Trusted guide instruction generation method based on four-dimensional track

Abstract

The invention discloses a credible guiding instruction generation method based on a four-dimensional flight path in an autonomous operation mode of a civil aircraft, and solves the problem of guiding the operation of an aircraft by a pilot in an autonomous operation / flight path-based operation mode in the future. The method comprises the following steps: firstly, fitting a four-dimensional track discrete point set into a binary high-order function by adopting an approximate fitting derivation method, and identifying four-dimensional track characteristics of a track; then, constructing a standard control guide instruction library, performing instruction mapping on the identified four-dimensional track characteristics, and converting four-dimensional track data into a control guide instruction which can be intuitively understood and executed by a pilot; and finally, constructing a pilot-aircraft man-machine interaction model and an aircraft flight control model, and optimizing the sending opportunity and the broadcasting speed of the credible guidance instruction, so that the pilot can control the aircraft to execute the nominal planning four-dimensional track to the maximum extent.

Description

Technical Field

[0001] The invention belongs to the technical field of air traffic management, and in particular relates to a method for generating a credible guidance instruction based on a four-dimensional trajectory. Background Art

[0002] With the rapid growth of air traffic, the traditional manual air traffic management method can no longer meet the needs of efficiency and safety. Therefore, the proposal of the autonomous operation mode of civil aircraft aims to improve the autonomy, intelligence and safety of aircraft through technical means to adapt to the growth of future air traffic. As one of the cores of this mode, the four-dimensional trajectory technology introduces the time dimension based on longitude, latitude and altitude to achieve more accurate prediction and control of the aircraft's flight status. This technology can not only improve the predictability of air traffic, but also better ensure flight safety, reduce fuel use, reduce costs, and reduce carbon dioxide emissions. However, to achieve four-dimensional trajectory guidance in the autonomous operation mode of civil aircraft, a series of technical problems need to be solved. Among them, how to generate reliable flight guidance instructions based on four-dimensional trajectory is one of the keys. Traditional flight guidance instruction generation methods often rely on manual experience and rules, and are difficult to adapt to complex and changing flight environments. Therefore, a method that can automatically and accurately generate guidance instructions based on four-dimensional trajectory information is needed. Based on the above needs, a reliable guidance instruction generation method based on four-dimensional trajectory in the autonomous operation mode of civil aircraft is proposed. This method continuously provides aircraft with reliable guidance instructions for target guidance tracks through four-dimensional track feature recognition, flight guidance instruction generation, and human-machine interactive optimization control. This method not only improves the accuracy and reliability of flight guidance instructions, but also reduces the task load of air traffic controllers, providing strong support for the future development of the civil aviation industry. Summary of the invention

[0003] To achieve the above object, the technical solution of the present invention is as follows:

[0004] The present invention discloses a method for generating credible guidance instructions based on four-dimensional trajectory in the autonomous operation mode of civil aircraft, which solves the problem of pilots guiding aircraft operation in the future autonomous operation / track-based operation mode. It includes the following steps:

[0005] Step S1: approximate fitting and differentiation to identify track features

[0006] Step S2: Track feature recognition based on curve derivation

[0007] Step S3: Human-computer interaction optimization control

[0008] Among them, step S1 approximate fitting and differentiation to identify track features includes two main processes:

[0009] Step S1.1, discrete track point fitting

[0010] Step S1.1 discrete track point fitting mainly includes (1) selecting a fitting function, (2) constructing an objective function, (3) solving fitting parameters, and (4) verifying the fitting effect.

[0011] Step S1.2: Track feature identification based on curve derivation

[0012] Step S1.2 is track feature identification based on curve derivation, which mainly includes (1) first-order derivative calculation, (2) second-order derivative calculation, and (3) track curvature calculation.

[0013] The track feature recognition based on curve derivation in step S2 includes two main processes:

[0014] Step S2.1: Flight guidance instruction formatting code

[0015] The instruction formatting code in step 2.1 is as follows:

[0016] Instruction code = TYPE_CODE + DATE_CODE + TIME_CODE +

[0017] AIRCRAFT_CALLSIGN_CODE

[0018] Among them, TYPE_CODE is the code of the instruction type, such as "T" for take-off instruction, "L" for landing instruction, etc.

