by dividing the actual standard deviation of the gyroscope measurement error by the set maximum standard deviation of the gyroscope measurement error; the accelerometer error index is obtained by dividing the actual standard deviation of the accelerometer measurement error by the set maximum standard deviation of the accelerometer measurement error; the flight risk assessment model is as follows: The flight risk assessment index equals 1 minus the negative value of the natural exponential function, whose exponent is the weighted sum of the environmental state index, flight state index, and flight phase index, multiplied by (the product of sensor error sensitivity and comprehensive sensor error index, plus 1). Among them, the flight phase index is calculated from the flight phase coefficient, which corresponds to the four phases of takeoff, cruise, operation, and landing; the comprehensive sensor error index is the arithmetic mean of the gyroscope error index and the accelerometer error index.
8. The intelligent remote sensing UAV according to claim 1, characterized in that, In the flight speed control model: The target flight speed is equal to the product of the maximum flight speed, the risk reduction factor, and the current speed relative position, plus the product of the minimum flight speed and the speed relative position complement; Among them, the risk reduction factor is obtained by subtracting the flight risk assessment index from 1; the current speed relative position is the normalized value of the current speed within the maximum and minimum speed range; the speed relative position complement is obtained by subtracting the relative position from 1.