A method and system for coordinated control of attitude and tilt mechanisms of an aircraft

Abstract

This invention relates to the field of flight control technology and discloses a method and system for coordinated control of an aircraft's attitude and tilt mechanisms. The method includes the following steps: acquiring real-time generalized state data and aerodynamic environment data containing real-time dynamic pressure parameters; establishing an open-loop dynamic model containing an aerodynamic pseudo-potential energy field and an aerodynamic natural dissipation matrix regulated by dynamic pressure parameters; establishing a desired closed-loop dynamic structure containing a desired interconnect matrix and a desired damping matrix based on target equilibrium state data; calculating an aerodynamic inertial two-port mapping matrix based on dynamic pressure parameters and a dynamic damping reconstruction matrix based on the current control command and physical saturation boundary values, used to update the desired closed-loop dynamic structure; solving the partial differential algebra matching equations by simultaneously solving the aforementioned model and structure, and extracting the nonlinear coordinated control law output to the actuator. This invention enables a smooth transition of control rights across modes of the aircraft and improves the system robustness when the actuator approaches its limits.

Description

Technical Field

[0001] This invention relates to the field of flight control technology, specifically to a method and system for coordinated control of the attitude and tilt mechanisms of an aircraft. Background Technology

[0002] During the transition between vertical takeoff and landing and high-speed cruise, aircraft equipped with tilting mechanisms exhibit highly nonlinear and strongly interconnected dynamic characteristics in their dynamic properties due to drastic changes in external airflow and dynamic pressure. Traditional control schemes typically rely on small-amplitude linear approximations of the dynamic model at specific operating points. This approach severs the intrinsic connection between rigid body inertia and the complex external aerodynamic environment in terms of global physical energy flow. Consequently, it becomes difficult for the aircraft to achieve accurate energy matching in the modal transition range, easily leading to abrupt changes in the system's dynamic characteristics and attitude oscillations.

[0003] As the flight envelope continues to expand, the aerodynamic and inertial coupling between the attitude control channel and the tilt motion channel becomes increasingly severe. Existing independent loop control architectures or static allocation strategies cannot adaptively and smoothly adjust the interconnected topology between multiple physical channels according to changes in external dynamic pressure, resulting in extremely abrupt transitions in control power during the transition between the rotor-dominant mode and the fixed-wing-dominant mode. Even more serious is that when the aircraft performs attitude control under complex airflow disturbances, aerodynamic control surfaces and other actuators are prone to approaching their absolute physical actuation limits and entering a saturation state. In traditional control law calculation architectures, once the core actuators lose their adjustment capabilities, without a cross-channel physical energy compensation mechanism based on constrained boundaries, the system not only fails to force actuators in other dimensions to intervene and share the control requirements, but also suffers from feedback signal distortion due to deadlock in the algebraic calculation loop. Ultimately, this causes the aircraft to completely lose its overall robustness and flight safety when facing extreme conditions or limited maneuvers. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for coordinated control of the attitude and tilt mechanism of an aircraft. It solves the problems of control abrupt changes caused by strong aerodynamic and inertial coupling during cross-mode transitions in aircraft equipped with tilt mechanisms, as well as the loss of cross-channel compensation capability and global robustness of the system when the actuator approaches the physical saturation boundary.

[0005] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of the present invention provides a method for coordinated control of the attitude and tilt mechanism of an aircraft, comprising the following steps: Acquire real-time generalized state data of the aircraft and real-time aerodynamic environment data including real-time dynamic pressure parameters; Based on real-time generalized state data and real-time aerodynamic environment data, an open-loop dynamic model is established, which includes an aerodynamic pseudo-potential energy field and an aerodynamic natural dissipation matrix that are regulated by real-time dynamic pressure parameters. Based on the received target equilibrium state data, a desired closed-loop dynamic structure including the desired interconnection matrix and the desired damping matrix is ​​established; The aerodynamic inertia dual-port mapping matrix is ​​calculated based on real-time dynamic pressure parameters, and the dynamic damping reconstruction matrix is ​​calculated based on the actual control input commands of the actuators and the physical saturation boundary values, which is used to update the desired closed-loop dynamic structure. By solving the partial differential algebraic matching equations using the simultaneous open-loop dynamics model and the updated desired closed-loop dynamics structure, the nonlinear cooperative control law is extracted, and the corresponding control commands are output to the aircraft's actuators.

