A pilot behavior detection method and system, an aircraft

Abstract

The application provides a pilot behavior detection method and system and an aircraft, and particularly relates to the technical field of flight, and comprises the following steps: calculating flight control quantity into control target flight parameters, and determining the real-time flight mode of the aircraft based on the real-time flight parameters of the sensor at the current time; matching an abnormal control judgment matrix according to the real-time flight mode, and inputting the control target flight parameters and the real-time flight parameters into the abnormal control judgment matrix to calculate a risk score; performing flight state evaluation on the aircraft based on the risk score, and determining that the pilot has abnormal behavior when the flight state is abnormal; or determining that the pilot does not have abnormal behavior when the flight state is normal. The application can identify whether the pilot has abnormal control by performing combined operation on the real-time flight parameters and the control target flight parameters; the application does not need to perform a large number of simulation tests or model training in advance, the calculation amount is small, and the demand and burden of the on-board processor operation amount are low.

Description

Technical Field

[0001] This application relates to the field of flight technology, and in particular to a pilot behavior detection method and system, and an aircraft. Background Technology

[0002] Aircraft operational safety has always been a focus of global attention. Since the birth of the aviation industry, countries around the world have been constantly seeking various methods and measures to improve flight safety. To date, these methods and measures are almost all designed for the reliability of aircraft hardware or software operation. As the core element in controlling aircraft safety, it is an indisputable fact that the behavior and rights of pilots need to be monitored and constrained. For this reason, there are currently a variety of schemes for monitoring pilots. For example, the existing literature CN115783278A proposes a pilot operation monitoring method based on "threshold-feature-result" matching. However, this literature has a narrow logical coverage when judging abnormal pilot operation, and it requires (2) to use a six-degree-of-freedom model for real-time calculation, which has a large amount of computation and a heavy load on the airborne processor. In addition, the existing literature CN110712765A proposes an aircraft abnormal operation localization method based on operation spectrum. However, this literature mainly focuses on the tracing of the pilot's historical operation, rather than the abnormal detection of the pilot's real-time operation, and has poor real-time performance.

[0003] Therefore, designing a solution with broad logical coverage that can detect pilots' real-time maneuvering behavior is a problem that urgently needs to be solved. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the purpose of this application is to provide a pilot behavior detection method and system, and an aircraft, to solve the problems of insufficient logic coverage and poor real-time performance in the prior art when judging abnormal pilot operation.

[0005] To achieve the above and other related objectives, this application provides a method for detecting pilot behavior, the method comprising the following steps:

[0006] In response to pilot input, the data processing device outputs target flight parameters based on the pilot input.

[0007] The flight mode of the aircraft at the current moment is determined based on the real-time flight parameters at the current moment, and is denoted as the real-time flight mode.

[0008] The abnormal manipulation judgment matrix is ​​matched according to the real-time flight mode, and the flight parameters of the manipulation target and the real-time flight parameters are input into the abnormal manipulation judgment matrix to calculate the risk score;

[0009] The flight status of the aircraft is assessed based on the risk score, and if the flight status is abnormal, it is determined that the pilot has abnormal behavior; or, if the flight status is normal, it is determined that the pilot has no abnormal behavior.

[0010] In one embodiment of this application, the process of inputting the flight parameters of the manipulated target and the real-time flight parameters into the abnormal manipulation judgment matrix to calculate the risk score includes:

[0011] Obtain the abnormal manipulation judgment state quantity corresponding to the abnormal manipulation judgment matrix;

[0012] The target flight parameters are used as the horizontal input to the abnormal control judgment matrix, and the real-time flight parameters are used as the vertical input to the abnormal control judgment matrix; or, the target flight parameters are used as the vertical input to the abnormal control judgment matrix, and the real-time flight parameters are used as the horizontal input to the abnormal control judgment matrix; or, the target flight parameters are used as both the horizontal and vertical inputs to the abnormal control judgment matrix.

[0013] All horizontal and vertical inputs are combined in pairs according to the abnormal manipulation judgment state variables, and a risk score is calculated for each combination of abnormal manipulation judgment state variables.

[0014] The risk score of the abnormal manipulation judgment matrix is ​​generated based on the risk score of each abnormal manipulation judgment state quantity combination.

[0015] In one embodiment of this application, the process of generating a risk score for the abnormal manipulation judgment matrix based on the risk score of each combination of abnormal manipulation judgment state variables includes:

[0016] The risk scores for all combinations of abnormal manipulation judgment state variables in each horizontal input are summed to obtain the summary risk score for each horizontal input, denoted as the first-level summary risk score; and,

[0017] The first-level summary risk score of all horizontal inputs is used as the risk score of the abnormal manipulation judgment matrix.

[0018] In one embodiment of this application, the process of assessing the flight status of the aircraft based on the risk score further includes:

[0019] The current / historical / future flight status of the aircraft is assessed based on the first-level aggregated risk score for each lateral input and / or the first-level aggregated risk score for all lateral inputs.

[0020] In one embodiment of this application, the real-time flight anomaly level of the aircraft is determined based on a first-level aggregated risk score;

[0021] The real-time flight anomaly level is compared with the flight anomaly level reference table, and the flight status of the aircraft is determined to be abnormal or normal based on the comparison result.

[0022] In one embodiment of this application, the process of generating the risk score of the abnormal manipulation judgment matrix based on the risk score of each abnormal manipulation judgment state variable combination further includes:

[0023] The first-level aggregated risk scores of all horizontal inputs are summed to obtain the second-level aggregated risk scores;

[0024] The secondary summary risk score is used as the risk score of the abnormal manipulation judgment matrix.

[0025] In one embodiment of this application, the process of assessing the flight status of the aircraft based on the risk score further includes:

[0026] The current / historical / future flight status of the aircraft is assessed based on the secondary summary risk score.

[0027] In one embodiment of this application, the process of assessing the flight status of the aircraft based on the risk score further includes:

[0028] The real-time flight anomaly level of the aircraft is determined based on the secondary aggregated risk score;

[0029] The real-time flight anomaly level is compared with the flight anomaly level reference table, and the flight status of the aircraft is determined to be abnormal or normal based on the comparison result.