[0019] DATE_CODE is the code for the date the instruction was issued, which can be a combination of year, month, and day, such as "20241022".

[0020] TIME_CODE is the code for the instruction release time, which can be a combination of hours, minutes and seconds, such as "143000".

[0021] AIRCRAFT_CALLSIGN_CODE is the aircraft call sign code, which can be an abbreviation of the call sign or some converted form.

[0022] Instruction formatting code example (based on C language):

[0023]

[0024] Step S2.2: Calculation of the transformation from 4D trajectory to command

[0025] Step S2.2: Calculation of the transformation from the four-dimensional trajectory to the command, mainly including (1) speed adjustment command generation, (2) heading adjustment command generation, and (3) altitude adjustment command generation.

[0026] Step S3 of human-machine interaction optimization control includes three main processes:

[0027] Step S3.1: Aircraft flight control model construction

[0028] Step S3.2: Pilot behavior model construction

[0029] Step S3.3: Human-machine interaction control optimization

[0030] Step S3.3 human-machine interactive control optimization mainly includes (1) establishing an aircraft flight state prediction model (predictor) based on the state space, and (2) establishing an aircraft control instruction optimization model (optimizer) based on the current flight state and the predictor.

[0031] Compared with the prior art, the beneficial effects of the present invention are as follows: the method greatly improves the autonomous operation capability and flight safety of civil aviation through highly intelligent technical means, and brings revolutionary changes to air traffic management. First, the application of this method significantly improves the fluency and efficiency of air traffic. Through precise four-dimensional trajectory planning, aircraft can efficiently and accurately complete flight missions within the predetermined time and space, effectively avoid flight conflicts and delays, and improve the operating efficiency of the entire air traffic system. Secondly, this method is of great significance for ensuring flight safety. Trusted guidance instructions based on four-dimensional trajectory can reflect the flight status of aircraft in real time and accurately, provide more reliable information support for pilots and air traffic controllers, help to timely discover and deal with potential flight risks, and ensure flight safety. In addition, this method also helps to reduce aviation operating costs and environmental impacts. By optimizing flight paths and speeds, unnecessary fuel consumption and carbon dioxide emissions are reduced, which is of great significance for promoting the development of green aviation and achieving sustainable development goals. In summary, the method for generating trusted guidance instructions based on four-dimensional trajectory in the autonomous operation mode of civil aircraft not only improves the intelligent level of air traffic management, but also provides strong support for the safe, efficient and green development of the aviation industry, and has extensive social benefits and far-reaching significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is the overall flow chart of the method for generating trusted guidance instructions based on four-dimensional trajectory in the autonomous operation mode of civil aircraft;

[0033] Figure 2 This is the overall framework diagram of the reliable guidance instruction generation method based on four-dimensional trajectory;

[0034] Figure 3 Generate samples for trusted guidance instructions based on 4D trajectory;

[0035] Figure 4 It is the deviation error diagram between the aircraft track and the nominal track executed based on the guidance instructions;

[0036] Figure 5 Error analysis of aircraft trajectory based on guidance instructions. DETAILED DESCRIPTION

[0037] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0038] Embodiment: This method aims to solve the problem of pilots guiding aircraft operation in the future autonomous operation / track-based operation mode, that is, after the airborne / ground track planning system outputs the four-dimensional track planning result, guide the pilot to perform the correct operation to complete the implementation of the planned four-dimensional track, as shown in the attached Figure 1 shown.

[0039] The present invention provides a method for generating a trusted guidance instruction based on a four-dimensional trajectory in an autonomous operation mode of a civil aircraft, comprising the following steps:

[0040] Step S1: Approximate fitting and differentiation to identify track features

[0041] Assume that the discrete point set of the four-dimensional trajectory planning result of the airborne / ground terminal is {(x i ,y i ,z i ,t i )} i＝1 n , where x i ,y i Indicates longitude and latitude (or two-dimensional spatial position), z i Indicates height, t i Indicates the timestamp of the track point.

[0042] Step S1.1: Discrete track point fitting

[0043] (1) Select the fitting function

[0044] Choose a two-variable high-order function as the fitting function, such as a polynomial function:

[0045]

[0046] Among them, m, n is the highest degree of the polynomial, a ij are the fitting parameters to be sought.