[0006] In the specific implementation principle of the first aspect, the steps of establishing the open-loop dynamic model further include: extracting the static aerodynamic moment coefficients from the pre-established aerodynamic database; scaling the static aerodynamic moment coefficients using real-time dynamic pressure parameters to obtain the absolute aerodynamic restoring moment; integrating the absolute aerodynamic restoring moment with the attitude angle vector to generate an aerodynamic pseudo-potential energy field; extracting the dynamic aerodynamic damping derivative; multiplying the dynamic aerodynamic damping derivative with the real-time dynamic pressure parameters; and allocating the calculation result to the diagonal elements of the aerodynamic natural dissipation matrix. Through the above process, the aerodynamic characteristics of the aircraft are embedded into the basic dynamic architecture.

[0007] In the specific implementation principle of the first aspect, the steps of setting the target equilibrium state and establishing the desired closed-loop dynamic structure further include: extracting the global symmetric positive definite inertia matrix from the open-loop dynamic model, setting the off-diagonal block elements representing the strong cross-linking between the tilt rotor space and the attitude subspace in the global symmetric positive definite inertia matrix to zero, and generating a diagonally dominant desired closed-loop inertia matrix. Based on real-time generalized state data, a current generalized coordinate vector is constructed, and a desired generalized coordinate vector is constructed based on the target equilibrium state; a quadratic mathematical model of the error between the current generalized coordinate vector and the desired generalized coordinate vector is set as the desired potential energy function, and it is verified that the first-order partial derivative of the desired potential energy function with respect to the generalized coordinates is equal to zero and the Hessian matrix formed by the second-order partial derivatives is a positive definite matrix; subsequently, the desired interconnection matrix is ​​defined as an antisymmetric matrix, and the desired damping matrix is ​​defined as a symmetric positive semi-definite matrix, thereby ensuring the global asymptotic stability of the closed-loop system.

[0008] In the specific implementation principle of the first aspect, the steps for generating the aero-inertial dual-port mapping matrix are as follows: Constructing a cross-axis interconnection operator with non-zero antisymmetric elements configured in the off-diagonal block region, and a self-axis interconnection operator with non-zero elements configured in the diagonal block region; calculating and generating normalized weights between 0 and 1, and then weighting and summing the cross-axis interconnection operator data and the self-axis interconnection operator data based on these normalized weights. The normalized weights are calculated as follows: when the real-time dynamic pressure parameter is less than or equal to a set first aerodynamic pressure threshold, a normalized weight approaching zero is output; when the real-time dynamic pressure parameter is greater than or equal to a set second aerodynamic pressure threshold, a normalized weight approaching 1 is output; when the parameter is between the two thresholds, a preset piecewise cubic polynomial interpolation algorithm is used to calculate the output, thereby achieving a smooth transition.

[0009] In the specific implementation principle of the first aspect, the steps for calculating and generating the dynamic damping reconstruction matrix are as follows: Calculate the reference positive definite expected damping matrix matching the current aerodynamic environment based on real-time dynamic pressure parameters; read the actual control input command from the previous control cycle as the feedback actual control input command for the actuators, and extract the absolute physical actuation extreme values ​​of each actuator as physical saturation boundary values; calculate the absolute margin value between the actual control input command and the physical saturation boundary values; use the sum of the absolute margin value and a preset strictly positive minimum constant as the denominator, and a preset positive penalty gain coefficient as the numerator to calculate the penalty damping element; construct the boundary penalty expected damping matrix using the penalty damping element, and add it to the reference positive definite expected damping matrix to generate the dynamic damping reconstruction matrix. During this process, a cross-mapping operation is performed. Since the aircraft's actuators include attitude actuators and tilting mechanisms, the penalty damping elements calculated based on the absolute margin values ​​of the attitude actuators are mapped and superimposed onto the diagonal block region corresponding to the tilting mechanism's degrees of freedom in the dynamic damping reconstruction matrix.