[0030] This application also provides a pilot behavior detection system, the system comprising:

[0031] The control response module is used to respond to the pilot's control input, and the data processing device outputs control target flight parameters based on the pilot's control input;

[0032] The flight mode module is used to determine the flight mode of the aircraft at the current moment based on the real-time flight parameters at the current moment, which is denoted as the real-time flight mode;

[0033] The risk score calculation module is used to match the abnormal operation judgment matrix according to the real-time flight mode, and input the flight parameters of the manipulation target and the real-time flight parameters into the abnormal operation judgment matrix to calculate the risk score;

[0034] The behavior detection module is used to assess the flight status of the aircraft based on the risk score, and determine that the pilot has abnormal behavior when the flight status is abnormal; or, determine that the pilot has no abnormal behavior when the flight status is normal.

[0035] This application also provides an aircraft that is used in any of the pilot behavior detection methods described above.

[0036] As described above, this application provides a pilot behavior detection method and system, and an aircraft, with the following beneficial effects: In response to pilot input, a data processing device outputs target flight parameters based on the pilot's input; then, based on the real-time flight parameters, it determines the aircraft's flight mode at the current moment, denoted as the real-time flight mode; it matches an abnormal control judgment matrix according to the real-time flight mode, and inputs the target flight parameters and real-time flight parameters into the abnormal control judgment matrix to calculate a risk score; based on the risk score, it assesses the aircraft's flight status, and if the flight status is abnormal, it determines that the pilot has abnormal behavior; or, if the flight status is normal, it determines that the pilot does not have abnormal behavior. Therefore, this application, by combining and calculating the real-time flight parameters and the target flight parameters, covers the flight status within the range described by the pilot's input flight parameters, thus identifying whether the pilot has abnormal control. Furthermore, this application does not require extensive prior simulation experiments or model training; it only needs to set thresholds and coefficients, and then identify pilot abnormal control through algebraic operations. This not only reduces the computational load but also lowers the computational requirements and burden on the airborne processor. Furthermore, the scope of this application's judgment on abnormal control depends on the manually selected parameter range, and can be limited to abnormal control of a single channel or broadly encompass abnormalities of all coupled channels. In addition, this application can provide abnormal control warnings for past, present, and future scenarios through differential, integral, and proportional adjustments, effectively detecting reciprocating control and preventing control resonance from threatening flight safety. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a schematic flowchart of a pilot behavior detection method provided in one embodiment of this application;

[0039] Figure 2 This is a schematic diagram of the pilot abnormal operation judgment logic provided in one embodiment of this application;

[0040] Figure 3 This is a schematic diagram illustrating the determination of an abnormal manipulation judgment matrix provided in one embodiment of this application;

[0041] Figure 4 This is a schematic diagram illustrating the calculation of a risk score according to an embodiment of this application;

[0042] Figure 5 A schematic diagram illustrating a flight status assessment provided in one embodiment of this application;

[0043] Figure 6 This is a schematic diagram of the hardware structure of a pilot behavior detection system provided in one embodiment of this application. Detailed Implementation

[0044] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0045] It should be noted that the illustrations provided in this embodiment are only schematic representations of the basic concept of this application. Therefore, the drawings only show the components related to this application and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0046] In one embodiment of this application, please refer to Figure 1 As shown, this embodiment provides a pilot behavior detection method, including the following steps:

[0047] S110, in response to the pilot's control input, the data processing device outputs and controls the target flight parameters based on the pilot's control input; wherein, the data processing device in this embodiment includes, but is not limited to, flight control system, controller, etc.

[0048] S120, based on the real-time flight parameters at the current moment, determines the flight mode of the aircraft at the current moment, denoted as the real-time flight mode;

[0049] S130 matches the abnormal manipulation judgment matrix according to the real-time flight mode, and inputs the flight parameters of the manipulation target and the real-time flight parameters into the abnormal manipulation judgment matrix to calculate the risk score;

[0050] S140 assesses the flight status of the aircraft based on a risk score, and determines that the pilot has engaged in abnormal behavior when the flight status is abnormal; or determines that the pilot has not engaged in abnormal behavior when the flight status is normal.

[0051] Therefore, this embodiment, by combining and calculating real-time flight parameters and target flight parameters, covers the flight states within the range of flight parameters input by the pilot, thus identifying any abnormal pilot maneuvers. Furthermore, this embodiment does not require extensive prior simulation experiments or model training; it only needs to set thresholds and coefficients, and then identify pilot maneuvering anomalies through algebraic calculations. This not only reduces computational load but also minimizes the computational demands and burden on the onboard processor. Moreover, the scope of abnormal maneuvering judgment in this embodiment depends on the manually selected parameter range, and can be applied to abnormal control in a single channel or broadly encompass anomalies in all coupled channels. In addition, this embodiment, through differential, integral, and proportional adjustments, can provide abnormal maneuvering warnings for past, present, and future scenarios, effectively detecting reciprocating maneuvers and preventing control resonance from threatening flight safety.

[0052] According to the above description, in an exemplary embodiment, the process of inputting the target flight parameters and real-time flight parameters into the abnormal manipulation judgment matrix to calculate the risk score in step S130 may include: obtaining the abnormal manipulation judgment state quantity corresponding to the abnormal manipulation judgment matrix; using the target flight parameters as the horizontal input of the abnormal manipulation judgment matrix and the real-time flight parameters as the vertical input of the abnormal manipulation judgment matrix; combining all horizontal inputs and all vertical inputs in pairs according to the abnormal manipulation judgment state quantity, and calculating the risk score for each combination of abnormal manipulation judgment state quantities; and generating the risk score of the abnormal manipulation judgment matrix based on the risk score of each combination of abnormal manipulation judgment state quantities.

[0053] As an example, the process of generating a risk score for the abnormal manipulation judgment matrix based on the risk score of each abnormal manipulation judgment state variable combination includes: summing the risk scores of all abnormal manipulation judgment state variable combinations in each lateral input to obtain a summary risk score for each lateral input, denoted as the first-level summary risk score; and using the first-level summary risk scores of all lateral inputs as the risk score of the abnormal manipulation judgment matrix. As another example, the process of generating a risk score for the abnormal manipulation judgment matrix based on the risk score of each abnormal manipulation judgment state variable combination may also include: summing the first-level summary risk scores of all lateral inputs to obtain a second-level summary risk score; and using the second-level summary risk score as the risk score of the abnormal manipulation judgment matrix. Therefore, this embodiment can use the first-level summary risk scores of all lateral inputs as the risk score of the abnormal manipulation judgment matrix, or it can sum the first-level summary risk scores of all lateral inputs as the risk score of the abnormal manipulation judgment matrix. Essentially, this embodiment can treat each lateral input as a control channel, and then perform risk scoring for each channel to conduct flight risk assessment. Simultaneously, this embodiment can also sum the first-level summary scores corresponding to all control channels to obtain a second-level summary risk score, which is a level of risk assessment for the entire aircraft control system. The advantage of this method, which divides the system into primary and secondary levels, is that it allows for a more comprehensive risk assessment of both single channels and the entire system.

