Multi-sensor height prediction system fusing LSTM (Long Short Term Memory) and Kalman filtering

Abstract

The invention discloses a multi-sensor height prediction system fusing LSTM and Kalman filtering, which relates to the technical field of multi-sensor height prediction and comprises an aircraft height monitoring and data acquisition module, a phase mismatch prediction and classification module, a self-adaptive weight adjustment module, an adjustment mechanism feedback evaluation module and a flight process monitoring and optimization module. The phase mismatch prediction and classification module analyzes the obtained prediction input information, predicts the phase mismatch degree between the prediction results of the LSTM model and the Kalman filter, and divides the prediction results into three classes of phase consistency, mild mismatch and serious mismatch according to the prediction results; and the self-adaptive weight adjustment module constructs a self-adaptive weight adjustment mechanism according to a division result, and performs different weight adjustment measures on three types of conditions of phase consistency, mild mismatch and serious mismatch. According to the invention, the problem of phase mismatch of height prediction of the aircraft under sudden environment change is solved, and real-time response and long-term stability of height prediction are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-sensor height prediction, and in particular to a multi-sensor height prediction system integrating LSTM and Kalman filtering. Background Art

[0002] In recent years, with the rapid development of aerospace technology, the safe flight of aircraft has put forward higher requirements for the accurate acquisition of flight altitude data. In order to ensure the safe and efficient operation of aircraft in complex airspace environments, the commonly used altitude measurement methods currently include global navigation satellite system (GNSS), inertial navigation system (INS) and barometric altimeter. GNSS can provide high-precision three-dimensional positioning and altitude information, inertial navigation system relies on gyroscopes and accelerometers to achieve autonomous navigation, and barometric altimeter measures the altitude by measuring atmospheric pressure. Various types of sensors have their own advantages in different flight scenarios. Through data fusion technology, integrating multi-source sensor data to improve the accuracy and robustness of altitude measurement has become an important research direction in the aviation field. In recent years, deep learning algorithms, especially long short-term memory neural networks (LSTM), have performed well in time series data processing, and Kalman filters, as classic data fusion algorithms, can effectively filter out noise in dynamic systems. Therefore, combining LSTM neural networks with Kalman filters to predict flight altitude by multi-sensor data fusion is becoming an important technical means to improve the reliability and stability of aviation flight data.

[0003] The existing multi-sensor altitude prediction technology that integrates LSTM and Kalman filtering can achieve accurate prediction of aircraft altitude by integrating data from the global navigation satellite system (GNSS), inertial navigation system (INS) and barometric altimeter. First, the LSTM neural network is used to process the time series data of flight altitude. By learning from historical data, a nonlinear mapping relationship between input features and altitude output is established to predict the altitude data at the next moment. Then, the LSTM prediction results are combined with real-time sensor data from GNSS, INS and barometer, and input into the Kalman filter for data fusion. During the fusion process, the Kalman filter takes the measurement values ​​of each sensor as observation input, and iteratively updates based on the system state model. By calculating the Kalman gain, the weights of different data sources are dynamically adjusted to obtain a more stable and accurate altitude estimate. The entire process effectively utilizes the prediction ability of LSTM for time series and the advantages of Kalman filtering for noise suppression to ensure the accuracy and reliability of aircraft altitude information in various complex environments.

[0004] The prior art has the following deficiencies: When an aircraft encounters sudden airflow or drastic changes in the environment, there will be sharp fluctuations in altitude data. In the existing fusion LSTM and Kalman filter system, the LSTM model relies on long-term series data for trend prediction, while the Kalman filter relies on short-term data for error correction. Therefore, when the environment changes suddenly, the LSTM long-sequence prediction based on historical data will lag behind the actual situation, while the Kalman filter attempts to correct the current altitude data through rapid adjustment, resulting in phase inconsistency between the two prediction results. This phase mismatch will cause the system to be unable to effectively balance the long-term trend prediction of LSTM and the short-term correction of Kalman filtering in a short period of time, causing the altitude prediction results to oscillate frequently in a short period of time and be difficult to stabilize. However, the existing technology cannot effectively deal with this sudden phase mismatch problem due to the failure to quickly adjust the weight distribution between LSTM and Kalman filtering, resulting in a response delay in the altitude control system when the aircraft descends or lands rapidly, thereby increasing the risk of altitude control failure during landing, which in turn causes the aircraft to land unsteadily, increase energy consumption and affect flight safety.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a multi-sensor height prediction system integrating LSTM and Kalman filtering to solve the problems in the above-mentioned background technology.

[0007] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: a multi-sensor altitude prediction system integrating LSTM and Kalman filtering, comprising an aircraft altitude monitoring and data acquisition module, a phase mismatch prediction and classification module, an adaptive weight adjustment module, an adjustment mechanism feedback evaluation module, and a flight process monitoring and optimization module; The aircraft altitude monitoring and data acquisition module monitors the aircraft altitude data in real time during the flight of the aircraft, and obtains the prediction input information required for generating LSTM model and Kalman filter prediction when a sharp fluctuation in the aircraft altitude data is detected; The phase mismatch prediction and classification module analyzes the obtained prediction input information, predicts the phase mismatch degree between the prediction results of the LSTM model and the Kalman filter, and divides them into three categories according to the prediction results: phase consistency, mild mismatch, and severe mismatch; The adaptive weight adjustment module builds an adaptive weight adjustment mechanism based on the division results, and performs different weight adjustment measures for the three situations of phase consistency, mild mismatch and severe mismatch; The adjustment mechanism feedback evaluation module obtains the adjustment mechanism feedback information in real time during the process of the adaptive weight adjustment mechanism adjusting the weight, and analyzes it after obtaining it to evaluate whether the adjustment degree of the adaptive weight adjustment mechanism can meet expectations, and optimizes the adaptive weight adjustment mechanism according to the evaluation results; The flight process monitoring and optimization module monitors and analyzes the altitude prediction data and adjustment records during the entire flight process in real time, and continuously optimizes the adaptive weight adjustment mechanism based on the monitoring results to ensure the altitude prediction stability and control accuracy of the aircraft in different flight phases and environmental conditions.

[0008] Preferably, in the phase mismatch prediction and classification module, the acquired prediction input information is preprocessed, and the prediction time synchronization information and prediction error comparison information in the preprocessed prediction input information are extracted, and analyzed after extraction to generate a prediction time deviation index and a height prediction error synchronization index, respectively; a phase mismatch evaluation model is constructed for the generated prediction time deviation index and the height prediction error synchronization index, a phase mismatch coefficient is generated, and the generated phase mismatch coefficient is compared with a pre-set phase mismatch coefficient threshold interval, and the phase mismatch degree between the prediction results of the LSTM model and the Kalman filter is predicted according to the comparison results, and it is divided into three categories according to the prediction results: phase consistency, mild mismatch and severe mismatch.

[0009] Preferably, the acquisition logic of the prediction time deviation index and the height prediction error synchronization index is as follows: Extract the prediction time synchronization information from the preprocessed prediction input information, including the timestamps of the LSTM model generating prediction results at different times within a period of time, the timestamps of the Kalman filter generating prediction results, and the timestamps of actually obtaining the aircraft altitude data, and mark them as , and , Indicates that within a period of time The timestamp of the moment when the LSTM model generates the prediction result, Indicates that within a period of time The timestamp of the moment the Kalman filter generates the prediction result, Indicates that within a period of time The timestamp of the actual acquisition of the aircraft altitude data at the moment, , is a positive integer; Calculate the forecast time deviation index. The specific calculation formula is as follows: In the formula, is the forecast time deviation index; Extract the prediction error comparison information from the preprocessed prediction input information, including the altitude prediction value of the LSTM model at different times in a period of time, the altitude prediction value of the Kalman filter, and the actual altitude of the aircraft obtained, and calibrate them as , and , Indicates that within a period of time The height prediction value of the LSTM model at the moment, Indicates that within a period of time The height prediction value of the Kalman filter at time instant, Indicates that within a period of time The actual altitude of the aircraft obtained at all times; Calculate the height prediction error synchronization index. The specific calculation formula is as follows: In the formula, is the height forecast error synchronization index.