[0010] In the specific implementation principle of the first aspect, the steps of solving the matching equation further include: extracting the input configuration matrix from the open-loop dynamic model and calculating its weighted pseudo-inverse matrix; constructing a partial differential algebraic matching equation by multiplying both sides of the equation on the left by the weighted pseudo-inverse matrix, and analytically extracting the nonlinear cooperative control law vector containing the mixed commands of body torque and tilting drive torque. During the solution process, the calculation result of the previous digital control cycle is latched by an internal first-order hysteresis filter buffer, and the calculation result is delayed by one clock cycle and then inserted into the expression of the current operation cycle to block the physically unsolvable state of the algebraic loop in the damping reconstruction calculation loop.

[0011] A second aspect of the present invention provides a coordinated control system for the attitude and tilt mechanism of an aircraft, applied to the coordinated control method provided in the first aspect, comprising: The state perception module is used to acquire real-time generalized state data and real-time aerodynamic environment data of the aircraft. The real-time generalized state data includes the airframe attitude angle, airframe attitude angular velocity, tilt mechanism deflection angle and tilt mechanism deflection angular velocity. The real-time aerodynamic environment data includes real-time dynamic pressure parameters. The global modeling module is used to establish an open-loop dynamic model containing an aerodynamic pseudo-potential energy field and an aerodynamic natural dissipation matrix based on real-time generalized state data and real-time aerodynamic environment data, wherein the aerodynamic pseudo-potential energy field is scheduled by real-time dynamic pressure parameters. The topology configuration module is used to receive the set target equilibrium state data, construct the corresponding desired closed-loop Hamiltonian energy function, and establish the desired closed-loop dynamic structure containing the desired interconnect matrix and the desired damping matrix. The matrix mapping and reconstruction module is used to calculate and generate the aerodynamic inertia two-port mapping matrix based on the input real-time dynamic pressure parameters, and to calculate and generate the dynamic damping reconstruction matrix based on the feedback of the actual control input command of the actuator and the preset physical saturation boundary value, which is used to update the desired closed-loop dynamic structure. The analysis and control allocation module is used to solve the partial differential algebraic matching equations by simultaneously solving the open-loop dynamics model and the updated desired closed-loop dynamics structure to extract the nonlinear cooperative control law, and output the corresponding control commands to the actuators of the aircraft according to the set input configuration matrix.

[0012] This invention provides a method and system for coordinated control of the attitude and tilt mechanisms of an aircraft. It offers the following advantages: 1. This invention utilizes real-time dynamic pressure parameters to calculate the static aerodynamic moment coefficient and dynamic aerodynamic damping derivative, establishing an open-loop dynamic model that includes an aerodynamic pseudo-potential energy field and an aerodynamic natural dissipation matrix. This successfully transforms traditional discrete aerodynamic data into a global energy function analysis framework. This mechanism effectively integrates rigid body inertial mechanics with the external aerodynamic environment, enabling the control base to accurately match the actual characteristics of the aircraft under different operating conditions.

[0013] 2. This invention calculates normalized weights by setting dynamic pressure threshold conditions related to flight state, and then performs weighted summation on the cross-axis interconnection operator and the self-axis interconnection operator based on these weights to generate an aero-inertial two-port mapping matrix. This feature can smoothly adjust the interconnection topology between the attitude channel and the tilt channel according to the continuous change of external dynamic pressure, effectively eliminating system oscillations caused by sudden changes in dynamic characteristics during the transition between the rotor dominant mode and the fixed wing dominant mode.