[0054] According to the above description, in an exemplary embodiment, the process of assessing the flight status of the aircraft based on the risk score in step S140 further includes: assessing the current / historical / future flight status of the aircraft based on the first-level aggregated risk score of each lateral input, and / or the first-level aggregated risk scores of all lateral inputs. As an example, in this embodiment, the abnormal manipulation judgment state quantity corresponding to each lateral input can be used as the target abnormal manipulation judgment state quantity, and the integral threshold, differential threshold, and proportional threshold of the target abnormal manipulation judgment state quantity can be obtained. The first-level aggregated risk score of each lateral input is integrally calculated to obtain the corresponding first-level real-time integral value; and the first-level aggregated risk score of each lateral input is differentially calculated to obtain the corresponding first-level real-time differential value; and the first-level aggregated risk score of each lateral input is proportionally calculated to obtain the corresponding first-level real-time proportional value. For each lateral input corresponding to an abnormal control judgment state quantity, the first-level real-time integral value of each lateral input is compared with the integral threshold of the corresponding target abnormal control judgment state quantity to obtain a first integral comparison result; and the first-level real-time differential value of each lateral input is compared with the differential threshold of the corresponding target abnormal control judgment state quantity to obtain a first differential comparison result; and the first-level real-time proportional value of each lateral input is compared with the proportional threshold of the corresponding target abnormal control judgment state quantity to obtain a first proportional comparison result. Based on the first integral comparison result, the first differential comparison result, and the first proportional comparison result, the current / historical / future flight status of the aircraft is assessed.

[0055] Specifically, the process of assessing the flight status of an aircraft based on the first integral comparison result, the first differential comparison result, and the first proportional comparison result includes: determining the target abnormal control judgment state quantity with integral comparison anomalies using the first integral comparison result; determining the target abnormal control judgment state quantity with differential comparison anomalies using the first differential comparison result; and determining the target abnormal control judgment state quantity with proportional comparison anomalies using the first proportional comparison result. The target abnormal control judgment state quantities with integral comparison anomalies, differential comparison anomalies, and proportional comparison anomalies are correlated, and the real-time flight anomaly level of the aircraft is determined based on the correlation result. The real-time flight anomaly level is compared with a flight anomaly level reference table, and the flight status of the aircraft is determined to be abnormal or normal based on the flight anomaly level comparison result.

[0056] Therefore, this embodiment divides risk scoring into Level 1 (single control quantity), which can obtain single state quantities to assess all risk levels in the past, present, and future, facilitating the location of the control channel where specific abnormal manipulation occurs. Simultaneously, through PID (Proportion Integration Differentiation) control, abnormal manipulation warnings can be issued for the past, present, and future scenarios. Furthermore, the integral term summarizing the Level 1 data also performs reciprocating manipulation detection on each channel to avoid the danger of control resonance.

[0057] According to the above description, in an exemplary embodiment, the process of assessing the flight status of an aircraft based on a risk score may further include: assessing the current / historical / future flight status of the aircraft based on a secondary aggregated risk score. As an example, in this embodiment, the abnormal manipulation judgment state quantity corresponding to each lateral input can be used as the target abnormal manipulation judgment state quantity, and the integral threshold, differential threshold, and proportional threshold of the target abnormal manipulation judgment state quantity can be obtained. The secondary aggregated risk score for each lateral input is integrally calculated to obtain the corresponding secondary real-time integral value; and the secondary aggregated risk score for each lateral input is differentially calculated to obtain the corresponding secondary real-time differential value; and the secondary aggregated risk score for each lateral input is proportionally calculated to obtain the corresponding secondary real-time proportional value. For each lateral input corresponding to an abnormal control judgment state quantity, the second-level real-time integral value of each lateral input is compared with the integral threshold of the corresponding target abnormal control judgment state quantity to obtain a second integral comparison result; and the second-level real-time differential value of each lateral input is compared with the differential threshold of the corresponding target abnormal control judgment state quantity to obtain a second differential comparison result; and the second-level real-time proportional value of each lateral input is compared with the proportional threshold of the corresponding target abnormal control judgment state quantity to obtain a second proportional comparison result. Based on the second integral comparison result, the second differential comparison result, and the second proportional comparison result, the current / historical / future flight status of the aircraft is assessed.

[0058] Specifically, the process of assessing the flight status of an aircraft based on the second integral comparison result, the second differential comparison result, and the second proportional comparison result includes: determining the target abnormal control judgment state quantity with integral comparison anomalies using the second integral comparison result; determining the target abnormal control judgment state quantity with differential comparison anomalies using the second differential comparison result; and determining the target abnormal control judgment state quantity with proportional comparison anomalies using the second proportional comparison result. The target abnormal control judgment state quantities with integral comparison anomalies, differential comparison anomalies, and proportional comparison anomalies are correlated, and the real-time flight anomaly level of the aircraft is determined based on the correlation result. The real-time flight anomaly level is compared with a flight anomaly level reference table, and the flight status of the aircraft is determined to be abnormal or normal based on the flight anomaly level comparison result.

[0059] Therefore, this embodiment uses a two-tiered risk scoring system (overall aircraft control) to assess the overall aircraft status across all risk levels in the past, present, and future, facilitating the identification of specific control channels where abnormal maneuvers occur. Simultaneously, PID control provides early warnings for abnormal maneuvers in the past, present, and future scenarios. Furthermore, the summation of the first-tier aggregated scores for all control channels yields a second-tier aggregated risk score, representing a risk assessment at the overall aircraft control level. The advantage of this hierarchical assessment method is that it allows for a more comprehensive risk assessment of both individual channels and the entire aircraft simultaneously.