[0010] Preferably, the generated prediction time deviation index and height forecast error synchronicity index Construct a phase mismatch evaluation model and generate a phase mismatch coefficient by weighted summation , and the generated phase mismatch coefficient The phase mismatch coefficient threshold interval is set A comparison is made, and the phase mismatch degree between the prediction results of the LSTM model and the Kalman filter is predicted based on the comparison results. The prediction results are divided into three categories: phase consistency, mild mismatch, and severe mismatch. The specific comparison analysis and classification are as follows: like , if the phase mismatch between the prediction results of the LSTM model and the Kalman filter is low, the situation is classified as phase consistent; like , the phase mismatch between the prediction results of the LSTM model and the Kalman filter is moderate, then the situation is classified as mild mismatch; like , if the phase mismatch between the prediction results of the LSTM model and the Kalman filter is high, the situation is classified as severe mismatch.

[0011] Preferably, in the adaptive weight adjustment module, an adaptive weight adjustment mechanism is constructed according to the division results, specifically: according to the division results of phase consistency, mild mismatch and severe mismatch, different weight adjustment parameters are set, corresponding to the weight allocation of the LSTM model and the Kalman filter in the prediction fusion, respectively, to form an adaptive weight adjustment mechanism; the adjustment mechanism is based on the range of the phase mismatch coefficient, and automatically determines the weight allocation strategy and adjustment range between the LSTM model and the Kalman filter through pre-set rules; Different weight adjustment measures are taken for the three situations of phase consistency, mild mismatch and severe mismatch. Specifically, in the case of phase consistency, a balanced weight strategy is adopted to assign equal weights to the LSTM model and the Kalman filter to maintain stable prediction; in the case of mild mismatch, the weight of the Kalman filter is adjusted to be higher than that of the LSTM model to increase the influence of short-term correction in the prediction, so that the system can maintain a balance between long-term and short-term predictions while enhancing its responsiveness to current environmental changes; in the case of severe mismatch, the weight of the Kalman filter is significantly increased, and the weight of the LSTM model is significantly reduced to ensure that the system can respond quickly to sudden changes and enhance the short-term stability of highly predicted results.

[0012] Preferably, in the adjustment mechanism feedback evaluation module, after the adjustment mechanism feedback information is acquired in real time, it is preprocessed, and the instantaneous prediction adjustment feedback information and the long-term adjustment cumulative feedback information in the preprocessed adjustment mechanism feedback information are extracted, and analyzed after extraction to generate an instant adjustment deviation index and a cumulative adjustment effect coefficient, respectively; an adjustment degree evaluation model is constructed for the generated instant adjustment deviation index and cumulative adjustment effect coefficient, an adjustment coefficient is generated, and the generated adjustment coefficient is compared with a pre-set adjustment coefficient threshold, and whether the adjustment degree of the adaptive weight adjustment mechanism can meet expectations is evaluated based on the comparison results, and the adaptive weight adjustment mechanism is optimized based on the evaluation results.

[0013] Preferably, the acquisition logic of the instant adjustment deviation index and the cumulative adjustment effect coefficient is as follows: The instantaneous prediction adjustment feedback information in the preprocessed adjustment mechanism feedback information is extracted, specifically including the difference between the predicted altitude of the LSTM model and Kalman filter combination and the actual altitude of the aircraft at different times within a period of time during the process of the adaptive weight adjustment mechanism adjusting the weight, the weight difference between the LSTM model and the Kalman filter, and the time interval of the system responding to the change of the aircraft altitude after the weight adjustment, and calibrated as , and , Indicates that during the period of time when the adaptive weight adjustment mechanism adjusts the weight The difference between the predicted altitude of the LSTM model and Kalman filter combination and the actual altitude of the aircraft at that moment, Indicates that during the period of time when the adaptive weight adjustment mechanism adjusts the weight The weight difference between the LSTM model and the Kalman filter at this moment, It represents the time interval for the system to respond to the change of aircraft altitude after weight adjustment. , is a positive integer; Calculate the instant adjustment deviation index. The specific calculation formula is as follows: In the formula, To adjust the deviation index instantly; The long-term adjustment cumulative feedback information in the preprocessed adjustment mechanism feedback information is extracted, specifically including the cumulative prediction error change after multiple weight adjustments within the time window in which the adaptive weight adjustment mechanism adjusts the weight, the cumulative amplitude of all weight adjustments within the time window, and the time length within the time window, and they are marked as , and , It represents the cumulative change in prediction error after multiple weight adjustments within the time window in which the adaptive weight adjustment mechanism adjusts the weight. Represents the cumulative magnitude of all weight adjustments within the time window, Indicates the length of time within the time window; Calculate the cumulative adjustment effect coefficient. The specific calculation formula is as follows: In the formula, It is the cumulative adjustment effect coefficient.

[0014] Preferably, the generated instant adjustment deviation index and the cumulative adjusted effect coefficient Construct an adjustment degree assessment model and generate adjustment coefficients through weighted summation , and the resulting adjustment factor With the preset adjustment factor threshold Compare and evaluate whether the adjustment degree of the adaptive weight adjustment mechanism can meet expectations based on the comparison results, and optimize the adaptive weight adjustment mechanism based on the evaluation results. The specific comparison and analysis are as follows: like , the adjustment degree of the adaptive weight adjustment mechanism can meet expectations, and there is no need to optimize the adaptive weight adjustment mechanism; like , the adjustment degree of the adaptive weight adjustment mechanism cannot meet expectations, and the adaptive weight adjustment mechanism needs to be optimized, including: increasing the weight adjustment frequency of the LSTM model and the Kalman filter to improve the system's response speed to environmental changes; resetting the upper limit of the weight adjustment range to ensure the short-term response effect during the adjustment process; optimizing the weight adjustment parameters, dynamically adjusting the weight ratio of the LSTM model and the Kalman filter to adapt to the current flight status requirements; combining historical data and real-time feedback information, analyzing the prediction error and response time, automatically adjusting the system threshold setting, and improving the system's adaptability and accuracy.

[0015] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. The present invention combines the long-term trend prediction of the LSTM model with the short-term error correction of the Kalman filter, so that the system can quickly adjust the prediction weight under sudden environmental changes. This design is particularly suitable for the altitude prediction needs of aircraft when the airflow changes drastically or the environment changes suddenly, ensuring that the weight distribution between the LSTM and the Kalman filter is dynamically adjusted with environmental changes, alleviating the problem of phase mismatch. The specific "aircraft altitude monitoring and data acquisition module" captures sensor data in real time and uses the data as prediction input information, so that the system can perceive altitude fluctuations in a timely manner and quickly make weight adjustment decisions, thereby significantly improving the real-time and accuracy of the prediction.

[0016] 2. The present invention realizes a comprehensive evaluation and dynamic feedback of the system adjustment process through the combination of the "phase mismatch prediction and classification module" and the "adaptive weight adjustment module". The instant adjustment deviation index and the cumulative adjustment effect coefficient provide feedback on the adjustment effect at both the short-term and long-term levels. The adjustment coefficient is generated by weighted summation and compared with the preset threshold to evaluate the overall adjustment effect of the system. This multi-level feedback mechanism ensures that no matter what kind of altitude fluctuation occurs during the flight, the system can effectively monitor and continuously optimize the adjustment mechanism based on the feedback. In this way, not only the stability of the aircraft altitude prediction is guaranteed, but also the response sensitivity and long-term stability of the system are improved, providing a solid technical guarantee for flight control safety.

[0017] 3. The present invention further enhances the self-adaptability and fault tolerance of the system through an automatic optimization and adjustment mechanism. Specific optimization measures include increasing the frequency of weight adjustment, resetting the upper limit of weight adjustment, and adjusting threshold settings based on historical data and real-time feedback. Such an adaptive optimization design enables the system to improve the accuracy of prediction and the stability of control by dynamically adjusting parameters in changing flight environments, such as rapid climbs, landings, and encountering sudden airflows. This intelligent optimization process greatly improves the applicability of the system in different flight scenarios, enabling it to maintain high prediction stability in complex and changing environments, reduce the risk of altitude control failure, and reduce energy consumption, thereby improving flight safety and the overall performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0019] Figure 1 This is a module schematic diagram of a multi-sensor height prediction system that integrates LSTM and Kalman filtering in the present invention. DETAILED DESCRIPTION

[0020] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0021] The present invention provides Figure 1 The multi-sensor altitude prediction system integrating LSTM and Kalman filtering shown in the figure includes an aircraft altitude monitoring and data acquisition module, a phase mismatch prediction and classification module, an adaptive weight adjustment module, an adjustment mechanism feedback evaluation module, and a flight process monitoring and optimization module; The aircraft altitude monitoring and data acquisition module monitors the aircraft altitude data in real time during the flight, and obtains the prediction input information required for generating LSTM model and Kalman filter prediction when the aircraft altitude data is detected to fluctuate sharply. The input information includes multiple sensor data, such as global navigation satellite system (GNSS), inertial navigation system (INS) and barometric altimeter data; During the flight of an aircraft, real-time monitoring of the aircraft's altitude data can be achieved through multiple sensors on the aircraft. The Global Navigation Satellite System (GNSS) can provide accurate ground positioning information to obtain the aircraft's altitude data. The Inertial Navigation System (INS) calculates the altitude changes of the aircraft through sensors such as accelerometers and gyroscopes, especially in areas where satellite signals cannot be received (such as tunnels or urban environments with dense high-rise buildings). It can still provide continuous altitude data. In addition, the barometric altimeter calculates the relative altitude of the aircraft by measuring changes in air pressure. The software system can regularly collect altitude data output by the sensor through the sensor data interface (API), and store this data in memory or transmit it to the central processing unit in real time. The system can determine the frequency of data collection based on timestamps or sampling frequencies, usually at a frequency of milliseconds, to ensure that the altitude data reflects the status of the aircraft in real time.