[0047] (2) Constructing the objective function

[0048] Use mean square error (MSE) as the objective function:

[0049]

[0050] Where E is the value of the fitting mean square error MSE, and n is the number of four-dimensional track points.

[0051] (3) Solving the fitting parameters

[0052] The objective function For each fitting parameter a ij Find the partial derivative and set it equal to 0 to form a linear equation system:

[0053] for all i,j

[0054] Solve this linear equation system and get the fitting parameter a ij The value of .

[0055] (4) Verify the fitting effect

[0056] Use the validation data set to calculate the prediction error of the fitted function, such as the root mean square error (RMSE):

[0057]

[0058] Among them, n val is the size of the validation dataset (the number of track points), z val,i , x val,i ,y val,i It is to verify the various attributes (longitude, latitude, altitude) of the points in the track dataset.

[0059] Step S1.2: Track feature identification based on curve derivation

[0060] (1) Calculation of first-order derivative

[0061] The first-order derivative represents the slope of the tangent line of a function at a certain point, reflecting the rate at which the function value changes with the independent variable. In track feature identification, the first-order derivative can be used to describe the rate of change of the altitude, longitude or latitude of the track with time or another dimension.

[0062] For example:

[0063] Rate of change of altitude: By calculating the first derivative of altitude with respect to time or longitude / latitude, ascending, descending, or level flight segments in a flight path can be identified.

[0064] Rate of Change of Direction: By calculating the first derivative of longitude or latitude with respect to time, turning points or straight flight segments in the track can be identified.

[0065] The calculation method of the first-order derivative of the fitted track curve is as follows:

[0066]

[0067] Among them, n is the number of coefficients before the higher-order variable, a ij are constant coefficients for each term, and f represents the equation of the curve.

[0068] (2) Calculation of second-order derivatives

[0069] The second-order derivative represents the rate of change of the tangent slope of a function at a certain point, reflecting the acceleration or concavity of the function value change. In track feature recognition, the second-order derivative can be used to describe the curvature change or acceleration change of the track:

[0070] Curvature change: By calculating the second derivative of altitude, longitude or latitude with respect to another dimension, the degree of curvature or complexity in the track can be identified. A second derivative greater than 0 indicates that the function graph is protruding upward (concave) at that point, and less than 0 indicates that it is protruding downward (convex), which helps to identify inflection points or complex shape areas in the track.

[0071] Acceleration changes: If the track data contains time information, the acceleration changes in the track can be identified by calculating the second-order derivative of altitude, longitude or latitude with respect to time.

[0072] The calculation method of the second-order derivative of the fitted track curve is as follows:

[0073]

[0074] Among them, n is the number of coefficients before the higher-order variable, a ij are constant coefficients for each term, and f represents the equation of the curve.

[0075] (3) Track curvature calculation

[0076] Curvature is a quantity that describes the curvature of a curve or surface. It reflects the rate at which the tangent direction of a curve at a certain point changes with position. In track feature recognition, curvature can be used to describe the curvature or complexity of a track:

[0077] Curvature: By calculating the curvature of a track, you can quantify how curvy the track is, which is useful for identifying complex shapes or unusual paths in a track.

[0078] Feature extraction: Points with larger curvature usually correspond to turning points or important feature points in the track. These points can be used as the basis for feature extraction for subsequent track classification, identification or tracking.

[0079] The curvature of the track curve after fitting is calculated as follows:

[0080]

[0081] Where K represents the curvature of the planned trajectory and f represents the equation of the curve.

[0082] Step S2: Generation of flight guidance instructions based on track characteristics

[0083] Step S2.1: Formatting the flight guidance instructions

[0084] According to the actual needs of autonomous operation / track-based operation scenarios, determine what types of guidance instructions need to be included, such as take-off instructions, landing instructions, route change instructions, etc. Design the structure of each instruction, including the basic format of the instruction, the fields included (such as aircraft call sign, time, position, speed, altitude, heading, etc.) and the order of the fields. Assign a unique code to each instruction type to facilitate recognition and processing by the computer system. Design the encoding rules for each field in the instruction to ensure that each field can be accurately parsed. According to the determined instruction type and structure, standardize the sorted guidance instructions to ensure that each instruction complies with the prescribed format and structure, and standardize the key fields in the instructions, such as unified units, precision, etc., and further use the database management system or file storage system to establish a standard control guidance instruction library, and store the standardized guidance instructions in the instruction library according to the encoding rules to ensure that each instruction can be quickly retrieved and called.