[0060] According to the above description, in an exemplary embodiment, the process of inputting the target flight parameters and real-time flight parameters into the abnormal maneuver judgment matrix to calculate the risk score in step S130 may include: obtaining the abnormal maneuver judgment state quantity corresponding to the abnormal maneuver judgment matrix; using the target flight parameters as the vertical input of the abnormal maneuver judgment matrix and the real-time flight parameters as the horizontal input of the abnormal maneuver judgment matrix; combining all horizontal inputs and all vertical inputs pairwise according to the abnormal maneuver judgment state quantity, and calculating the risk score for each combination of abnormal maneuver judgment state quantities; and generating the risk score of the abnormal maneuver judgment matrix based on the risk score of each combination of abnormal maneuver judgment state quantities. In this embodiment, the risk score of the abnormal maneuver judgment matrix is ​​generated based on the risk score of each combination of abnormal maneuver judgment state quantities. The process of generating the risk score of the abnormal maneuver judgment matrix can be found in some of the above embodiments, and will not be repeated here.

[0061] According to the above description, in an exemplary embodiment, the process of inputting the target flight parameters and real-time flight parameters into the abnormal control judgment matrix to calculate the risk score in step S130 may further include: obtaining the abnormal control judgment state quantity corresponding to the abnormal control judgment matrix; using the target flight parameters as the lateral and longitudinal inputs of the abnormal control judgment matrix, so that the target flight parameters can also be combined; combining all lateral and longitudinal inputs pairwise according to the abnormal control judgment state quantity, and calculating the risk score for each combination of abnormal control judgment state quantities; and generating the risk score of the abnormal control judgment matrix based on the risk score of each combination of abnormal control judgment state quantities. In this embodiment, by using the target flight parameters as the lateral and longitudinal inputs of the abnormal control judgment matrix, the target flight parameters can be combined pairwise, for example, pitch control and throttle control can be combined, thereby generating the risk score of the abnormal control judgment matrix based on the risk score of each combination of abnormal control judgment state quantities. The process of generating the risk score of the abnormal control judgment matrix can be found in some of the above embodiments, and will not be repeated here.

[0062] In another exemplary embodiment of this application, the embodiment also provides a method for detecting abnormal pilot maneuvering behavior, such as... Figure 2 As shown, Figure 2 A schematic diagram illustrating the logic for judging abnormal pilot maneuvers is provided. Specifically,

[0063] All parameters input from the sensors first undergo flight mode determination logic to determine the aircraft's current flight mode. For example, tiltrotor aircraft have three main flight modes: rotor mode, tilt transition mode, and fixed-wing mode. This flight mode, along with all sensor data required by the abnormal control determination logic, is then fed into the abnormal control determination matrix. The abnormal control determination matrix has three types of inputs: pilot control input, flight status input, and flight mode input. In this embodiment, the process for determining the abnormal control determination matrix is ​​as follows: Figure 3 As shown. In Figure 3In this embodiment, the flight mode input is used to switch the corresponding matrix. Different flight modes correspond to the abnormal control judgment matrix of the current flight mode. This embodiment switches matrices by distinguishing flight modes, which has the advantage of more accurately and comprehensively covering the entire flight mission profile for abnormal control risk assessment. As an example, in the tiltrotor aircraft example mentioned above, there will be three different abnormal control judgment matrices. Each abnormal control judgment matrix can be regarded as a two-dimensional matrix composed of two vectors: pilot control input and flight state input. In this embodiment or some other embodiments, the abnormal control judgment matrix can also use a judgment matrix with higher or lower dimensions. As an example, the abnormal control judgment matrix can be a one-dimensional judgment matrix for risk assessment of control variables, or a new dimension that adds other variables in addition to control variables and state variables. As another example, the abnormal control judgment matrix in this embodiment can also use multiple parallel judgment matrices. For example, if the designer is concerned about the risk of many variables, but putting them all in the same matrix would result in an excessively large matrix and many unconsidered coupling terms, then it can be split into multiple smaller matrices for parallel assessment.

[0064] Figure 4 A schematic diagram for calculating a risk score is provided. For Figure 4 The matrix is ​​judged by abnormal manipulation. Figure 4 The left dimension (or lateral input) represents the target control variables [p_c, q_c, r_c, u_c, v_c, w_c, throttle_c] calculated from the pilot's input after processing by the flight control system. Here, p_c represents the target roll rate calculated from the pilot's input; q_c represents the target pitch rate calculated from the pilot's input; r_c represents the target yaw rate calculated from the pilot's input; u_c represents the target forward velocity calculated from the pilot's input; v_c represents the target lateral velocity calculated from the pilot's input; w_c represents the target vertical velocity calculated from the pilot's input; and throttle_c represents the target throttle. Figure 4 The upper dimension (or longitudinal input) is the current value of these state variables [p, q, r, u, v, w, throttle]; where p represents the current roll rate; q represents the current pitch rate; r represents the current yaw rate; u represents the current forward velocity; v represents the current lateral velocity; w represents the current vertical velocity; and throttle represents the current throttle position. Figure 4The seven state variables selected are merely examples. In this embodiment or other embodiments, different state variables can be selected to form an abnormal manipulation judgment matrix according to the actual situation. In this embodiment, each target state control value is combined with different current state values ​​to calculate the risk score of the target state control value for the actual state value. For example, the risk assessment of control and state values ​​within the same channel is: f(p,p_c)=a*(p_c-p); where a is the evaluation coefficient, which can be manually set, and p_c and p are the control quantity and current state value within the same channel. The control quantity within the same channel can be angular rate / angle or velocity, etc. The calculation formula is also manually set. For another example, for the risk assessment of coupled channels, the calculation formula can be: f(p,r_c)=a*(b*r_c-p); the above formula is an example of the coupled risk assessment of yaw rate control quantity and roll rate, where a / b are both set coefficients.

[0065] In this embodiment, the risk scores of the corresponding channel state variables and their corresponding coupled state variables for the same control variable are summed (within the same row of the matrix) to obtain the first-level summary risk score, which is for the risk score of a single control variable. The first-level summary scores for all control channels are summed to obtain the second-level summary risk score, which is for risk assessment at the overall system control level. The advantage of this hierarchical assessment method is that it allows for a more comprehensive risk assessment of both single channels and the entire system. It should be noted that the control dimension and state dimension do not need to be two independent dimensions as in the example. For example, control variables can exist in both the upper and left dimensions, allowing for risk assessment between control variables, such as: f(p_c,r_c)=a*(r_c*p_c).