[0022] Detecting sharp fluctuations in aircraft altitude data can be achieved by calculating the rate of change of altitude data. The software system analyzes the altitude data collected each time and calculates the difference between adjacent sampling points. If the altitude change exceeds the preset threshold within a certain period of time, it means that there is a sharp fluctuation. For example, if the altitude data of the aircraft changes by more than the set safe fluctuation range (such as a percentage of a certain altitude range or a specific altitude value) in a short period of time (such as within 1 second), it is considered to have a sharp fluctuation. The software can use a sliding window algorithm to analyze the altitude changes within a certain time window. If the average rate of change or standard deviation within the window exceeds the predetermined range, the alarm mechanism is triggered. In this way, the system can monitor the altitude changes of the aircraft in real time and quickly detect the sharp altitude fluctuations caused by sudden airflow, wind speed changes or other emergencies.

[0023] When the system detects a sharp fluctuation in the aircraft's altitude data, the software can automatically retrieve and combine data from multiple sensors to generate the prediction input information required by the LSTM model and Kalman filter. This data includes the aircraft's real-time altitude data (such as data provided by GNSS and INS), flight speed (obtained through INS or external sensors), flight direction (through trajectory information from the inertial measurement unit IMU or GPS), and air pressure information (provided by the barometric altimeter). When a sharp fluctuation occurs, the system will construct a multidimensional feature vector based on the real-time collected sensor data as the input of the LSTM and Kalman filter. This input information can be standardized and normalized through the data preprocessing module to ensure the accuracy and applicability of the data. The software system can input this data into the LSTM and Kalman filter models for further prediction and error correction to ensure the stability of the aircraft's altitude prediction in complex environments.

[0024] The purpose of this is to accurately capture the altitude fluctuations of the aircraft when the environment changes suddenly, and to adjust the prediction mechanism in time to deal with the phase mismatch problem between LSTM and Kalman filter. When the aircraft encounters sudden airflow or drastic changes in the environment, the LSTM model relies on long-term series data for trend prediction, while the Kalman filter performs error correction based on short-term data. Due to the hysteresis of LSTM prediction, it may conflict with the short-term adjustment of Kalman filter, resulting in phase mismatch, which in turn causes frequent oscillation and instability of altitude prediction. By monitoring the altitude data of the aircraft in real time, detecting sharp fluctuations and obtaining prediction input information in time, the system can quickly analyze and identify the degree of phase mismatch when the aircraft encounters sudden environmental changes, thereby adjusting the weight distribution between LSTM and Kalman filter, avoiding frequent fluctuations or lags in prediction results, and ensuring the accuracy of aircraft altitude control and flight stability. This solution can dynamically adapt to the real-time status of the aircraft, ensuring that the altitude prediction of the aircraft under sudden changes will not be affected by lag or mismatch, thereby greatly improving the safety and control accuracy of the aircraft.

[0025] The phase mismatch prediction and classification module analyzes the obtained prediction input information, predicts the phase mismatch degree between the prediction results of the LSTM model and the Kalman filter, and divides them into three categories according to the prediction results: phase consistency, mild mismatch, and severe mismatch; In this embodiment, in the phase mismatch prediction and classification module, the acquired prediction input information is preprocessed, and the prediction time synchronization information and prediction error comparison information in the preprocessed prediction input information are extracted, and analyzed after extraction to generate a prediction time deviation index and a height prediction error synchronization index, respectively; a phase mismatch evaluation model is constructed for the generated prediction time deviation index and the height prediction error synchronization index, and a phase mismatch coefficient is generated, and the generated phase mismatch coefficient is compared with a pre-set phase mismatch coefficient threshold interval, and the phase mismatch degree between the prediction results of the LSTM model and the Kalman filter is predicted according to the comparison results, and it is divided into three categories according to the prediction results: phase consistency, mild mismatch and severe mismatch.

[0026] In the phase mismatch prediction and classification module, the obtained prediction input information is preprocessed to ensure the quality and consistency of the input data, thereby improving the accuracy and reliability of the subsequent phase mismatch assessment. First, preprocessing can eliminate noise and outliers to avoid data errors caused by sensor accuracy limitations or environmental interference affecting the analysis results; second, preprocessing can standardize the scale of the data so that data from different sources can be analyzed under the same dimension to ensure the uniformity of the model processing. In the specific preprocessing process, denoising can be performed first, such as using moving average filtering or low-pass filtering methods to smooth the fluctuations of height data and eliminate high-frequency noise. Then, outlier detection is performed, such as using Z scores or box plots to identify and eliminate abnormal height values ​​that exceed the set threshold range to prevent extreme data from interfering with the results. Finally, normalization is performed to scale the data of different sensors to a uniform range (for example, between 0 and 1) according to a certain ratio. The minimum-maximum normalization method is usually used to standardize the data. Through these preprocessing steps, the noise, outliers and data scale differences in the prediction input information can be effectively controlled, providing high-quality input data for subsequent phase mismatch assessment.

[0027] The "pre-set phase mismatch coefficient threshold interval" can be determined by statistical analysis and experimental calibration based on historical data to ensure the accuracy and robustness of phase mismatch classification. First, the prediction results of the LSTM model and Kalman filter can be extracted from a large amount of historical flight data, and the phase mismatch coefficients generated by these data under different flight conditions (such as coefficient values ​​in normal flight, slight disturbance, and severe disturbance) are calculated. Then, the phase mismatch coefficients are classified by cluster analysis (such as K-means clustering) to determine the coefficient range corresponding to each category, so as to find the preliminary dividing point between phase consistency, slight mismatch, and severe mismatch. In order to improve accuracy, statistical methods can be used, such as calculating the mean and standard deviation of each type of phase mismatch coefficient, and setting the threshold interval to the mean of each type of coefficient ±2 times the standard deviation to cover most data points. Finally, the set threshold interval is verified and fine-tuned through software simulation testing to observe whether the performance of actual flight data in these intervals meets the expected classification criteria. Through this process, a robust phase mismatch coefficient threshold interval can be obtained, so as to effectively classify the phase mismatch degree of LSTM and Kalman filter in practical applications.

[0028] In this embodiment, the acquisition logic of the prediction time deviation index and the height prediction error synchronization index is as follows: Extract the prediction time synchronization information from the preprocessed prediction input information, including the timestamps of the LSTM model generating prediction results at different times within a period of time, the timestamps of the Kalman filter generating prediction results, and the timestamps of actually obtaining the aircraft altitude data, and mark them as , and , Indicates that within a period of time The timestamp of the moment when the LSTM model generates the prediction result, Indicates that within a period of time The timestamp of the moment the Kalman filter generates the prediction result, Indicates that within a period of time The timestamp of the actual acquisition of the aircraft altitude data at the moment, , is a positive integer; To extract the prediction time synchronization information from the preprocessed prediction input information, it is necessary to extract the corresponding timestamps from the LSTM model, Kalman filter, and sensor data respectively. First, for the timestamp of the prediction result generated by the LSTM model, the time when the LSTM model outputs the predicted height result can be recorded in real time by the software. Usually, these timestamps are automatically generated by the system's internal clock and attached to the output result. Every time the LSTM model generates a new prediction result within a time window, the system automatically saves the prediction result and its generated timestamp to form time series data. Next, the timestamp of the Kalman filter can be recorded in the same way, that is, every time the Kalman filter performs error correction on the sensor data and generates a new height value, the system will also save the corresponding timestamp to ensure that it forms a comparable data pair with the timestamp of the LSTM prediction result. Finally, the timestamp of the actual aircraft height data is collected by sensors (such as GNSS, INS, barometric altimeter). Every time the sensor collects new height data, the system automatically attaches the timestamp collected by the sensor. Usually, these timestamps are directly generated by the system clock provided by the hardware, reflecting the time when the actual data was obtained. Through the software's time synchronization module, the timestamp data from different data sources are aligned along the time axis to form a set of synchronized records of three types of timestamp data. Then, the timestamp data of LSTM, Kalman, and sensors are extracted in sequence within the set time window and stored as predicted time synchronization information for a period of time. This ensures that each timestamp data corresponds accurately and provides high-quality input information for subsequent time deviation analysis.

