Coach aircraft pilot physiological status monitoring and early warning system
By fusion of multimodal physiological signals to assess cardiopulmonary coupling strength, system recovery capacity, and cognitive load, dynamic resilience indicators are generated, solving the problems of delayed early warning and high false alarm rate in traditional methods. This enables accurate and proactive early warning of pilots' physiological state, thereby improving flight safety.
Patent Information
- Application Number
- CN202511535327.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-27
- Publication Date
- 2026-03-03
AI Technical Summary
In existing technologies, pilot physiological state monitoring methods rely on single or a few physiological indicators, resulting in delayed warnings, high false alarm rates, and an inability to deeply analyze the intrinsic correlation between the cardiovascular, respiratory, and central nervous systems, or to capture the early dynamic characteristics of system instability.
A physiological state monitoring and early warning system for trainer aircraft pilots was constructed. By fusing multimodal physiological signals, the system's cardiopulmonary coupling strength, recovery capacity, and cognitive load were assessed to generate dynamic resilience indicators. The weights were adjusted according to the flight phase to achieve multi-level early warning.
Significantly reduces false alarm rate, detects physiological disability risks in advance, provides accurate multi-level early warning, and improves flight safety and efficiency.
Smart Images
Figure CN121587689A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of aviation physiological monitoring and early warning technology, specifically to a physiological state monitoring and early warning system for trainer aircraft pilots. Background Technology
[0002] With the improvement of the performance of modern trainer aircraft and the increasing complexity of training subjects, the physiological and psychological load that pilots bear during flight is becoming increasingly severe. The pilot's physiological state is directly related to flight safety and training effectiveness, and it is crucial to ensure the stability of their physiological functions during critical phases such as high-G maneuvers and complex instrument operation. Currently, traditional methods for monitoring pilots' physiological state mostly rely on static threshold judgments of single or a few physiological indicators such as heart rate and blood oxygen. Technicians identify physiological abnormalities by setting fixed thresholds. However, this monitoring method has inherent defects: First, it is a reactive monitoring method, and warnings are usually triggered only when physiological indicators show obvious abnormalities, resulting in delayed warnings and failing to meet the need for advanced prediction. Second, single indicators are easily affected by non-critical factors such as individual differences among pilots and motion artifacts, often leading to a high false alarm rate and reducing the reliability of the system. In addition, this method fails to deeply analyze the intrinsic correlation between the cardiovascular, respiratory, and central nervous systems, and cannot capture the early dynamic characteristics of system instability. Therefore, how to construct a monitoring and early warning method that can comprehensively assess the dynamic characteristics of multiple physiological systems and achieve early and accurate prediction of the risk of physiological disability has become a technical problem that urgently needs to be solved in this field. Summary of the Invention
[0003] To address the aforementioned technical problems, this invention provides a physiological state monitoring and early warning system for trainer aircraft pilots. Specifically, the technical solution of this invention includes: The data acquisition and preprocessing module is used to acquire the pilot's multimodal physiological signals and flight parameters, and process them to generate RR interval sequences; this module is also used to generate continuous respiratory signals, generate multi-lead EEG signals, and generate synchronized G-overload values; The autonomic nervous system resilience assessment module is used to calculate the cross-modal phase synchronization index based on the RR interval sequence and the continuous respiratory signal; and the autonomic nervous system resilience assessment module is used to calculate the convergence rate of the state space trajectory based on the G overload value and the physiological state vector; and the autonomic nervous system resilience assessment module is used to calculate the modular coefficient of the brain functional network based on the multi-lead EEG signal. The dynamic resilience index fusion module is used to combine the cross-modal phase synchronization index, the state space trajectory convergence rate, and the brain functional network modularity coefficient, and generate a dynamic resilience index based on preset weight parameters. The early warning generation module is used to determine the early warning level based on the dynamic resilience index.
[0004] Preferably, the autonomic nervous system resilience assessment module calculates the convergence rate of the state-space trajectory, including: In response to the G overload value exceeding a preset impact threshold, the Euclidean distance attenuation process between the physiological state vector and the dynamic baseline point is tracked. Based on the Euclidean distance decay process, the convergence rate of the state space trajectory is calculated to characterize the system's recovery capability.