[0066] Figure 5 A schematic diagram for flight status assessment is provided. (For example...) Figure 5As shown, the abnormal manipulation judgment state quantity corresponding to each left dimension is taken as the target abnormal manipulation judgment state quantity, and the integral threshold, differential threshold, and proportional threshold of the target abnormal manipulation judgment state quantity are obtained. The first-level aggregated risk score of each left dimension is integrally calculated to obtain the corresponding first-level real-time integral value; the first-level aggregated risk score of each left dimension is differentially calculated to obtain the corresponding first-level real-time differential value; and the first-level aggregated risk score of each left dimension is proportionally calculated to obtain the corresponding first-level real-time proportional value. According to the abnormal manipulation judgment state quantity corresponding to each left dimension, the first-level real-time integral value of each left dimension is compared with the integral threshold of the corresponding target abnormal manipulation judgment state quantity to obtain the first integral comparison result; the first-level real-time differential value of each left dimension is compared with the differential threshold of the corresponding target abnormal manipulation judgment state quantity to obtain the first differential comparison result; and the first-level real-time proportional value of each left dimension is compared with the proportional threshold of the corresponding target abnormal manipulation judgment state quantity to obtain the first proportional comparison result. The system determines the target abnormal manipulation judgment state quantity with an integral comparison anomaly based on the first integral comparison result; the system determines the target abnormal manipulation judgment state quantity with a differential comparison anomaly based on the first differential comparison result; and the system determines the target abnormal manipulation judgment state quantity with a proportional comparison anomaly based on the first proportional comparison result. The system correlates the target abnormal manipulation judgment state quantities with integral, differential, and proportional comparison anomalies, and determines the real-time flight anomaly level of the aircraft based on the correlation results. The real-time flight anomaly level is compared with a flight anomaly level reference table, and the aircraft's flight state is determined to be abnormal or normal based on the flight anomaly level comparison results. Alternatively, the abnormal manipulation judgment state quantity corresponding to each left dimension is used as the target abnormal manipulation judgment state quantity, and the integral threshold, differential threshold, and proportional threshold of the target abnormal manipulation judgment state quantity are obtained. For each left-hand dimension, the secondary aggregated risk score is integrated to obtain the corresponding secondary real-time integral value; and for each left-hand dimension, the secondary aggregated risk score is differentiated to obtain the corresponding secondary real-time differential value; and for each left-hand dimension, the secondary aggregated risk score is proportionally calculated to obtain the corresponding secondary real-time proportional value. Based on the abnormal manipulation judgment state quantity corresponding to each left-hand dimension, the secondary real-time integral value of each left-hand dimension is compared with the integral threshold of the corresponding target abnormal manipulation judgment state quantity to obtain a second integral comparison result; and for each left-hand dimension, the secondary real-time differential value is compared with the differential threshold of the corresponding target abnormal manipulation judgment state quantity to obtain a second differential comparison result; and for each left-hand dimension, the secondary real-time proportional value is compared with the proportional threshold of the corresponding target abnormal manipulation judgment state quantity to obtain a second proportional comparison result.The system uses the second integral comparison result to determine the target abnormal control judgment state quantity with an integral comparison anomaly; the second differential comparison result to determine the target abnormal control judgment state quantity with a differential comparison anomaly; and the second proportional comparison result to determine the target abnormal control judgment state quantity with a proportional comparison anomaly. It then correlates the target abnormal control judgment state quantities with integral, differential, and proportional comparison anomalies, and determines the real-time flight anomaly level of the aircraft based on the correlation results. Finally, it compares the real-time flight anomaly level with a flight anomaly level reference table, and determines whether the aircraft's flight status is abnormal or normal based on the flight anomaly level comparison results.

[0067] Therefore, this embodiment compares the proportional / integral / derivative thresholds of the two-level risk scores output by the matrix. This process yields assessments of all risk levels for past, present, and future for both individual state variables and the overall state. It also facilitates locating the control channel where specific abnormal manipulations occur. Furthermore, this embodiment performs reciprocating manipulation detection on each channel within the first-level summary integral term to avoid the danger of control resonance.

[0068] In summary, this application provides a pilot behavior detection method. Responding to pilot input, a data processing device outputs target flight parameters based on the pilot's input; then, based on real-time flight parameters, it determines the aircraft's flight mode at the current moment, denoted as the real-time flight mode; it matches an abnormal control judgment matrix according to the real-time flight mode, and inputs the target flight parameters and real-time flight parameters into the abnormal control judgment matrix to calculate a risk score; based on the risk score, it assesses the aircraft's flight status, and if the flight status is abnormal, it determines that the pilot has abnormal behavior; or, if the flight status is normal, it determines that the pilot has no abnormal behavior. Therefore, this method, by combining real-time flight parameters and target flight parameters, covers the flight status within the range described by the pilot's input flight parameters, thus identifying whether the pilot has abnormal control. Furthermore, this method does not require extensive pre-simulation experiments or model training; it only needs to set thresholds and coefficients, and then identify pilot abnormal control through algebraic operations. This not only reduces computational load but also lowers the computational requirements and burden on the airborne processor. Furthermore, the scope of this method's judgment on abnormal control depends on the manually selected parameter range; it can target only single-channel abnormal control or be broadly applicable to anomalies in all coupled channels. In addition, through differential, integral, and proportional adjustments, this method can provide early warnings of abnormal control in past, present, and future scenarios, effectively detecting reciprocating control and preventing control resonance from threatening flight safety.

[0069] like Figure 6As shown, this application also provides a pilot behavior detection system, including:

[0070] The flight parameter module 610 is used to respond to the pilot's control input, and the data processing device outputs and controls the target flight parameters based on the pilot's control input; wherein, the data processing device in this embodiment includes, but is not limited to, the flight control system, the controller, etc.

[0071] The flight mode module 620 is used to determine the flight mode of the aircraft at the current moment based on the real-time flight parameters at the current moment, which is denoted as the real-time flight mode;

[0072] The risk score calculation module 630 is used to match the abnormal operation judgment matrix according to the real-time flight mode, and input the flight parameters of the manipulation target and the real-time flight parameters into the abnormal operation judgment matrix to calculate the risk score;

[0073] The behavior detection module 640 is used to assess the flight status of the aircraft based on the risk score, and to determine that the pilot has abnormal behavior when the flight status is abnormal; or to determine that the pilot has no abnormal behavior when the flight status is normal.