[0029] Calculate the forecast time deviation index. The specific calculation formula is as follows: In the formula, is the forecast time deviation index; This formula is used to calculate the forecast time deviation index , which aims to quantify the degree of synchronization between the LSTM model and the Kalman filter on the time axis, thereby evaluating their time deviation when generating prediction results. The formula is divided into several key steps. First, Indicates at a point in time The time difference between the prediction result generated by the LSTM model at the moment and the time when the actual height data is obtained, which reflects the delay or advance of the LSTM prediction result. Similarly, It represents the time difference between the Kalman filter prediction result and the actual height data, reflecting the time offset of the Kalman filter. The denominator is added with "+1" to prevent division by zero and make the calculation more stable. Next, the absolute ratio of the LSTM time difference to the Kalman time difference is squared to enhance the penalty effect when the deviation is large and highlight the time asynchrony. The step of summing and averaging unifies the time deviation of each time point into an average value, thereby reflecting the time synchronization of the LSTM and Kalman filter in the entire time window. Through this formula, The overall level of time deviation between LSTM and Kalman filter can be quantified, providing an accurate reference indicator for subsequent analysis and classification of phase mismatch.

[0030] Forecast Time Deviation Index The size of directly reflects the degree of synchronization between the LSTM model and the Kalman filter prediction results on the time axis, and thus reflects the degree of phase mismatch between them. Specifically, the smaller The value of indicates that the time deviation between the LSTM model and the Kalman filter is small, that is, the time when the two generate prediction results is closer, indicating that their prediction results are more synchronized in time and the phases are basically consistent; while the larger The value indicates that there is a significant deviation in the time between the prediction results of LSTM and Kalman filter, and the generated prediction time is inconsistent, resulting in phase asynchrony. The greater the asynchrony, the more serious the phase mismatch. Therefore, The larger the value, the higher the phase mismatch between the two, which may cause the system to be unable to balance the long-term trend prediction of LSTM and the short-term correction of Kalman filtering, thus affecting the stability and accuracy of the overall prediction.

[0031] Extract the prediction error comparison information from the preprocessed prediction input information, including the altitude prediction value of the LSTM model at different times in a period of time, the altitude prediction value of the Kalman filter, and the actual altitude of the aircraft obtained, and calibrate them as , and , Indicates that within a period of time The height prediction value of the LSTM model at the moment, Indicates that within a period of time The height prediction value of the Kalman filter at time instant, Indicates that within a period of time The actual altitude of the aircraft obtained at all times; Extracting prediction error comparison information from the preprocessed prediction input information mainly involves obtaining the altitude prediction value of the LSTM model, the altitude prediction value of the Kalman filter, and the actual altitude of the aircraft over a period of time. These data can be extracted from different modules through the software system and aligned in the same time window. First, the altitude prediction value of the LSTM model is generated by the deep learning algorithm based on historical sensor data. It can be automatically recorded by the software each time the prediction result is calculated. These prediction values ​​are usually used as model outputs and are stored in the database with a timestamp attached when they are generated, forming a time series of altitude predictions. Secondly, the altitude prediction value of the Kalman filter is a real-time error correction based on the current sensor data. The software will also store the prediction value with the corresponding timestamp every time the Kalman filter updates the altitude prediction result. In order to achieve data consistency, the system ensures that each altitude prediction value is stored in chronological order so that it can be aligned with the LSTM model output. Finally, the actual altitude data of the aircraft is collected in real time by sensors (such as GNSS, INS, and barometric altimeter). The altitude value output by the sensor is directly obtained through the software interface, and the timestamp of the sensor is attached. The software system will extract the values ​​corresponding to the time point in the set time window from the LSTM prediction value, Kalman filter value and the actual height value of the sensor, and synchronize these data in time to ensure that the three types of values ​​are obtained at the same time point. In this way, the system can generate consistent prediction error comparison information, ensure that the LSTM prediction value, Kalman filter prediction value and actual height data at each moment accurately correspond, and provide high-quality input data for subsequent error analysis and phase mismatch assessment.

[0032] Calculate the height prediction error synchronization index. The specific calculation formula is as follows: In the formula, is the height forecast error synchronization index.

[0033] This formula is used to calculate the height forecast error synchronicity index. , which aims to quantify the synchronization between the LSTM model and the Kalman filter in the predicted height error to evaluate the phase consistency between them. Indicates at a point in time At the moment, the error between the LSTM model's predicted height and the actual height, which quantifies the LSTM model's prediction bias. Similarly, represents the error between the Kalman filter's predicted height and the actual height, which reflects the Kalman filter's prediction deviation. By calculating the difference between these two errors, the formula can obtain the degree of asynchrony between the prediction errors of the LSTM model and the Kalman filter. It plays a normalization role, which is used to reduce the impact of different actual height ranges on error calculation, and "+1" can avoid zero division errors when the actual height is zero. The absolute value operation is used to eliminate the impact of positive and negative deviations, making the calculation of the synchronization index more stable. Finally, by summing up the error asynchrony degree at all time points and taking the average value, The error synchronization between the LSTM and Kalman filter prediction results in the entire time window can be measured. The larger the value, the more asynchronous the error changes between the two and the higher the phase mismatch. A small value indicates that the error changes of the two are relatively consistent, that is, the phase synchronization is good.

[0034] Height forecast error synchronicity index The size of directly reflects the synchronization of the LSTM model and the Kalman filter in predicting the height error, and thus reflects the degree of phase mismatch between them. Specifically, the smaller The value of indicates that the changes in the prediction errors of the LSTM model and the Kalman filter are synchronized, that is, the deviations of the two from the actual height are basically consistent, indicating that their prediction results are relatively consistent in phase; on the contrary, a larger The value shows that the changes in the prediction errors of LSTM and Kalman filter are not synchronized, indicating that there is a large difference in the deviation of the actual height between the two, resulting in phase mismatch. The larger the value, the greater the phase mismatch, which will affect the system's accurate prediction of the aircraft's altitude and control stability. The value indicates that the LSTM and Kalman filter are more synchronized in time, with a lower phase mismatch and better coordination of the prediction results.

[0035] In this embodiment, the generated prediction time deviation index and height forecast error synchronicity index Construct a phase mismatch evaluation model and generate a phase mismatch coefficient by weighted summation , and the generated phase mismatch coefficient The phase mismatch coefficient threshold interval is set A comparison is made, and the phase mismatch degree between the prediction results of the LSTM model and the Kalman filter is predicted based on the comparison results. The prediction results are divided into three categories: phase consistency, mild mismatch, and severe mismatch. The specific comparison analysis and classification are as follows: like , if the phase mismatch between the prediction results of the LSTM model and the Kalman filter is low, the situation is classified as phase consistent; This situation shows that the prediction results of the LSTM model and the Kalman filter are highly synchronized in time, and the error changes are basically the same, indicating that the prediction results of the two have almost no deviation in phase. In this case, the long-term trend prediction of the LSTM model and the short-term correction of the Kalman filter can work in coordination to achieve accurate prediction of the aircraft's altitude. The phase-consistent state can ensure the stability and accuracy of the system's prediction, provide reliable altitude data support for the aircraft, and facilitate smooth flight and precise control.

[0036] like , the phase mismatch between the prediction results of the LSTM model and the Kalman filter is moderate, then the situation is classified as mild mismatch; This situation indicates that there is a certain time deviation or error change asynchrony between the prediction results of the LSTM model and the Kalman filter, but the deviation is small, which is a mild mismatch state. In this case, the prediction results of the two are slightly out of sync in phase, which may cause slight fluctuations or instability in the system when predicting altitude. Although this degree of mismatch will not significantly affect the safety of the aircraft, it may cause the aircraft to respond slightly slowly or inaccurately in flight missions that require highly accurate predictions, so the system may need to make small adjustments to improve the prediction effect.

[0037] like , if the phase mismatch between the prediction results of the LSTM model and the Kalman filter is high, the situation is classified as severe mismatch.