[0005] Preferably, the dynamic resilience index fusion module is specifically used for: The convergence rate of the discretely calculated state-space trajectory is converted into a continuous-time convergence rate signal through an event update and hold strategy. The convergence rate signal over continuous time is normalized using a sigmoid mapping function to generate a dimensionless restoring force score. The dynamic resilience index is generated by weighted fusion of the cross-modal phase synchronization index, the resilience score, and the deviation of the brain functional network modularity coefficient from the individual cognitive baseline.
[0006] Preferably, the early warning generation module is further used to calculate the time derivative of the dynamic resilience index; The warning generation module determines the warning level, including: When the dynamic resilience index is lower than the first warning threshold, the warning level is determined to be a Level 1 warning. When the dynamic resilience index is lower than the second warning threshold, the warning level is determined to be a level two warning, wherein the second warning threshold is lower than the first warning threshold; When the dynamic resilience index is lower than the third warning threshold or the absolute value of the time derivative exceeds the rate threshold, the warning level is determined to be a level three warning, wherein the third warning threshold is lower than the second warning threshold.
[0007] Preferred options also include: The weight dynamic configuration module is used to identify the current flight phase based on the flight parameters and select the weight parameters that match the current flight phase for the dynamic resilience index fusion module.
[0008] Preferably, the weighting parameters include a first weight characterizing the cardiopulmonary coupling strength, a second weight characterizing the system's recovery capacity, and a third weight for assessing cognitive load; The weight dynamic configuration module loads a configuration that increases the second weight during the high-G maneuver phase and loads a configuration that increases the third weight during the instrument approach phase.
[0009] Preferably, the physiological state vector consists of heart rate and heart rate variability index; The dynamic baseline point is the stable centroid determined by the physiological state vector during the low-load phase of flight.
[0010] Preferably, the reference convergence rate in the S-shaped mapping function is determined based on statistical data from a large number of pilots in baseline conditions.
[0011] Compared with the prior art, the present invention has the following beneficial effects: 1. By integrating multiple physiological indicators that assess cardiopulmonary coupling strength, system recovery capacity, and cognitive load status, this invention constructs a comprehensive dynamic resilience index. This multi-dimensional assessment mechanism overcomes the shortcomings of traditional methods that rely on only a single physiological indicator and are easily interfered with, significantly reducing the false alarm rate and enhancing the robustness and reliability of the early warning results. 2. By introducing the analysis of the rate of change of comprehensive physiological state indicators and assessing the pilot's systemic recovery ability after experiencing a high-G impact, this invention can capture the dynamic trend of physiological state deterioration; this changes the traditional method of triggering an alarm only after physiological abnormalities occur, and realizes the advance prediction of physiological disability risk, giving pilots and ground personnel valuable reaction time. 3. This invention establishes a multi-level early warning mechanism based on physiological state levels and their rate of change, which can generate different levels of early warning according to the severity and urgency of the risk. This refined risk classification avoids the coarseness of traditional single-threshold early warning, making the early warning information more hierarchical and operable, and improving the efficiency of risk management. 4. This invention introduces a dynamic weighting configuration mechanism, which can automatically adjust the weight of various physiological indicators in the comprehensive evaluation according to the current flight phase. For example, it emphasizes system recovery capability during high-G maneuvering phases and cognitive load during instrument approach phases. This adaptive adjustment enables the monitoring system to focus on the most critical sources of physiological risk under different flight missions, significantly improving the pertinence and accuracy of early warning. Attached Figure Description
[0012] The present invention will be further explained below with reference to the accompanying drawings and embodiments: Figure 1 This is a structural diagram of the system of the present invention. Detailed Implementation
[0013] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.
[0014] Example 1: Please see Figure 1A physiological state monitoring and early warning system for trainer aircraft pilots, comprising: The data acquisition and preprocessing module is used to acquire the pilot's multimodal physiological signals and flight parameters, and process them to generate RR interval sequences; this module is also used to generate continuous respiratory signals, generate multi-lead EEG signals, and generate synchronized G-overload values; The autonomic nervous system resilience assessment module is used to calculate the cross-modal phase synchronization index based on RR interval sequences and continuous respiratory signals; calculate the convergence rate of the state space trajectory based on G overload value and physiological state vector; and calculate the modular coefficient of brain functional network based on multi-lead EEG signals. The dynamic resilience index fusion module is used to combine the cross-modal phase synchronization index, the convergence rate of the state space trajectory and the modular coefficient of brain functional network, and generate dynamic resilience index according to preset weight parameters. The early warning generation module is used to determine the early warning level based on dynamic resilience indicators.