[0085] Each guidance instruction can be assigned a unique code, which can be generated based on factors such as instruction type, date, time, aircraft call sign, etc.

[0086] Instruction code = TYPE_CODE + DATE_CODE + TIME_CODE +

[0087] AIRCRAFT_CALLSIGN_CODE

[0088] Among them, TYPE_CODE is the code of the instruction type, such as "T" for take-off instruction, "L" for landing instruction, etc.

[0089] DATE_CODE is the code for the date the instruction was issued, which can be a combination of year, month, and day, such as "20241022".

[0090] TIME_CODE is the code for the instruction release time, which can be a combination of hours, minutes and seconds, such as "143000".

[0091] AIRCRAFT_CALLSIGN_CODE is the aircraft call sign code, which can be an abbreviation of the call sign or some converted form.

[0092] Instruction formatting code example (based on C language):

[0093]

[0094]

[0095] Note: The formulas here are only examples. The actual coding rules should be determined according to specific requirements and system design.

[0096] Step S2.2 Calculation of the transformation from 4D trajectory to command

[0097] The known four-dimensional track characteristics can be represented as a series of discrete points, each of which contains longitude (Lon), latitude (Lat), altitude (Alt) and time (Time) information.

[0098] In order to match the instruction library, we may need to quantify these characteristics into specific numerical indicators, such as average speed, average altitude change rate, average heading change rate, etc. The instruction library contains a series of predefined flight guidance instructions, each of which corresponds to a specific track feature pattern. By comparing the real-time track characteristics with the feature patterns in the instruction library, we can select the most appropriate instruction for execution.

[0099] (1) Speed ​​adjustment command generation

[0100] If you need to adjust the speed, you can use the following formula to calculate the target speed V target :

[0101] V target =V current +△V

[0102] Among them, V current is the current speed, and △V is the speed increment that needs to be adjusted.

[0103] (2) Heading adjustment command generation

[0104] The heading adjustment can be done by calculating the target heading HDG target With the current heading HDG current The difference between:

[0105] △HDG=HDG target -HDG current

[0106] Among them, △HDG is the heading difference that needs to be adjusted.

[0107] (3) Height adjustment command generation

[0108] Height adjustment can be done by calculating the target height Alt target Alt with current height current The difference between , and combined with the climb rate or descent rate to achieve:

[0109]

[0110] Where ΔTime is the time required to reach the target altitude.

[0111] Step S3: Human-computer interaction optimization control

[0112] Step S3.1: Aircraft flight control model construction

[0113] Aircraft flight control models are used to describe how its physical properties such as mass, inertia, thrust, drag and lift affect flight conditions. These conditions usually include position (longitude, latitude, altitude), speed (ground speed, airspeed, wind speed components), acceleration, attitude angle (pitch angle, roll angle, yaw angle) and angular velocity.

[0114] The aircraft flight control model can be expressed by a set of nonlinear differential equations:

[0115] x=f(x,u,w)

[0116] in:

[0117] x is the state vector, which contains all the dynamic states of the aircraft (such as position, velocity, attitude, etc.).

[0118] u is the control input vector, which contains the control instructions (such as thrust, control surface deflection, etc.) provided by the pilot or the autopilot system.

[0119] w is the external interference vector, which includes external environmental factors such as wind, temperature, and air pressure.

[0120] f is a function that describes the dynamic behavior of the aircraft.

[0121] Control logic is the process of how the aircraft determines the next control action based on the current state and control instructions. This can be a simple open-loop control system (such as fixed route flight) or a complex closed-loop control system (such as an autopilot).

[0122] The control logic can be expressed by the following function:

[0123] u=g(x,r)

[0124] in:

[0125] r is a reference vector containing the desired flight state (such as desired heading, altitude, speed, etc.).

[0126] g is the control logic function, which calculates the control input based on the current state and the desired state.

[0127] Sensor feedback is the process of how an aircraft obtains information about its current state. This information usually includes position, velocity, attitude, and acceleration, which are measured by various sensors on the aircraft (such as GPS, radar altimeter, pitot tube, gyroscope, etc.).