[0038] This situation shows that the prediction results of the LSTM model and the Kalman filter are significantly out of sync in time, with large differences in error changes, and are in a serious mismatch state. In this case, the phases of the prediction results of the two are significantly out of sync, and they cannot work together effectively, which greatly reduces the system's prediction accuracy of the aircraft's altitude, and may cause frequent prediction errors and instability. This serious mismatch will affect the altitude control of the aircraft, increase the risk of the aircraft when performing precise operations (such as landing or obstacle avoidance), and may cause control response failure or instability. Therefore, active adjustment measures need to be taken to improve phase synchronization and ensure flight safety.

[0039] To build a phase mismatch estimation model and generate phase mismatch coefficients First, we need to generate the forecast time deviation index and height forecast error synchronicity index Perform weighted summation. Specifically, It is used to measure the synchronization between the LSTM model and the Kalman filter on the time axis. It is used to evaluate the synchronization of the two in terms of prediction error. , two weight coefficients can be set and To adjust and The influence of the weight coefficient on the final phase mismatch coefficient. The selection of the weight coefficient should be based on the degree of concern for time synchronization and error synchronization in practical applications. For example, in flight missions with high requirements for time synchronization, Can be set slightly larger to increase In scenarios where higher accuracy is required, the The final phase mismatch coefficient Calculated by the following formula: ,in and The value of needs to be optimized through a large amount of experimental data analysis or actual flight tests to ensure that the phase mismatch evaluation model can accurately reflect the phase synchronization of LSTM and Kalman filter, thereby providing a reliable basis for the classification of the degree of phase mismatch.

[0040] The adaptive weight adjustment module builds an adaptive weight adjustment mechanism based on the division results, and performs different weight adjustment measures for the three situations of phase consistency, mild mismatch and severe mismatch; In this embodiment, in the adaptive weight adjustment module, an adaptive weight adjustment mechanism is constructed according to the division results, specifically: according to the division results of phase consistency, mild mismatch and severe mismatch, different weight adjustment parameters are set, corresponding to the weight allocation of the LSTM model and the Kalman filter in the prediction fusion, respectively, to form an adaptive weight adjustment mechanism; the adjustment mechanism is based on the range of the phase mismatch coefficient, and automatically determines the weight allocation strategy and adjustment range between the LSTM model and the Kalman filter through pre-set rules; In the adaptive weight adjustment module, in order to automatically adjust the weight distribution of the LSTM model and the Kalman filter according to the division results of phase consistency, mild mismatch and severe mismatch, it can be achieved through dynamic weight distribution algorithm and rule setting. Specifically, it is necessary to first define the range of the phase mismatch coefficient and the corresponding weight adjustment rules in the software, that is, the numerical interval of the preset phase mismatch coefficient (such as the threshold of phase consistency, mild mismatch, and severe mismatch) and the weight ratio of the LSTM and Kalman filter corresponding to each interval. For example, in the software, the balance weight is set for the phase consistency state, the slightly higher weight of the Kalman filter is set for the mild mismatch state, and the higher weight of the Kalman filter is set for the severe mismatch state to ensure that the system has an appropriate response strategy under different mismatch states. Then, the system will calculate the current phase mismatch coefficient in real time and automatically match the corresponding weight adjustment rules according to the value of the coefficient. In this way, the software can automatically adjust the weight distribution of the LSTM and Kalman filter based on the current mismatch degree, so as to dynamically balance the relationship between long-term trend prediction (LSTM) and short-term error correction (Kalman filter). Through this implementation, the system can adapt to different phase mismatch situations, improve the accuracy and stability of the prediction, and thus ensure the reliability of the aircraft altitude prediction.

[0041] Different weight adjustment measures are taken for the three situations of phase consistency, mild mismatch and severe mismatch. Specifically, in the case of phase consistency, a balanced weight strategy is adopted to assign equal weights to the LSTM model and the Kalman filter to maintain stable prediction; in the case of mild mismatch, the weight of the Kalman filter is adjusted to be higher than that of the LSTM model to increase the influence of short-term correction in the prediction, so that the system can maintain a balance between long-term and short-term predictions while enhancing its responsiveness to current environmental changes; in the case of severe mismatch, the weight of the Kalman filter is significantly increased, and the weight of the LSTM model is significantly reduced to ensure that the system can respond quickly to sudden changes and enhance the short-term stability of highly predicted results.

[0042] In order to achieve different weight adjustment measures for the three types of situations, namely phase consistency, mild mismatch and severe mismatch, the software can automatically implement them through the dynamic weight control algorithm combined with the conditional judgment rules. Specifically, the software can first set the weight allocation rules for phase consistency, mild mismatch and severe mismatch. For example, when the phase is consistent, the LSTM model and the Kalman filter are assigned equal weights; when the mismatch is mild, the weight ratio of the Kalman filter is increased and the weight of the LSTM is slightly reduced; when the mismatch is severe, the weight of the Kalman filter is significantly increased, and the weight of the LSTM is significantly reduced. The system will determine which mismatch state it is in through the phase mismatch coefficient calculated in real time, and automatically select the corresponding weight configuration strategy. Whenever the mismatch state changes, the software will automatically adjust the weights of the LSTM and Kalman filter based on the preset weight rules to adapt to the current environmental requirements. The purpose of this dynamic control method is to maintain the stability of the prediction in the phase consistency state, moderately improve the short-term correction capability in the mild mismatch state, and give priority to rapid response to sudden changes in the severe mismatch state, thereby enhancing the overall prediction accuracy and stability of the system.

[0043] The adjustment mechanism feedback evaluation module obtains the adjustment mechanism feedback information in real time during the process of the adaptive weight adjustment mechanism adjusting the weight, and analyzes it after obtaining it to evaluate whether the adjustment degree of the adaptive weight adjustment mechanism can meet expectations, and optimizes the adaptive weight adjustment mechanism according to the evaluation results; In this embodiment, in the adjustment mechanism feedback evaluation module, after the adjustment mechanism feedback information is acquired in real time, it is preprocessed, and the instantaneous prediction adjustment feedback information and the long-term adjustment cumulative feedback information in the preprocessed adjustment mechanism feedback information are extracted, and analyzed after extraction to generate an immediate adjustment deviation index and a cumulative adjustment effect coefficient, respectively; an adjustment degree evaluation model is constructed for the generated immediate adjustment deviation index and cumulative adjustment effect coefficient, an adjustment coefficient is generated, and the generated adjustment coefficient is compared with a pre-set adjustment coefficient threshold, and whether the adjustment degree of the adaptive weight adjustment mechanism can meet expectations is evaluated based on the comparison results, and the adaptive weight adjustment mechanism is optimized based on the evaluation results.

[0044] In the process of the adaptive weight adjustment mechanism adjusting the weight, real-time acquisition of the adjustment mechanism feedback information can be achieved through data monitoring and automatic recording. Specifically, the software system can set a real-time monitoring module to automatically capture the output prediction value, real-time error and currently adjusted weight parameters of the LSTM model and Kalman filter after each adjustment. Whenever the weight adjustment mechanism fine-tunes the weights of the LSTM and Kalman filter, the system will immediately call the corresponding sensor interface to obtain the actual altitude data from the aircraft's altitude sensor (such as GNSS, INS, etc.). At the same time, the system will take the current predicted altitude, actual altitude, real-time error and weight parameters as part of the feedback information, attach a timestamp and store them in a temporary data cache to ensure the time series integrity of the data. At each time interval of the entire adjustment process, the system automatically records this feedback information and synchronizes it to the feedback evaluation module for subsequent analysis and processing. Through this software automatic monitoring and recording method, the system can achieve real-time acquisition of the adjustment mechanism feedback information, ensure the timeliness and integrity of the feedback data, and thus support immediate adjustment effect evaluation.

[0045] In the adjustment mechanism feedback evaluation module, the adjustment mechanism feedback information obtained in real time needs to be preprocessed to ensure the accuracy, consistency and reliability of the data, so as to improve the accuracy of the evaluation. First, the feedback information may contain sensor noise and outliers, so it is necessary to remove the noise through filtering (such as Kalman filtering or moving average filtering), and use threshold filtering to remove abnormal readings. Next, in order to ensure that the feedback information from the LSTM model, Kalman filter and actual height can be compared and analyzed at the same time point, the system needs to time align the data to keep the data sequence consistent. In addition, data normalization is also necessary to standardize data from different sources to a unified scale or relative proportion to ensure comparison in the same dimension. Through such preprocessing, the feedback data is smoother, standardized and consistent, which can provide a reliable basis for the subsequent evaluation of the adjustment effect.