[0015] This invention discloses a physiological state monitoring and early warning system for trainer aircraft pilots. The system aims to solve the problems of delayed early warning and high false alarm rate caused by relying on static threshold monitoring of a single physiological indicator in the prior art. By deeply analyzing the intrinsic correlation of multiple physiological signals, it can achieve advanced prediction of the risk of physiological disability. A physiological state monitoring and early warning system for trainer aircraft pilots, comprising: The data acquisition and preprocessing module aims to acquire and generate a high-quality, synchronized data stream for subsequent analysis from multi-source heterogeneous signals in real time. In this embodiment, the module acquires the pilot's multimodal physiological signals and flight parameters in real time through airborne sensors such as wearable ECG, photoelectric, and EEG sensors and a flight data bus. Specifically, the module's functions include: acquiring ECG signals; detecting R-wave peaks and calculating the time interval between adjacent R-waves to generate an RR interval sequence. The RR interval sequence refers to a continuous time record of the heartbeat cycle, serving as a reference for heart rate and heart rate variation. The basic data for heterogeneity analysis comes from the raw ECG signal after artifact correction. Continuous respiratory signals are reconstructed by filtering and feature extraction of the photoplethysmography (PPG) signal, which provides respiratory phase information for subsequent cardiopulmonary coupling analysis. The raw EEG signal is filtered, denoised, and artifact-corrected to output high-quality multi-lead EEG signals, which provide information on brain functional activity for cognitive load analysis. The G-overload value obtained from the flight data bus is precisely aligned with the above physiological signals on the time axis to ensure the synchronization of flight events and physiological responses. The autonomic nervous system resilience assessment module aims to quantitatively evaluate the regulatory resilience of a pilot's autonomic nervous system in parallel across three dimensions: cardiopulmonary coordination, systemic resilience, and cognitive efficiency. This module receives data from the data acquisition and preprocessing module and performs the following calculations: based on the RR interval sequence and continuous respiratory signals, it uses a nonlinear time series analysis method to calculate the cross-modal phase synchronization index. This index is a... The dimensionless scalar within the interval directly characterizes the strength of cardiopulmonary coupling; based on the G overload value and the physiological state vector, the convergence rate of the state-space trajectory is calculated. The physical dimensions of this rate are The larger the value, the faster the system recovers from disturbances; based on multi-lead EEG signals, graph theory methods are used to calculate the modularity coefficient of brain functional networks with functional connectivity. This coefficient is a dimensionless scalar, and its deviation from an individual's baseline reflects changes in cognitive state. The purpose of the dynamic resilience index fusion module is to integrate the three dimensions of assessment information, which are different in nature, into a single, comprehensive dynamic resilience index. This module aims to achieve a comprehensive assessment of the pilot's overall physiological state. It performs continuous and normalized processing on the convergence rate of the discrete-time state-space trajectory, then combines the baseline deviation of the cross-modal phase synchronization index and the modular coefficients of the brain functional network with a set of preset weighting parameters for linear weighted fusion, ultimately generating a dynamic resilience index. ; The early warning generation module aims to provide graded early warnings for the risk of physiological disability based on dynamic resilience indicators; this module continuously monitors dynamic resilience indicators. The value is used to determine the warning level based on its relationship with a set of preset thresholds. When the temperature falls below a certain threshold, the system will generate a warning signal of the corresponding level. This embodiment achieves advanced prediction of pilot physiological disability risk by constructing a complete computational framework from multi-dimensional dynamic characteristic assessment to comprehensive index fusion. Compared with existing technologies, this system can capture early signals of system instability caused by cardiopulmonary decoupling, decreased system resilience, or cognitive overload, rather than waiting for obvious abnormalities in physiological indicators. Its comprehensive assessment mechanism integrates information from the cardiovascular, respiratory, and central nervous systems, enhancing the robustness and reliability of the assessment results. By dynamically quantifying the pilot's physiological resilience, this invention can provide earlier and more accurate warnings, effectively reducing flight safety risks caused by physiological dysfunction. In comparative tests under simulated flight scenarios, compared with traditional warning methods that rely solely on static physiological indicator thresholds, the system described in this invention can advance the effective warning time of physiological disability risk by an average of 15 to 30 seconds, while reducing the false alarm rate caused by motion artifacts or fluctuations in single indicators by about 25%, thereby significantly improving the effectiveness of flight safety assurance.