[0128] Sensor feedback can be represented by a function:

[0129] γ=h(x,v)where:

[0130] γ is the measurement vector, which contains the state information provided by the sensor.

[0131] v is the measurement noise vector, representing the uncertainty or error in the sensor measurement.

[0132] h is the sensor measurement function, which calculates the measurement value based on the current state and measurement noise.

[0133] Step S3.2: Pilot behavior model construction

[0134] The pilot's behavior can be viewed as an input-output system, in which the input includes the aircraft's status information, external environment information, and the pilot's internal cognitive state (such as goals, intentions, emotions, etc.), and the output is the pilot's control instructions for the aircraft, such as changing heading, speed, altitude, etc.

[0135] The pilot's behavior can be represented by the following function:

[0136]

[0137] Where x, y are the positions of the aircraft, v x ,v y is the velocity component, a x ,a y is the acceleration component, Control_Command is the pilot's control command, Aircraft_Parameters is the physical parameters of the aircraft (such as mass, moment of inertia, etc.), and External_Forces is the external force (such as wind, gravity, etc.).

[0138] The pilot's control command Control_Command can be further expressed as:

[0139] Control_Command=f(Aircraft_State,External_Environment,Pilot_Cognition)

[0140] Among them, Aircraft_State represents the aircraft state, External_Environment represents the external environment, and Pilot_Cognition represents the pilot's reaction / cognitive state.

[0141] (2) Human-computer interaction interface

[0142] The human-machine interface is the medium for information exchange between the pilot and the aircraft. The following functions can be used to represent the flow of information on the interface:

[0143] Displayed_Information=h(Aircraft_State,External_Environment)

[0144] Updated_Pilot_Congnition=k(Displayed_Information,Pilot_Cognition)

[0145] Among them, Displayed_Information is the information displayed to the pilot on the interface, Updated_Pilot_Cognition is the pilot's cognitive state after updating based on the displayed information, Aircraft_State is the aircraft's own state information, and External_Environment is the aircraft's surrounding operating environment information.

[0146] Step S3.3: Human-machine interaction control optimization

[0147] Combining the models in step S3, we can get a complete aircraft control model:

[0148] 1. Use the dynamics model to calculate the next state of the aircraft.

[0149] 2. Use sensor feedback to obtain measurements of the current state.

[0150] 3. Use human-computer interaction and control logic to calculate the next control input.

[0151] 4. Apply control inputs to the dynamics model to update the state of the aircraft.

[0152] This process is usually an iterative process that is repeated in each time step until the aircraft reaches the desired flight state or encounters some termination condition (such as the end of the flight, a failure, etc.).

[0153] (1) Establishing an aircraft flight state prediction model (predictor) based on state space

[0154] Based on the kinematic model, the aircraft state prediction equation based on the state space is established, such as the CV (constant velocity) model, CA (constant acceleration) model, CTRV (constant turn rate and velocity) model, etc. The predictor uses different models to calculate the next state according to the aircraft action, expressed as s(t+δt)=p(t,u), where s represents the state of the aircraft, s(t+δt) is the state of the aircraft predicted at the future time t+δt based on the current time t, p(t,u) represents the predictor, t represents the current time, and u represents the aircraft action.

[0155] (2) Establish an aircraft control command optimization model (optimizer) based on the current flight status and predictor

[0156] The lateral trajectory error (CTE) is used to measure the difference between the current position of the aircraft and the target track. CET = d·sinθ, where d represents the distance between the current position of the aircraft and the nearest path point in front of the aircraft, and θ represents the angular difference between the aircraft heading and the nearest path point.

[0157] The error function e(t) between the current and guided trajectory of the aircraft is established, and the objective function of the optimization model is established to make the aircraft position at the next k moments as close to the expected trajectory as possible. The aircraft position information at the future moment is predicted by the predictor based on the aircraft position information at the previous moment by taking action u(t+(k-1)δt).