[0046] The "pre-set adjustment coefficient threshold" can be determined based on historical data analysis and simulation test results to ensure that the threshold adapts to different flight scenarios and meets the stability and response requirements of the system. Specifically, the software system can first import a large amount of flight data, including real-time feedback data of the LSTM model and Kalman filter during the operation of the adjustment mechanism under different flight conditions. Then, these historical data are statistically processed through the data analysis module to calculate the mean, variance and distribution range of each adjustment coefficient in different scenarios. Subsequently, the system can use the simulation environment to conduct multiple rounds of weight adjustment tests, simulate various emergencies and altitude changes, and observe the impact of different adjustment coefficients on system performance. According to the simulation results, the software will automatically record the minimum adjustment coefficient value that ensures prediction stability and the maximum adjustment coefficient value for rapid response in each test, and then determine a suitable threshold range. Finally, the system will use this range as the basis for setting the adjustment coefficient threshold, and fine-tune it according to the evaluation results in actual applications to optimize the adjustment effect and meet the height control requirements of the aircraft in various environments.

[0047] In this embodiment, the logic for obtaining the instant adjustment deviation index and the cumulative adjustment effect coefficient is as follows: The instantaneous prediction adjustment feedback information in the preprocessed adjustment mechanism feedback information is extracted, specifically including the difference between the predicted altitude of the LSTM model and Kalman filter combination and the actual altitude of the aircraft at different times within a period of time during the process of the adaptive weight adjustment mechanism adjusting the weight, the weight difference between the LSTM model and the Kalman filter, and the time interval of the system responding to the change of the aircraft altitude after the weight adjustment, and calibrated as , and , Indicates that during the period of time when the adaptive weight adjustment mechanism adjusts the weight The difference between the predicted altitude of the LSTM model and Kalman filter combination and the actual altitude of the aircraft at that moment, Indicates that during the period of time when the adaptive weight adjustment mechanism adjusts the weight The weight difference between the LSTM model and the Kalman filter at this moment, It represents the time interval for the system to respond to the change of aircraft altitude after weight adjustment. , is a positive integer; When extracting the instantaneous prediction adjustment feedback information from the preprocessed adjustment mechanism feedback information, the software system can obtain and record the difference between the predicted height and the actual height of the LSTM model and the Kalman filter, the weight difference, and the time interval of the system response through the real-time data acquisition and data synchronization processing modules. The specific extraction process is as follows: First, during the adjustment process of the adaptive weight adjustment mechanism, the software will monitor and record the predicted height value after the combination of LSTM and Kalman filter and the actual height value from the aircraft height sensor (such as data provided by sensors such as GNSS and INS). The system will automatically calculate the difference between the two as real-time prediction error data. This error value can reflect the deviation between the prediction result and the actual height, indicating the prediction effect after adjustment. At the same time, the software will automatically record the current weight difference between LSTM and Kalman filter (that is, the relative change of the weights of the two), which indicates the specific adjustment range of the weight during the adjustment process, which helps to quantify the strength of the adjustment mechanism. In addition, the software will mark the adjusted response time, that is, record the time interval of the system's response to the actual height change after the weight adjustment takes effect, and track the start and end of each adjustment through the built-in timestamp, so as to calculate the response speed of the system. Through this set of real-time data collection and synchronous calculation mechanisms, the software can automatically extract the required instantaneous prediction adjustment feedback information, providing accurate basic data for subsequent evaluation of adjustment effects.

[0048] Calculate the instant adjustment deviation index. The specific calculation formula is as follows: In the formula, To adjust the deviation index instantly; Instantly adjust the deviation index The calculation formula is designed to measure the response effect of the adaptive weight adjustment mechanism in a short period of time, and each calculation step in the formula serves this goal. First, represents the instantaneous error between the predicted height and the actual height after each adjustment, which reflects whether the weight adjustment can reduce the prediction deviation in time. Represents the weight difference between the LSTM model and the Kalman filter, which is used to quantify the strength of weight adjustment; Indicates the adjusted system response time, reflecting the immediate response speed of the adjustment mechanism. The weighted denominator structure can balance the relationship between error and adjustment amplitude, ensuring that a slow response speed or a large adjustment amplitude will not excessively affect the calculation result. Finally, the sum of the squares of all instantaneous errors is divided by the number of adjustments. The averaging process can smooth out short-term fluctuations and provide an overall evaluation of the immediate response effect. Overall, this formula effectively balances the error, adjustment range, and response speed, and can objectively evaluate the immediate adjustment effect of the adjustment mechanism.

[0049] Instantly adjust the deviation index The size of directly reflects the improvement effect of the adaptive weight adjustment mechanism on the prediction accuracy in the short term, and then evaluates whether the adjustment mechanism has achieved the expected adjustment degree. Specifically, The smaller the value, the faster the error between the predicted height and the actual height decreases after the weight adjustment, and the faster the system responds to the height change, which indicates that the immediate effect of the weight adjustment mechanism is better and meets the expected adjustment target. On the contrary, if If the value is large, it means that after the weight adjustment, the prediction error is still large or the response speed after adjustment is slow, and it cannot quickly adapt to the highly changing needs, indicating that the immediate adjustment effect of the adjustment mechanism is not good and does not meet the expected requirements. The numerical value can be used to determine whether the immediate response capability and accuracy of the current adjustment mechanism meet the set performance indicators.

[0050] The long-term adjustment cumulative feedback information in the preprocessed adjustment mechanism feedback information is extracted, specifically including the cumulative prediction error change after multiple weight adjustments within the time window in which the adaptive weight adjustment mechanism adjusts the weight, the cumulative amplitude of all weight adjustments within the time window, and the time length within the time window, and they are marked as , and , It represents the cumulative change in prediction error after multiple weight adjustments within the time window in which the adaptive weight adjustment mechanism adjusts the weight. Represents the cumulative magnitude of all weight adjustments within the time window, Indicates the length of time within the time window; Extracting the long-term adjustment cumulative feedback information can be achieved through the long-term data monitoring and cumulative calculation module of the software system, so that the system can evaluate the cumulative effect of multiple adjustments within the time window when the adaptive weight adjustment mechanism is running. Specifically, the system automatically records the prediction error at each weight adjustment, that is, the difference between the predicted height and the actual height of the LSTM and Kalman filter combination. The system will accumulate these instantaneous errors and calculate the cumulative prediction error change, thereby reflecting the overall error change trend within the time window. At the same time, the system will also record the weight change amplitude of the LSTM and Kalman filter at each adjustment, that is, the weight difference between the LSTM and Kalman filter in each adjustment, and accumulate these weight differences one by one within the time window to calculate the cumulative amplitude of the weight adjustment, indicating the total strength of the adjustment within this window. In addition, the system marks the time of each adjustment with a timestamp, and calculates the length of time from the beginning to the end at the end of the window, that is, the time span within the entire window. These data are automatically extracted and stored through the software's cumulative calculation and data recording modules, providing a complete data basis for the subsequent evaluation of the long-term effect of weight adjustment.

[0051] Calculate the cumulative adjustment effect coefficient. The specific calculation formula is as follows: In the formula, It is the cumulative adjustment effect coefficient.

[0052] Cumulative Adjustment Effect Coefficient The calculation formula of is intended to evaluate the error convergence effect of the adaptive weight adjustment mechanism in the long term. Each operation step in the formula has a specific evaluation function. First, the numerator It indicates the cumulative change in prediction error after multiple weight adjustments within the time window, reflecting whether the prediction error is gradually reduced after multiple adjustments. is smaller, indicating that the error is gradually decreasing and the adjustment effect is better. It represents the cumulative magnitude of weight adjustments, that is, the total magnitude of all weight adjustments made within this window, which measures the strength of the adjustment mechanism in the long term. and The ratio of can reflect the convergence of the prediction error under multiple adjustments. In order to further quantify the degree of convergence of the error, this ratio is squared to emphasize the significance of the error change after adjustment. Finally, the whole result is divided by ,in is the length of the time window, ensuring that the value decreases when the time is longer, indicating the persistence of the adjustment effect in the long-term evaluation. The smaller the value, the better the effect of weight adjustment in the long run, and the error tends to be stable; A larger value indicates that the adjustment effect is not ideal and the error is difficult to converge.