[0016] Example 2: The autonomic nervous system resilience assessment module calculates the convergence rate of the state-space trajectory, including: In response to the G overload value exceeding the preset impact threshold, the Euclidean distance between the physiological state vector and the dynamic baseline point is tracked to decay. Based on the Euclidean distance decay process, the convergence rate of the state-space trajectory is calculated to characterize the system's recovery capability.
[0017] The physiological state vector consists of heart rate and heart rate variability index; The dynamic baseline point is the stable centroid determined by the physiological state vector during the low-load phase of flight.
[0018] Based on Example 1, this embodiment specifies the process for calculating the convergence rate of the state space trajectory in the autonomic nervous system resilience assessment module; this specification ensures the objectivity and repeatability of assessing the dynamic recovery ability of the pilot's physiological system. To calculate the convergence rate of the state-space trajectory, the implementation process is limited to: Construct an m-dimensional physiological state vector Physiological state vector This refers to a mathematical construct used to characterize a pilot's current physiological state in a multidimensional space, whose function is to comprehensively reflect the instantaneous state of the physiological system; in this embodiment, the physiological state vector is composed of heart rate and heart rate variability index, for example, its components It can be heart rate (HR) and time-domain heart rate variability index (SDNN) calculated in real time from RR interval sequences; During low-load phases of flight, such as level cruise, the system determines the stable centroid of the state vector, which serves as the dynamically updated dynamic baseline point. Dynamic baseline points refer to reference points for pilots in a physiologically stable state. Their purpose is to provide a benchmark for assessing deviations and the recovery process; they are derived from data collected during low-load phases. The data is obtained by statistical averaging. In response to the G overload value exceeding the preset impact threshold, the Euclidean distance decay process between the physiological state vector and the dynamic baseline point is tracked; preset impact threshold This refers to a G-value threshold used to identify significant external shock events such as high-G maneuvers. Its function is to trigger restoring force analysis, and it is derived from typical maneuvering maneuvers in the trainer aircraft training syllabus, for example, set to 4G. When the G-value exceeds this threshold, the system identifies a shock event and starts from this event point. Begin by continuously calculating the state vector. With baseline point Euclidean distance between ; Based on the Euclidean distance decay process, the convergence rate of the state-space trajectory is calculated to characterize the system's recovery capability; the decay process of this distance can be derived from an exponential model. To describe; through actual observation By fitting the sequence using the least squares method, the convergence rate of the state-space trajectory can be calculated. This parameter The larger the value, the faster the state vector returns to the baseline, thus objectively quantifying the physiological system's recovery ability after being subjected to shock; Through the above specific implementation methods, this invention provides a clear and operable computational path for the key feature of state-space trajectory convergence rate; by defining high-G maneuvering as an external disturbance and tracking the Euclidean distance decay process of the multidimensional physiological state vector, the assessment of recovery ability is no longer fuzzy and qualitative, but specific and quantifiable; in particular, by constructing the state vector from heart rate and heart rate variability indicators and using dynamically updated baseline points, the physiological significance of the assessment and its adaptability to individual differences are further enhanced, thereby bringing about a gain effect of more accurate assessment results.
[0019] Example 3: The dynamic resilience index fusion module is specifically used for: The convergence rate of the discrete-time state-space trajectory is converted into a continuous-time convergence rate signal through an event update and hold strategy. The convergence rate signal over continuous time is normalized using a sigmoid mapping function to generate a dimensionless restoring force score. We weighted and fused the deviations of the cross-modal phase synchronization index, resilience score, and brain functional network modularity coefficient from the individual cognitive baseline to generate a dynamic resilience index.
[0020] The reference convergence rate in the S-shaped mapping function is determined based on statistical data from a large number of pilots under baseline conditions.