[0158] Subsequently, the predictor and optimizer are combined into an aircraft trusted guidance instruction generator based on MPC (model predictive control), and the current flight state of the aircraft and the planned four-dimensional guidance trajectory are obtained. The current flight state of the aircraft provided by the airborne equipment, including longitude and latitude, horizontal velocity, vertical velocity, heading angle, angular velocity and other information, is obtained as the input state of the optimizer, and the four-dimensional guidance trajectory result obtained by aircraft trajectory planning and negotiation is obtained as the input trajectory of the optimizer. The input state and input trajectory are input into the aircraft trusted guidance instruction generator, and the optimization result is calculated. The current aircraft flight state and target trajectory are used as parameters of the optimization model, and the optimal control scheme of the aircraft from the current moment to the future k-1 moment is solved based on the optimization model. The control variable u(t) at the current moment in the optimal control scheme is output as the final result, and the corresponding text instruction is obtained through mapping as the aircraft trusted guidance instruction.

[0159] It should be noted that the above content only illustrates the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. For ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications all fall within the protection scope of the claims of the present invention.

Claims

1. A method for generating a trusted guidance instruction based on a four-dimensional trajectory, characterized in that: The method comprises the following steps, Step S1, approximate fitting and differentiation to identify track features; Wherein step S1 comprises: step S1.1, discrete track point fitting, step S1.2, track feature identification based on curve derivation; Step S2, track feature identification based on curve derivation; Wherein step S2 comprises: step S2.1, formatting and encoding the flight guidance instruction, step S2.2, calculating the transformation from the four-dimensional trajectory to the instruction; Step S3, human-computer interaction optimization control; Step S3 includes: step S3.1, aircraft flight control model construction, step S3.2, pilot behavior model construction, step S3.3, human-computer interaction control optimization.

2. The method for generating a trusted guidance instruction based on a four-dimensional trajectory according to claim 1, characterized in that: The step S1.1 discrete track point fitting includes the following parts: Step S1.11, selecting a binary high-order function as a fitting function; Among them, m, n is the highest degree of the polynomial, a ij is the fitting parameter to be sought; Step S1.12, using mean square error as the objective function; Where E is the value of the fitting mean square error MSE, and n is the number of four-dimensional track points; Step S1.13, solving the fitting parameters; The objective function For each fitting parameter a ij Find the partial derivative and set it equal to 0 to form a linear equation system: for all i,j Solve this linear equation system and get the fitting parameter a ij The value of Step S1.14, verify the fitting effect; Calculate the prediction error of the fitted function using the validation dataset; Among them, n val is the size of the validation dataset, z val,i , x val,i ,y val,i It is to verify the various attributes of the points in the track dataset.

3. The method for generating a trusted guidance instruction based on a four-dimensional trajectory according to claim 1, characterized in that: The track feature identification based on curve derivation in step S1.2 includes the following parts: Step S1.21, first-order derivative calculation, the first-order derivative calculation method of the fitted track curve is as follows: Among them, n is the number of coefficients before the higher-order variable, a ij are constant coefficients for each term, and f represents the equation of the curve; Step S1.22, second-order derivative calculation, the second-order derivative of the fitted track curve is calculated as follows: Among them, n is the number of coefficients before the higher-order variable, a ij are constant coefficients for each term, and f represents the equation of the curve; Step S1.23, track curvature calculation, the curvature of the track curve after fitting is calculated as follows: Where K represents the curvature of the planned trajectory and f represents the equation of the curve.

4. The method for generating a trusted guidance instruction based on a four-dimensional trajectory according to claim 1, characterized in that: The step S2.1 determines what types of guidance instructions need to be included according to the actual needs of the autonomous operation / track-based operation scenarios, designs the structure of each instruction, including the basic format of the instruction, the fields included, and the order of the fields, assigns a unique code to each instruction type, designs the encoding rules for each field in the instruction, ensures that each field can be accurately parsed, and standardizes the sorted guidance instructions according to the determined instruction type and structure to ensure that each instruction complies with the prescribed format and structure, and normalizes the key fields in the instructions, further uses a database management system or a file storage system to establish a standard control guidance instruction library, and stores the standardized guidance instructions in the instruction library according to the encoding rules to ensure that each instruction can be quickly retrieved and called.