[0074] Therefore, this embodiment, by combining and calculating real-time flight parameters and target flight parameters, covers the flight states within the range of flight parameters input by the pilot, thus identifying any abnormal pilot maneuvers. Furthermore, this embodiment does not require extensive prior simulation experiments or model training; it only needs to set thresholds and coefficients, and then identify pilot maneuvering anomalies through algebraic calculations. This not only reduces computational load but also minimizes the computational demands and burden on the onboard processor. Moreover, the scope of abnormal maneuvering judgment in this embodiment depends on the manually selected parameter range, and can be applied to abnormal control in a single channel or broadly encompass anomalies in all coupled channels. In addition, this embodiment, through differential, integral, and proportional adjustments, can provide abnormal maneuvering warnings for past, present, and future scenarios, effectively detecting reciprocating maneuvers and preventing control resonance from threatening flight safety.

[0075] According to the above description, in an exemplary embodiment, the process of inputting the target flight parameters and real-time flight parameters into the abnormal maneuver judgment matrix to calculate the risk score in the risk score calculation module 630 may include: obtaining the abnormal maneuver judgment state quantity corresponding to the abnormal maneuver judgment matrix; using the target flight parameters as the horizontal input of the abnormal maneuver judgment matrix and the real-time flight parameters as the vertical input of the abnormal maneuver judgment matrix; combining all horizontal inputs and all vertical inputs in pairs according to the abnormal maneuver judgment state quantity, and calculating the risk score for each combination of abnormal maneuver judgment state quantities; and generating the risk score of the abnormal maneuver judgment matrix based on the risk score of each combination of abnormal maneuver judgment state quantities.

[0076] As an example, the process of generating a risk score for the abnormal manipulation judgment matrix based on the risk score of each abnormal manipulation judgment state variable combination includes: summing the risk scores of all abnormal manipulation judgment state variable combinations in each lateral input to obtain a summary risk score for each lateral input, denoted as the first-level summary risk score; and using the first-level summary risk scores of all lateral inputs as the risk score of the abnormal manipulation judgment matrix. As another example, the process of generating a risk score for the abnormal manipulation judgment matrix based on the risk score of each abnormal manipulation judgment state variable combination may also include: summing the first-level summary risk scores of all lateral inputs to obtain a second-level summary risk score; and using the second-level summary risk score as the risk score of the abnormal manipulation judgment matrix. Therefore, this embodiment can use the first-level summary risk scores of all lateral inputs as the risk score of the abnormal manipulation judgment matrix, or it can sum the first-level summary risk scores of all lateral inputs as the risk score of the abnormal manipulation judgment matrix. Essentially, this embodiment can treat each lateral input as a control channel, and then perform risk scoring for each channel to conduct flight risk assessment. Simultaneously, this embodiment can also sum the first-level summary scores corresponding to all control channels to obtain a second-level summary risk score, which is a level of risk assessment for the entire aircraft control system. The advantage of this method, which divides the system into primary and secondary levels, is that it allows for a more comprehensive risk assessment of both single channels and the entire system.

[0077] According to the above description, in an exemplary embodiment, the process of evaluating the flight status of the aircraft based on risk scores in the behavior detection module 640 further includes: evaluating the current / historical / future flight status of the aircraft based on the first-level aggregated risk score of each lateral input, and / or the first-level aggregated risk scores of all lateral inputs. As an example, in this embodiment, the abnormal manipulation judgment state quantity corresponding to each lateral input can be used as the target abnormal manipulation judgment state quantity, and the integral threshold, differential threshold, and proportional threshold of the target abnormal manipulation judgment state quantity can be obtained. The first-level aggregated risk score of each lateral input is integrally calculated to obtain the corresponding first-level real-time integral value; and the first-level aggregated risk score of each lateral input is differentially calculated to obtain the corresponding first-level real-time differential value; and the first-level aggregated risk score of each lateral input is proportionally calculated to obtain the corresponding first-level real-time proportional value. For each lateral input corresponding to an abnormal control judgment state quantity, the first-level real-time integral value of each lateral input is compared with the integral threshold of the corresponding target abnormal control judgment state quantity to obtain a first integral comparison result; and the first-level real-time differential value of each lateral input is compared with the differential threshold of the corresponding target abnormal control judgment state quantity to obtain a first differential comparison result; and the first-level real-time proportional value of each lateral input is compared with the proportional threshold of the corresponding target abnormal control judgment state quantity to obtain a first proportional comparison result. Based on the first integral comparison result, the first differential comparison result, and the first proportional comparison result, the current / historical / future flight status of the aircraft is assessed.

[0078] Specifically, the process of assessing the flight status of an aircraft based on the first integral comparison result, the first differential comparison result, and the first proportional comparison result includes: determining the target abnormal control judgment state quantity with integral comparison anomalies using the first integral comparison result; determining the target abnormal control judgment state quantity with differential comparison anomalies using the first differential comparison result; and determining the target abnormal control judgment state quantity with proportional comparison anomalies using the first proportional comparison result. The target abnormal control judgment state quantities with integral comparison anomalies, differential comparison anomalies, and proportional comparison anomalies are correlated, and the real-time flight anomaly level of the aircraft is determined based on the correlation result. The real-time flight anomaly level is compared with a flight anomaly level reference table, and the flight status of the aircraft is determined to be abnormal or normal based on the flight anomaly level comparison result.

[0079] Therefore, this embodiment divides risk scoring into Level 1 (single control quantity), which can obtain single state quantities to assess all risk levels in the past, present, and future, facilitating the location of the control channel where specific abnormal manipulation occurs. Simultaneously, through PID control, abnormal manipulation warnings can be issued for the past, present, and future scenarios. Furthermore, the integral term summarizing the Level 1 data also performs reciprocating manipulation detection on each channel to avoid the danger of control resonance.