[0021] Based on Example 1, this embodiment provides a detailed explanation of the specific implementation method of the dynamic resilience index fusion module. To ensure the mathematical rigor and logical consistency of the fusion process, this implementation method aims to solve the inconsistency problem in data characteristics and units of evaluation information from different dimensions, thereby achieving scientific and effective fusion. The specific implementation method of the dynamic resilience index fusion module is as follows: To address the data type mismatch issue, the convergence rate of the discrete-time state-space trajectory is converted into a continuous-time convergence rate signal using an event-driven update and hold strategy. This is because the convergence rate of the state-space trajectory... It is calculated only once after each high-G impact event, and is a discrete value that cannot be directly compared to a continuous one. This module integrates various indicators; therefore, it employs an event update and hold strategy to generate a continuous time signal. The implementation mechanism is as follows: the system maintains an internal variable whose value remains unchanged when no new high-G impact event occurs; when the state-space restoring force analysis submodule calculates a new value at a certain moment... After the value is changed, the internal variable is immediately updated to the new value; thus, A step function is formed, whose value at any given time is equal to the calculated result of the most recent impact event; to ensure that this signal is defined during the initial stage of the flight mission or during prolonged periods without high-G impacts, the system will... An initial value is assigned, which can be determined based on the pilot's historical data or the statistical mean of a similar group of pilots; To address the issue of dimension inconsistency, a sigmoid mapping function is used to normalize the continuous-time convergence rate signal to generate a dimensionless restoring force score; the dimension of $α_{conv}(t)$ is... Since there are dimensional differences compared to other dimensionless indices, this embodiment uses a sigmoid mapping function to achieve this transformation, the formula of which is: ; in, For resilience scoring; It is a continuous-time convergence rate signal; The convergence rate, used as a reference, is determined based on statistical data from a large number of pilots under baseline conditions; for example, it can be calculated from all valid events. The 50th percentile of the value distribution; The kurtosis parameter of the mapping function, measured in seconds (s), serves to ensure that the exponential term is dimensionless while adjusting the scoring effect. Sensitivity to change; To ensure the system's robustness in real flight environments, this module also incorporates data quality assessment and fault tolerance mechanisms. Before fusion computation, the system evaluates the validity of each input signal. When a physiological characteristic of a certain modality becomes invalid within the current time window, the system will initiate a degraded fusion strategy. Under this strategy, the weight of the invalid feature is temporarily set to zero, and its weight is distributed to the remaining valid features according to a preset ratio. Simultaneously, the system records and indicates that it is currently in degraded evaluation mode, thereby ensuring dynamic resilience indicators. The continuity and availability of the data are assessed. Based on the above processing results, the deviations of the cross-modal phase synchronization index, resilience score, and brain functional network modularity coefficient from the individual cognitive baseline are weighted and fused to generate a dynamic resilience index. Moving averages are applied to each input parameter to obtain a smooth value. The following fusion formula is constructed: ; in, It is a dynamic resilience index; This is the smoothed value of the cardiopulmonary coupling index; This is a smoothed value for the resilience score; The deviation of cognitive load, where It is an individual cognitive baseline calculated by collecting EEG signals during the resting or standardized testing phase before a flight mission; These are weight parameters; This embodiment cleverly solves the problem of integrating discrete, dimensional resilience indicators with other continuous, dimensionless physiological indicators by introducing an event update and maintenance strategy and a sigmoid mapping function. This not only makes the final dynamic resilience index more mathematically rigorous, but also determines it based on a large amount of statistical data. This ensures the standardization and comparability of the scoring; the final weighted fusion formula clearly integrates information from the three dimensions, generating a comprehensive, dynamic, and logically consistent physiological state assessment index, which greatly improves the comprehensiveness and accuracy of the assessment results.
[0022] Example 4: The early warning generation module is also used to calculate the time derivative of the dynamic resilience index; The early warning generation module determines the early warning level, including: When the dynamic resilience index is lower than the first warning threshold, the warning level is determined to be Level 1 warning. When the dynamic resilience index is lower than the second warning threshold, the warning level is determined to be a level two warning, wherein the second warning threshold is lower than the first warning threshold; When the dynamic resilience index is lower than the third warning threshold or the absolute value of the time derivative exceeds the rate threshold, the warning level is determined to be a level three warning, wherein the third warning threshold is lower than the second warning threshold.