5. The method for generating a trusted guidance instruction based on a four-dimensional trajectory according to claim 1, characterized in that: The step S2.2, the calculation of the transformation from the four-dimensional trajectory to the instruction includes: Step S2.21, speed adjustment instruction generation; If you need to adjust the speed, you can use the following formula to calculate the target speed V target : V target =V current +△V Among them, V current is the current speed, △V is the speed increment that needs to be adjusted; Step S2.22, generating a heading adjustment instruction; The heading adjustment can be done by calculating the target heading HDG target With the current heading HDG current The difference between: △HDG=HDG target -HDG current Among them, △HDG is the heading difference that needs to be adjusted; Step S2.3, height adjustment instruction generation; Height adjustment can be done by calculating the target height Alt target Alt with current height current The difference between , and combined with the climb rate or descent rate to achieve: Where ΔTime is the time required to reach the target altitude.

6. The method for generating a trusted guidance instruction based on a four-dimensional trajectory according to claim 1, characterized in that: The step S3.1 comprises the following steps, Aircraft flight control model construction: The aircraft flight control model is represented by a set of nonlinear differential equations: x=f(x,u,w) Where: x is the state vector, which contains all the dynamic states of the aircraft, u is the control input vector, which contains the control instructions provided by the pilot or the autopilot system, w is the external disturbance vector, which contains external environmental factors such as wind, temperature, and air pressure, and f is a function that describes the dynamic behavior of the aircraft; The control logic is represented by the following function: u=g(x,r) Where: r is the reference vector, containing the desired flight state, g is the control logic function, which calculates the control input based on the current state and the desired state; Sensor feedback is represented by a function: γ=h(x,v) Where: γ is the measurement vector, which contains the state information provided by the sensor, v is the measurement noise vector, which represents the uncertainty or error in the sensor measurement, and h is the sensor measurement function, which calculates the measurement value based on the current state and measurement noise.

7. The method for generating a trusted guidance instruction based on a four-dimensional trajectory according to claim 1, characterized in that: The step S3.2 comprises the following steps, The pilot's behavior is represented by the following function: Where x, y are the positions of the aircraft, v x ,v y is the velocity component, a x ,a y is the acceleration component, Control_Command is the pilot's control command, Aircraft_Parameters is the physical parameters of the aircraft, and External_Forces is the external force.

8. The method for generating a trusted guidance instruction based on a four-dimensional trajectory according to claim 1, characterized in that: The step S3.3 comprises the following steps, Establish an aircraft flight state prediction model based on state space; Based on the kinematic model, the aircraft state prediction equation based on the state space is established. Different models are used inside the predictor to calculate the next state according to the aircraft action, which is expressed as s(t+δt)=p(t,u), where s represents the state of the aircraft, s(t+δt) is the state of the aircraft predicted at the future time t+δt based on the current time t, p(t,u) represents the predictor, t represents the current time, and u represents the aircraft action; Establish an aircraft control command optimization model based on current flight status and predictors; The lateral trajectory error is used to measure the difference between the current position of the aircraft and the target track. CET = d·sinθ, where d represents the distance between the current position of the aircraft and the nearest path point in front of the aircraft, and θ represents the angular difference between the aircraft heading and the nearest path point.

9. The method for generating a trusted guidance instruction based on a four-dimensional trajectory according to claim 1, characterized in that: The step S3.3 comprises the following steps, The error function e(t) between the current and guided trajectory of the aircraft is established, and the objective function of the optimization model is established to make the aircraft position at the next k moments as close to the expected trajectory as possible. The aircraft position information at the future moment is predicted by the predictor based on the aircraft position information at the previous moment by taking action u(t+(k-1)δt).

10. The method for generating a trusted guidance instruction based on a four-dimensional trajectory according to claim 1, characterized in that: The step 3.3 also includes forming an aircraft trusted guidance instruction generator based on MPC with a predictor and an optimizer, obtaining the current flight state of the aircraft and a planned four-dimensional guidance trajectory, obtaining the current flight state of the aircraft provided by the onboard equipment as the input state of the optimizer, obtaining the four-dimensional guidance trajectory result obtained by aircraft trajectory planning and negotiation as the input trajectory of the optimizer, inputting the input state and the input trajectory into the aircraft trusted guidance instruction generator, calculating the optimization result, using the current aircraft flight state and the target trajectory as parameters of the optimization model, solving the optimal control plan of the aircraft from the current moment to the future k-1 moments based on the optimization model, outputting the control variable u(t) at the current moment in the optimal control plan as the final result, and obtaining the corresponding text instruction through mapping as the aircraft trusted guidance instruction.

Images