[0053] Cumulative Adjustment Effect Coefficient The size of directly reflects the ability of the adaptive weight adjustment mechanism to control the prediction error in the long term and evaluates whether its adjustment effect has achieved the expected result. Specifically, The smaller the value, the more the cumulative change of the forecast error tends to converge within the time window after the weight adjustment, and the amplitude of the weight adjustment is relatively stable, which means that the adjustment mechanism effectively reduces the error in the long term, achieves a good error control effect, and achieves the expected adjustment goal; on the contrary, if If the value is large, it means that the adjustment mechanism has not been able to effectively reduce the error over a long period of time, the error changes greatly or the adjustment range is unstable, indicating that the long-term effect of the adjustment mechanism is not ideal and does not meet the expected requirements. The numerical value of can be used to determine whether the current adjustment mechanism has achieved the expected adjustment effect in long-term error control.

[0054] In this embodiment, the generated instant adjustment deviation index and the cumulative adjusted effect coefficient Construct an adjustment degree assessment model and generate adjustment coefficients through weighted summation , and the resulting adjustment factor With the preset adjustment factor threshold Compare and evaluate whether the adjustment degree of the adaptive weight adjustment mechanism can meet expectations based on the comparison results, and optimize the adaptive weight adjustment mechanism based on the evaluation results. The specific comparison and analysis are as follows: like , the adjustment degree of the adaptive weight adjustment mechanism can meet expectations, and there is no need to optimize the adaptive weight adjustment mechanism; This situation means that the current adjustment degree of the adaptive weight adjustment mechanism has achieved the expected effect, which can effectively balance the weight distribution of the LSTM model and the Kalman filter and accurately track the change of flight altitude. In this case, the system does not need to further optimize the adjustment mechanism, indicating that the current adjustment strategy has good stability and adaptability in practical applications, and can maintain the accuracy of altitude prediction and the sensitivity of system response under current environmental conditions.

[0055] like , the adjustment degree of the adaptive weight adjustment mechanism cannot meet expectations, and the adaptive weight adjustment mechanism needs to be optimized, including: increasing the weight adjustment frequency of the LSTM model and the Kalman filter to improve the system's response speed to environmental changes; resetting the upper limit of the weight adjustment range to ensure the short-term response effect during the adjustment process; optimizing the weight adjustment parameters, dynamically adjusting the weight ratio of the LSTM model and the Kalman filter to adapt to the current flight status requirements; combining historical data and real-time feedback information, analyzing the prediction error and response time, automatically adjusting the system threshold setting, and improving the system's adaptability and accuracy.

[0056] This situation indicates that the adjustment effect of the adaptive weight adjustment mechanism has not met expectations, which may lead to an increase in prediction error or insufficient system response speed, making it difficult to cope with complex or sudden flight environments. In this case, the adjustment mechanism needs to be optimized by increasing the adjustment frequency, resetting the weight adjustment range, and dynamically adjusting the weight parameters to improve the system's error convergence speed and adjustment adaptability, so as to ensure that the aircraft can still maintain a high degree of prediction accuracy and stability in a changing environment.

[0057] In order to optimize the adaptive weight adjustment mechanism so that it meets the adjustment requirements under current conditions, the software can achieve this optimization process in the following ways. First, by adding a monitoring module for the weight adjustment frequency to the system, the software can automatically increase the weight update rate of the LSTM model and the Kalman filter, thereby shortening the weight adjustment cycle, ensuring that the system can respond to environmental changes more promptly, and improve its adaptability to emergencies. Secondly, the software can dynamically set the upper limit of the weight adjustment amplitude, that is, set the maximum allowable weight change amplitude in each adjustment to ensure effective response in a short period of time. This setting can be adjusted based on real-time error feedback. When the error increases, the adjustment amplitude also increases accordingly to ensure that the system can make rapid adjustments when the error increases suddenly. In addition, the software can optimize the weight adjustment parameters through intelligent algorithms, such as using adaptive parameter adjustment methods to adjust the weight ratio of the LSTM model and the Kalman filter in real time according to the current flight status (such as speed, attitude, etc.), so as to optimize the prediction effect according to different flight status requirements. Finally, the software can combine historical data and real-time feedback information, calculate the average and changing trend of prediction error and response time through the data analysis module, and automatically adjust the threshold setting of the system, so as to appropriately lower the threshold in a highly changing environment to increase sensitivity, and appropriately increase the threshold in a stable environment to enhance stability. These optimization measures not only enable the system to adapt more flexibly to different flight scenarios and mutations, but also improve the overall prediction accuracy and stability of the system, achieving continuous optimization.

[0058] When constructing the adjustment degree assessment model, the software system will generate an instant adjustment deviation index based on the generated and the cumulative adjusted effect coefficient To evaluate the overall adjustment effect of the adaptive weight adjustment mechanism. and Perform a weighted sum to generate a comprehensive adjustment factor , which is used to measure the comprehensive performance of the adjustment effect. Assume that the weight coefficients are and , then the calculation formula of the adjustment coefficient is: .in, and Respectively represent and These coefficients are set according to the needs of adjusting the model. Usually, The value of can be set higher to emphasize the immediate response effect of the system and ensure that the adjustment mechanism responds quickly in the short term; It is used to balance the long-term stability of the system and the error convergence effect, so it can be dynamically adjusted according to different flight environments. For example, in a stable environment, you can increase The weight of Through this weighted summation, the adjustment coefficient can comprehensively reflect the adjustment effect of the adaptive weight adjustment mechanism in both the immediate and long-term dimensions, thereby achieving accurate adjustment degree evaluation.

[0059] The flight process monitoring and optimization module monitors and analyzes the altitude prediction data and adjustment records during the entire flight process in real time, and continuously optimizes the adaptive weight adjustment mechanism based on the monitoring results to ensure the altitude prediction stability and control accuracy of the aircraft in different flight phases and environmental conditions.

[0060] The flight process monitoring and optimization module can achieve real-time monitoring and dynamic optimization of the flight process by establishing a closed-loop system of data collection, analysis and feedback adjustment. Specifically, the software first collects the aircraft's altitude prediction data and adjustment records in real time through the sensor interface, including the real-time predicted altitude, actual altitude, weight adjustment frequency and amplitude of the LSTM model and Kalman filter. These data are stored in the system's time series database with precise timestamps for historical analysis. Then, the software uses the data analysis module to perform statistical and trend analysis on real-time data through algorithms, such as calculating the error mean, variance, and prediction error fluctuations, to evaluate the current prediction stability and accuracy. Based on these analysis results, the software can identify the error characteristics and control requirements under different flight stages (such as climb, cruise, descent) and environmental changes (such as airflow, temperature changes). Then, the system will dynamically adjust the parameters of the adaptive weight adjustment mechanism based on the analysis results, such as increasing the weight adjustment frequency or response speed in the case of large errors or sudden changes in the environment to enhance the system's adaptability, and reducing the adjustment frequency when the environment is stable to ensure the stability and energy efficiency of the system. This closed-loop system of real-time monitoring and feedback optimization ensures that the aircraft can maintain highly predictable accuracy and control stability in all flight phases and environmental conditions, providing data support and optimization guarantees for flight safety.

[0061] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0062] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0063] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0064] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0065] In the several embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0066] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0067] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0068] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A multi-sensor height prediction system integrating LSTM and Kalman filtering, characterized by: It includes aircraft altitude monitoring and data acquisition module, phase mismatch prediction and classification module, adaptive weight adjustment module, adjustment mechanism feedback evaluation module and flight process monitoring and optimization module; The aircraft altitude monitoring and data acquisition module monitors the aircraft altitude data in real time during the flight of the aircraft, and obtains the prediction input information required for generating LSTM model and Kalman filter prediction when a sharp fluctuation in the aircraft altitude data is detected; The phase mismatch prediction and classification module analyzes the obtained prediction input information, predicts the phase mismatch degree between the prediction results of the LSTM model and the Kalman filter, and divides them into three categories according to the prediction results: phase consistency, mild mismatch, and severe mismatch; The adaptive weight adjustment module builds an adaptive weight adjustment mechanism based on the division results, and performs different weight adjustment measures for the three situations of phase consistency, mild mismatch and severe mismatch; The adjustment mechanism feedback evaluation module obtains the adjustment mechanism feedback information in real time during the process of the adaptive weight adjustment mechanism adjusting the weight, and analyzes it after obtaining it to evaluate whether the adjustment degree of the adaptive weight adjustment mechanism can meet expectations, and optimizes the adaptive weight adjustment mechanism according to the evaluation results; The flight process monitoring and optimization module monitors and analyzes the altitude prediction data and adjustment records during the entire flight process in real time, and continuously optimizes the adaptive weight adjustment mechanism based on the monitoring results to ensure the altitude prediction stability and control accuracy of the aircraft in different flight phases and environmental conditions.