[0023] Based on Example 1, this embodiment specifies the method by which the early warning generation module determines the early warning level, and introduces time derivative and multi-level thresholds, thereby constructing a more refined and sensitive early warning mechanism. The early warning generation module receives the dynamic resilience index calculated by the dynamic resilience index fusion module. Then, it is further used to calculate the time derivative of the dynamic resilience index. Time derivative This refers to the rate of change of the dynamic resilience index over time. Its function is to capture the trend and speed of deterioration in a pilot's physiological state. Its source is... The time series is obtained by difference calculation; Based on this, the early warning generation module determines the early warning level, including a level based on... Absolute value and its rate of change Multi-level decision-making logic: When the dynamic resilience index falls below the first warning threshold, the warning level is determined to be Level 1; the first warning threshold It is a threshold used to identify mild deviations from a physiological state; when At that time, the system triggered a Level 1 warning; When the dynamic resilience index falls below the second warning threshold, the warning level is determined to be Level II, where the second warning threshold is lower than the first warning threshold; the second warning threshold... It is a threshold used to identify moderate deterioration of physiological state; when At that time, the system triggered a level-two warning; When the dynamic resilience index falls below the third warning threshold or the absolute value of the time derivative exceeds the rate threshold, the warning level is determined to be Level III, where the third warning threshold is lower than the second warning threshold; the third warning threshold... and rate threshold Together they are used to identify severe or acute deterioration of physiological conditions; when or At that time, the system triggered the highest level, Level 3, early warning; All threshold parameters The sources are all determined based on statistical analysis of a large amount of historical flight data; in this embodiment, these thresholds are set at specific percentiles of the corresponding data distribution to ensure their statistical significance and the reliability of the warning. This embodiment introduces the time derivative of the dynamic resilience index, enabling the early warning system to respond not only to the level of physiological state but also to the rate of change in state. This dual judgment criterion of absolute value and rate of change significantly improves the sensitivity and foresight of the early warning, allowing for earlier detection of rapidly deteriorating physiological states and providing pilots and ground personnel with more valuable reaction time. The tiered early warning mechanism also makes the alarm information more hierarchical and operable, avoiding the crude all-or-nothing warnings brought about by a single threshold, thereby achieving more refined and efficient risk management.
[0024] Example 5: It also includes a weighted dynamic configuration module, which is used to identify the current flight phase based on flight parameters and select weight parameters that match the current flight phase for the dynamic resilience index fusion module.
[0025] The weighting parameters include a first weight characterizing the strength of cardiopulmonary coupling, a second weight characterizing the system's recovery capacity, and a third weight assessing cognitive load. The weighted dynamic configuration module loads a configuration that increases the second weight during the high-G maneuver phase and a configuration that increases the third weight during the instrument approach phase.
[0026] Based on the system of Example 1, this embodiment introduces a functional module to achieve adaptive adjustment and optimizes the fusion process of dynamic resilience indicators so that it can adjust according to the real-time changes of the flight mission. The system further includes a weighted dynamic configuration module, which aims to intelligently select and match the most suitable weight parameters for the dynamic resilience index fusion module based on the pilot's current flight phase; this module identifies the current flight phase by analyzing flight parameters such as G-value, altitude, and speed obtained from the flight data bus. When calculating the dynamic resilience index, the weighting parameters used include the first weight, which characterizes the cardiopulmonary coupling strength. The second weight characterizing the system's resilience And the third weight for assessing cognitive load. ; The specific working logic of the weight dynamic configuration module is as follows: During high-G maneuvers, the second weight is increased. The configuration; its technical principle lies in the fact that the main impact on the pilot's physiological system during high-G maneuvers comes from G-force overload, and the stability of the cardiovascular system and its recovery ability after being disturbed are key to determining its physiological state. Therefore, the system automatically increases loading. The weighting configuration makes the dynamic resilience index It is more sensitive to changes in the convergence rate of the state-space trajectory; During the instrument approach phase, the third weight is increased. The configuration; its technical principle lies in the fact that during the complex decision-making and delicate operation phase of instrument approach, the pilot's cognitive load is the main source of physiological stress, so the system automatically switches to increasing... The weight configuration makes It is more sensitive to the degree to which the modularity coefficient of brain functional networks deviates from the baseline; The specific weight configuration parameters for each flight phase are obtained through supervised learning training optimization using a large amount of simulator flight data and real training data; By adding a dynamic weighting configuration module, the monitoring and early warning system of this invention evolves from a static model into a dynamic intelligent system that can adapt to mission requirements. It recognizes that different flight phases have different emphases on the pilot's physiological capabilities and makes adaptive adjustments accordingly. This dynamic weighting mechanism enables the fused dynamic resilience index to more accurately reflect the most critical sources of physiological risk in the current flight phase, thereby significantly improving the pertinence and accuracy of early warnings, bringing timely and appropriate technical benefits, and further reducing the possibility of false alarms and missed alarms.