[0080] According to the above description, in an exemplary embodiment, the process of assessing the flight status of an aircraft based on a risk score may further include: assessing the current / historical / future flight status of the aircraft based on a secondary aggregated risk score. As an example, in this embodiment, the abnormal manipulation judgment state quantity corresponding to each lateral input can be used as the target abnormal manipulation judgment state quantity, and the integral threshold, differential threshold, and proportional threshold of the target abnormal manipulation judgment state quantity can be obtained. The secondary aggregated risk score for each lateral input is integrally calculated to obtain the corresponding secondary real-time integral value; and the secondary aggregated risk score for each lateral input is differentially calculated to obtain the corresponding secondary real-time differential value; and the secondary aggregated risk score for each lateral input is proportionally calculated to obtain the corresponding secondary real-time proportional value. For each lateral input corresponding to an abnormal control judgment state quantity, the second-level real-time integral value of each lateral input is compared with the integral threshold of the corresponding target abnormal control judgment state quantity to obtain a second integral comparison result; and the second-level real-time differential value of each lateral input is compared with the differential threshold of the corresponding target abnormal control judgment state quantity to obtain a second differential comparison result; and the second-level real-time proportional value of each lateral input is compared with the proportional threshold of the corresponding target abnormal control judgment state quantity to obtain a second proportional comparison result. Based on the second integral comparison result, the second differential comparison result, and the second proportional comparison result, the current / historical / future flight status of the aircraft is assessed.

[0081] Specifically, the process of assessing the flight status of an aircraft based on the second integral comparison result, the second differential comparison result, and the second proportional comparison result includes: determining the target abnormal control judgment state quantity with integral comparison anomalies using the second integral comparison result; determining the target abnormal control judgment state quantity with differential comparison anomalies using the second differential comparison result; and determining the target abnormal control judgment state quantity with proportional comparison anomalies using the second proportional comparison result. The target abnormal control judgment state quantities with integral comparison anomalies, differential comparison anomalies, and proportional comparison anomalies are correlated, and the real-time flight anomaly level of the aircraft is determined based on the correlation result. The real-time flight anomaly level is compared with a flight anomaly level reference table, and the flight status of the aircraft is determined to be abnormal or normal based on the flight anomaly level comparison result.

[0082] Therefore, this embodiment uses a two-tiered risk scoring system (overall aircraft control) to assess the overall aircraft status across all risk levels in the past, present, and future, facilitating the identification of specific control channels where abnormal maneuvers occur. Simultaneously, PID control provides early warnings for abnormal maneuvers in the past, present, and future scenarios. Furthermore, the summation of the first-tier aggregated scores for all control channels yields a second-tier aggregated risk score, representing a risk assessment at the overall aircraft control level. The advantage of this hierarchical assessment method is that it allows for a more comprehensive risk assessment of both individual channels and the entire aircraft simultaneously.

[0083] According to the above description, in an exemplary embodiment, the process of inputting the target flight parameters and real-time flight parameters into the abnormal maneuver judgment matrix to calculate the risk score in the risk score calculation module 630 may include: obtaining the abnormal maneuver judgment state quantity corresponding to the abnormal maneuver judgment matrix; using the target flight parameters as the vertical input of the abnormal maneuver judgment matrix and the real-time flight parameters as the horizontal input of the abnormal maneuver judgment matrix; combining all horizontal inputs and all vertical inputs pairwise according to the abnormal maneuver judgment state quantity, and calculating the risk score for each combination of abnormal maneuver judgment state quantities; and generating the risk score of the abnormal maneuver judgment matrix based on the risk score of each combination of abnormal maneuver judgment state quantities. In this embodiment, the risk score of the abnormal maneuver judgment matrix is ​​generated based on the risk score of each combination of abnormal maneuver judgment state quantities. The process of generating the risk score of the abnormal maneuver judgment matrix can be found in some of the above embodiments, and will not be repeated here.

[0084] According to the above description, in an exemplary embodiment, the process of inputting the target flight parameters and real-time flight parameters into the abnormal control judgment matrix to calculate the risk score in the risk score calculation module 630 may further include: obtaining the abnormal control judgment state quantity corresponding to the abnormal control judgment matrix; using the target flight parameters as the lateral and longitudinal inputs of the abnormal control judgment matrix, so that the target flight parameters can also be combined; combining all lateral and longitudinal inputs pairwise according to the abnormal control judgment state quantity, and calculating the risk score for each combination of abnormal control judgment state quantities; and generating the risk score of the abnormal control judgment matrix based on the risk score of each combination of abnormal control judgment state quantities. In this embodiment, by using the target flight parameters as the lateral and longitudinal inputs of the abnormal control judgment matrix, the target flight parameters can be combined pairwise, for example, pitch control and throttle control can be combined, thereby generating the risk score of the abnormal control judgment matrix based on the risk score of each combination of abnormal control judgment state quantities. The process of generating the risk score of the abnormal control judgment matrix can be found in some of the above embodiments, and will not be repeated here.

[0085] In summary, this application provides a pilot behavior detection system. Responding to pilot input, a data processing device outputs target flight parameters based on the pilot's input; then, based on real-time flight parameters, it determines the aircraft's flight mode at the current moment, denoted as the real-time flight mode; it matches an abnormal control judgment matrix according to the real-time flight mode, and inputs the target flight parameters and real-time flight parameters into the abnormal control judgment matrix to calculate a risk score; based on the risk score, it assesses the aircraft's flight status, and if the flight status is abnormal, it determines that the pilot has abnormal behavior; or, if the flight status is normal, it determines that the pilot has no abnormal behavior. Therefore, this system, by combining real-time flight parameters and target flight parameters, covers the flight status within the range described by the pilot's input flight parameters, thus identifying whether the pilot has abnormal control. Furthermore, this system does not require extensive pre-simulation experiments or model training; it only needs to set thresholds and coefficients, and then identify pilot abnormal control through algebraic operations. This not only reduces computational load but also lowers the computational requirements and burden on the onboard processor. Furthermore, the system's judgment range for abnormal control depends on the manually selected parameter range, and can be applied to abnormal control of a single channel or broadly to encompass abnormalities in all coupled channels. In addition, through differential, integral, and proportional adjustments, the system can provide early warnings of abnormal control in past, present, and future scenarios, effectively detecting reciprocating controls and preventing control resonance from threatening flight safety.

[0086] In another exemplary embodiment of this application, an aircraft is also provided; wherein the aircraft can be applied to the pilot behavior detection method or pilot behavior detection system described in any of the above embodiments. As an example, the aircraft in this embodiment can be an airplane. When the aircraft provided in this embodiment is applied to the pilot behavior detection method or pilot behavior detection system described in any of the above embodiments, the corresponding technical principles and functional effects are as described in the above embodiments, and will not be repeated here.