2. The multi-sensor height prediction system integrating LSTM and Kalman filtering according to claim 1 is characterized in that: In the phase mismatch prediction and classification module, the acquired prediction input information is preprocessed, and the prediction time synchronization information and prediction error comparison information in the preprocessed prediction input information are extracted, and analyzed after extraction to generate a prediction time deviation index and a height prediction error synchronization index, respectively; a phase mismatch evaluation model is constructed for the generated prediction time deviation index and the height prediction error synchronization index to generate a phase mismatch coefficient, and the generated phase mismatch coefficient is compared with a pre-set phase mismatch coefficient threshold interval, and the phase mismatch degree between the prediction results of the LSTM model and the Kalman filter is predicted according to the comparison results, and it is divided into three categories according to the prediction results: phase consistency, mild mismatch and severe mismatch.

3. The multi-sensor height prediction system integrating LSTM and Kalman filtering according to claim 2 is characterized in that: The acquisition logic of the prediction time deviation index and the height prediction error synchronization index is as follows: Extract the prediction time synchronization information from the preprocessed prediction input information, including the timestamps of the LSTM model generating prediction results at different times within a period of time, the timestamps of the Kalman filter generating prediction results, and the timestamps of actually obtaining the aircraft altitude data, and mark them as , and , Indicates that within a period of time The timestamp of the moment when the LSTM model generates the prediction result, Indicates that within a period of time The timestamp of the moment the Kalman filter generates the prediction result, Indicates that within a period of time The timestamp of the actual acquisition of the aircraft altitude data at the moment, , is a positive integer; Calculate the forecast time deviation index. The specific calculation formula is as follows: In the formula, is the forecast time deviation index; Extract the prediction error comparison information from the preprocessed prediction input information, including the altitude prediction value of the LSTM model at different times in a period of time, the altitude prediction value of the Kalman filter, and the actual altitude of the aircraft obtained, and calibrate them as , and , Indicates that within a period of time The height prediction value of the LSTM model at the moment, Indicates that within a period of time The height prediction value of the Kalman filter at time instant, Indicates that within a period of time The actual altitude of the aircraft obtained at all times; Calculate the height prediction error synchronization index. The specific calculation formula is as follows: In the formula, is the height forecast error synchronization index.

4. The multi-sensor height prediction system integrating LSTM and Kalman filtering according to claim 3 is characterized in that: The time deviation index of the generated forecast and height forecast error synchronicity index Construct a phase mismatch evaluation model and generate a phase mismatch coefficient by weighted summation , and the generated phase mismatch coefficient The phase mismatch coefficient threshold interval is set A comparison is made, and the phase mismatch degree between the prediction results of the LSTM model and the Kalman filter is predicted based on the comparison results. The prediction results are divided into three categories: phase consistency, mild mismatch, and severe mismatch. The specific comparison analysis and classification are as follows: like , if the phase mismatch between the prediction results of the LSTM model and the Kalman filter is low, the situation is classified as phase consistent; like , the phase mismatch between the prediction results of the LSTM model and the Kalman filter is moderate, then the situation is classified as mild mismatch; like , if the phase mismatch between the prediction results of the LSTM model and the Kalman filter is high, the situation is classified as severe mismatch.

5. The multi-sensor height prediction system integrating LSTM and Kalman filtering according to claim 4 is characterized in that: In the adaptive weight adjustment module, an adaptive weight adjustment mechanism is constructed according to the division results. Specifically, different weight adjustment parameters are set according to the division results of phase consistency, mild mismatch and severe mismatch, which correspond to the weight allocation of LSTM model and Kalman filter in prediction fusion, forming an adaptive weight adjustment mechanism. The adjustment mechanism automatically determines the weight allocation strategy and adjustment range between LSTM model and Kalman filter based on the range of phase mismatch coefficient through pre-set rules. Different weight adjustment measures are taken for the three situations of phase consistency, mild mismatch and severe mismatch. Specifically, in the case of phase consistency, a balanced weight strategy is adopted to assign equal weights to the LSTM model and the Kalman filter to maintain stable prediction; in the case of mild mismatch, the weight of the Kalman filter is adjusted to be higher than that of the LSTM model to increase the influence of short-term correction in the prediction, so that the system can maintain a balance between long-term and short-term predictions while enhancing its responsiveness to current environmental changes; in the case of severe mismatch, the weight of the Kalman filter is significantly increased, and the weight of the LSTM model is significantly reduced to ensure that the system can respond quickly to sudden changes and enhance the short-term stability of highly predicted results.

6. The multi-sensor height prediction system integrating LSTM and Kalman filtering according to claim 5 is characterized in that: In the adjustment mechanism feedback evaluation module, after the adjustment mechanism feedback information is obtained in real time, it is preprocessed, and the instantaneous prediction adjustment feedback information and the long-term adjustment cumulative feedback information in the preprocessed adjustment mechanism feedback information are extracted, and analyzed after extraction to generate an immediate adjustment deviation index and a cumulative adjustment effect coefficient respectively; an adjustment degree evaluation model is constructed for the generated immediate adjustment deviation index and cumulative adjustment effect coefficient to generate an adjustment coefficient, and the generated adjustment coefficient is compared with a pre-set adjustment coefficient threshold, and whether the adjustment degree of the adaptive weight adjustment mechanism can meet expectations is evaluated based on the comparison results, and the adaptive weight adjustment mechanism is optimized based on the evaluation results.

7. The multi-sensor height prediction system integrating LSTM and Kalman filtering according to claim 6, characterized in that: The acquisition logic of the instant adjustment deviation index and the cumulative adjustment effect coefficient is as follows: The instantaneous prediction adjustment feedback information in the preprocessed adjustment mechanism feedback information is extracted, specifically including the difference between the predicted altitude of the LSTM model and Kalman filter combination and the actual altitude of the aircraft at different times within a period of time during the process of the adaptive weight adjustment mechanism adjusting the weight, the weight difference between the LSTM model and the Kalman filter, and the time interval of the system responding to the change of the aircraft altitude after the weight adjustment, and calibrated as , and , Indicates that during the period of time when the adaptive weight adjustment mechanism adjusts the weight The difference between the predicted altitude of the LSTM model and Kalman filter combination and the actual altitude of the aircraft at that moment, Indicates that during the period of time when the adaptive weight adjustment mechanism adjusts the weight The weight difference between the LSTM model and the Kalman filter at this moment, It represents the time interval for the system to respond to the change of aircraft altitude after weight adjustment. , is a positive integer; Calculate the instant adjustment deviation index. The specific calculation formula is as follows: In the formula, To adjust the deviation index instantly; The long-term adjustment cumulative feedback information in the preprocessed adjustment mechanism feedback information is extracted, specifically including the cumulative prediction error change after multiple weight adjustments within the time window in which the adaptive weight adjustment mechanism adjusts the weight, the cumulative amplitude of all weight adjustments within the time window, and the time length within the time window, and they are marked as , and , It represents the cumulative change in prediction error after multiple weight adjustments within the time window in which the adaptive weight adjustment mechanism adjusts the weight. Represents the cumulative magnitude of all weight adjustments within the time window, Indicates the length of time within the time window; Calculate the cumulative adjustment effect coefficient. The specific calculation formula is as follows: In the formula, It is the cumulative adjustment effect coefficient.

8. The multi-sensor height prediction system integrating LSTM and Kalman filtering according to claim 7, characterized in that: Instantly adjust the deviation index for the generated and the cumulative adjusted effect coefficient Construct an adjustment degree assessment model and generate adjustment coefficients through weighted summation , and the resulting adjustment factor With the preset adjustment factor threshold Compare and evaluate whether the adjustment degree of the adaptive weight adjustment mechanism can meet expectations based on the comparison results, and optimize the adaptive weight adjustment mechanism based on the evaluation results. The specific comparison and analysis are as follows: like , the adjustment degree of the adaptive weight adjustment mechanism can meet expectations, and there is no need to optimize the adaptive weight adjustment mechanism; like , the adjustment degree of the adaptive weight adjustment mechanism cannot meet expectations, and the adaptive weight adjustment mechanism needs to be optimized, including: increasing the weight adjustment frequency of the LSTM model and the Kalman filter to improve the system's response speed to environmental changes; resetting the upper limit of the weight adjustment range to ensure the short-term response effect during the adjustment process; optimizing the weight adjustment parameters, dynamically adjusting the weight ratio of the LSTM model and the Kalman filter to adapt to the current flight status requirements; combining historical data and real-time feedback information, analyzing the prediction error and response time, automatically adjusting the system threshold setting, and improving the system's adaptability and accuracy.

Images