[0027] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims
1. A physiological state monitoring and early warning system for trainer aircraft pilots, characterized in that, include: The data acquisition and preprocessing module is used to acquire the pilot's multimodal physiological signals and flight parameters, and process them to generate RR interval sequences; this module is also used to generate continuous respiratory signals, generate multi-lead EEG signals, and generate synchronized G-overload values; The autonomic nervous system resilience assessment module is used to calculate the cross-modal phase synchronization index based on the RR interval sequence and the continuous respiratory signal; and the autonomic nervous system resilience assessment module is used to calculate the convergence rate of the state space trajectory based on the G overload value and the physiological state vector; and the autonomic nervous system resilience assessment module is used to calculate the modular coefficient of the brain functional network based on the multi-lead EEG signal. The dynamic resilience index fusion module is used to combine the cross-modal phase synchronization index, the state space trajectory convergence rate, and the brain functional network modularity coefficient, and generate a dynamic resilience index based on preset weight parameters. The early warning generation module is used to determine the early warning level based on the dynamic resilience index.
2. The physiological state monitoring and early warning system for trainer aircraft pilots according to claim 1, characterized in that, The autonomic nervous system resilience assessment module calculates the convergence rate of the state-space trajectory, including: In response to the G overload value exceeding a preset impact threshold, the Euclidean distance attenuation process between the physiological state vector and the dynamic baseline point is tracked. Based on the Euclidean distance decay process, the convergence rate of the state space trajectory is calculated to characterize the system's recovery capability.
3. The physiological state monitoring and early warning system for trainer aircraft pilots according to claim 1, characterized in that, The dynamic resilience index fusion module is specifically used for: The convergence rate of the discretely calculated state-space trajectory is converted into a continuous-time convergence rate signal through an event update and hold strategy. The convergence rate signal over continuous time is normalized using a sigmoid mapping function to generate a dimensionless restoring force score. The dynamic resilience index is generated by weighted fusion of the cross-modal phase synchronization index, the resilience score, and the deviation of the brain functional network modularity coefficient from the individual cognitive baseline.
4. The physiological state monitoring and early warning system for trainer aircraft pilots according to claim 1, characterized in that, The early warning generation module is also used to calculate the time derivative of the dynamic resilience index; The warning generation module determines the warning level, including: When the dynamic resilience index is lower than the first warning threshold, the warning level is determined to be a Level 1 warning. When the dynamic resilience index is lower than the second warning threshold, the warning level is determined to be a level two warning, wherein the second warning threshold is lower than the first warning threshold; When the dynamic resilience index is lower than the third warning threshold or the absolute value of the time derivative exceeds the rate threshold, the warning level is determined to be a level three warning, wherein the third warning threshold is lower than the second warning threshold.
5. The physiological state monitoring and early warning system for trainer aircraft pilots according to claim 1, characterized in that, Also includes: The weight dynamic configuration module is used to identify the current flight phase based on the flight parameters and select the weight parameters that match the current flight phase for the dynamic resilience index fusion module.
6. The physiological state monitoring and early warning system for trainer aircraft pilots according to claim 5, characterized in that, The weighting parameters include a first weight characterizing the cardiopulmonary coupling strength, a second weight characterizing the system's recovery capacity, and a third weight for assessing cognitive load. The weight dynamic configuration module loads a configuration that increases the second weight during the high-G maneuver phase and loads a configuration that increases the third weight during the instrument approach phase.
7. A physiological state monitoring and early warning system for trainer aircraft pilots according to claim 2, characterized in that, The physiological state vector is composed of heart rate and heart rate variability index; The dynamic baseline point is the stable centroid determined by the physiological state vector during the low-load phase of flight.
8. A physiological state monitoring and early warning system for trainer aircraft pilots according to claim 3, characterized in that, The reference convergence rate in the S-shaped mapping function is determined based on statistical data from a large number of pilots in baseline conditions.