[0087] It should be understood that the above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

[0088] It should be understood that the structures, proportions, sizes, etc., depicted in the accompanying drawings are merely for illustrative purposes to aid those skilled in the art and are not intended to limit the scope of this application. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in proportions, or adjustments to size, without affecting the effectiveness and purpose of this application, should still fall within the scope of the technical content disclosed in this application. Furthermore, the terms such as "upper," "lower," "left," "right," "middle," and "one" used in this specification are merely for clarity and not intended to limit the scope of this application. Changes or adjustments to their relative relationships, without substantially altering the technical content, should also be considered within the scope of this application's implementation.

[0089] It should be understood that although terms such as first, second, third, etc., may be used to describe preset ranges in the embodiments of this application, these preset ranges should not be limited to these terms. These terms are only used to distinguish preset ranges from one another. For example, without departing from the scope of the embodiments of this application, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

Claims

1. A method for detecting pilot behavior, characterized in that, The method includes the following steps: In response to pilot input, the data processing device outputs target flight parameters based on the pilot input. The flight mode of the aircraft at the current moment is determined based on the real-time flight parameters at the current moment, and is denoted as the real-time flight mode. The abnormal manipulation judgment matrix is ​​matched according to the real-time flight mode, and the flight parameters of the manipulation target and the real-time flight parameters are input into the abnormal manipulation judgment matrix to calculate the risk score; The flight status of the aircraft is assessed based on the risk score, and if the flight status is abnormal, it is determined that the pilot has engaged in abnormal behavior; or, if the flight status is normal, it is determined that the pilot has not engaged in abnormal behavior. The process of inputting the target flight parameters and the real-time flight parameters into the abnormal manipulation judgment matrix to calculate the risk score includes: Obtain the abnormal manipulation judgment state quantity corresponding to the abnormal manipulation judgment matrix; The target flight parameters are used as the horizontal input to the abnormal control judgment matrix, and the real-time flight parameters are used as the vertical input to the abnormal control judgment matrix; or, the target flight parameters are used as the vertical input to the abnormal control judgment matrix, and the real-time flight parameters are used as the horizontal input to the abnormal control judgment matrix; or, the target flight parameters are used as both the horizontal and vertical inputs to the abnormal control judgment matrix. All horizontal and vertical inputs are combined in pairs according to the abnormal manipulation judgment state variables, and a risk score is calculated for each combination of abnormal manipulation judgment state variables. The risk score of the abnormal manipulation judgment matrix is ​​generated based on the risk score of each abnormal manipulation judgment state quantity combination.

2. The pilot behavior detection method according to claim 1, characterized in that, The process of generating the risk score of the abnormal manipulation judgment matrix based on the risk score of each combination of abnormal manipulation judgment state variables includes: The risk scores for all combinations of abnormal manipulation judgment state variables in each horizontal input are summed to obtain the summary risk score for each horizontal input, denoted as the first-level summary risk score; and, The first-level summary risk score of all horizontal inputs is used as the risk score of the abnormal manipulation judgment matrix.

3. The pilot behavior detection method according to claim 2, characterized in that, The process of assessing the flight status of the aircraft based on the risk score also includes: The current / historical / future flight status of the aircraft is assessed based on the first-level aggregated risk score for each lateral input and / or the first-level aggregated risk score for all lateral inputs.

4. The pilot behavior detection method according to claim 2, characterized in that, The real-time flight anomaly level of the aircraft is determined based on the primary aggregated risk score; The real-time flight anomaly level is compared with the flight anomaly level reference table, and the flight status of the aircraft is determined to be abnormal or normal based on the comparison result.

5. The pilot behavior detection method according to any one of claims 2 to 4, characterized in that, The process of generating the risk score of the abnormal manipulation judgment matrix based on the risk score of each combination of abnormal manipulation judgment state variables further includes: The first-level aggregated risk scores of all horizontal inputs are summed to obtain the second-level aggregated risk scores; The secondary summary risk score is used as the risk score of the abnormal manipulation judgment matrix.

6. The pilot behavior detection method according to claim 5, characterized in that, The process of assessing the flight status of the aircraft based on the risk score also includes: The current / historical / future flight status of the aircraft is assessed based on the secondary summary risk score.

7. The pilot behavior detection method according to claim 6, characterized in that, The process of assessing the flight status of the aircraft based on the risk score also includes: The real-time flight anomaly level of the aircraft is determined based on the secondary aggregated risk score; The real-time flight anomaly level is compared with the flight anomaly level reference table, and the flight status of the aircraft is determined to be abnormal or normal based on the comparison result.

8. A pilot behavior detection system, characterized in that, The system includes: The flight parameter module is used to respond to pilot control inputs, and the data processing device outputs control target flight parameters based on the pilot control inputs; The flight mode module is used to determine the aircraft's flight mode at the current moment based on the real-time flight parameters at the current moment, which is referred to as the real-time flight mode. A risk score calculation module is used to match an abnormal manipulation judgment matrix according to the real-time flight mode, and input the target flight parameters and the real-time flight parameters into the abnormal manipulation judgment matrix to calculate a risk score. The process of inputting the target flight parameters and the real-time flight parameters into the abnormal manipulation judgment matrix to calculate the risk score includes: obtaining abnormal manipulation judgment state variables corresponding to the abnormal manipulation judgment matrix; using the target flight parameters as the horizontal input of the abnormal manipulation judgment matrix and the real-time flight parameters as the vertical input; or, using the target flight parameters as the vertical input of the abnormal manipulation judgment matrix and the real-time flight parameters as the horizontal input; or, using the target flight parameters as both the horizontal and vertical inputs; combining all horizontal and vertical inputs pairwise according to the abnormal manipulation judgment state variables, and calculating the risk score for each combination of abnormal manipulation judgment state variables; and generating the risk score of the abnormal manipulation judgment matrix based on the risk score of each combination of abnormal manipulation judgment state variables. The behavior detection module is used to assess the flight status of the aircraft based on the risk score, and determine that the pilot has abnormal behavior when the flight status is abnormal; or, determine that the pilot has no abnormal behavior when the flight status is normal.

9. An aircraft, characterized in that, The aircraft is used in the pilot behavior detection method as described in any one of claims 1 to 7.

Images