[0014] 3. This invention generates dynamic penalty damping elements by extracting the absolute margin between actual control commands and physical actuation extreme values, and performs cross-mapping operations to write the constrained state of the attitude actuator into the closed-loop equation of the tilt mechanism in the form of virtual damping. When the attitude control surface faces the risk of saturation, this mechanism can spontaneously force the tilt mechanism to intervene and share the control requirements. At the same time, it works in conjunction with a first-order hysteresis filter buffer to block unsolvable states in the algebraic loop, ensuring the safe and stable operation of the system under extreme boundary conditions. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a schematic diagram of the system modules of the present invention; Figure 3 This is a comparative schematic diagram of the elevator deflection angle response of the present invention; Figure 4 This is a schematic diagram comparing the pitch angle tracking response of the aircraft according to the present invention. Detailed Implementation

[0016] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] See attached document Figure 1 The present invention provides a method for coordinated control of the attitude and tilt mechanism of an aircraft, which may include the following steps: The system acquires real-time generalized state data and real-time aerodynamic environment data of the aircraft. The real-time generalized state data includes the aircraft attitude angle, the aircraft attitude angular velocity, the tilt mechanism deflection angle, and the tilt mechanism deflection angular velocity. The real-time aerodynamic environment data includes real-time dynamic pressure parameters.

[0018] An open-loop dynamic model based on port Hamiltonian theory is constructed. In this open-loop dynamic model, an aerodynamic pseudo-potential energy field and an aerodynamic natural dissipation matrix are set according to real-time dynamic pressure parameters to describe the physical coupling relationship between body attitude, tilting mechanism motion and external flow field.

[0019] Define the target equilibrium state and the desired closed-loop Hamiltonian energy function, and establish the desired closed-loop dynamic structure including the desired interconnect matrix and the desired damping matrix.

[0020] Based on real-time dynamic pressure parameters and physical boundary constraint parameters, the desired interconnection matrix and desired damping matrix are configured to generate an aerodynamic inertial two-port mapping matrix and a dynamic damping reconstruction matrix, thus establishing an energy transfer path between the body attitude subspace and the tilting mechanism subspace.

[0021] By combining the open-loop dynamics model with the desired closed-loop dynamics structure, a nonlinear cooperative control law is generated, and the corresponding control commands are output to the actuators of the aircraft based on the nonlinear cooperative control law.

[0022] See attached document Figure 2 The present invention provides a coordinated control system for the attitude and tilt mechanism of an aircraft, which may include the following modules: The state awareness module is used to collect real-time generalized state data through airborne sensors and to acquire real-time aerodynamic environment data through an atmospheric data system.

[0023] The global modeling module communicates with the state perception module and is used to establish an open-loop dynamic model that includes an aerodynamic pseudo-potential energy field and an aerodynamic natural dissipation matrix based on the received real-time generalized state data and real-time aerodynamic environment data.

[0024] The topology configuration module communicates with the global modeling module to receive the set target equilibrium state data, construct the corresponding desired closed-loop Hamiltonian energy function, and establish the desired closed-loop dynamic structure containing the desired interconnect matrix and the desired damping matrix.

[0025] The matrix mapping and reconstruction module communicates with the topology configuration module. It is used to calculate and generate the aerodynamic inertia two-port mapping matrix based on the input real-time dynamic pressure parameters, and to calculate and generate the dynamic damping reconstruction matrix based on the feedback current control command value of the actuator and the preset physical saturation boundary value. The calculation results are fed back to the topology configuration module to update the desired closed-loop dynamic structure.

[0026] The analysis and control allocation module communicates with the matrix mapping and reconstruction module and the global modeling module, respectively. It is used to solve the partial differential algebraic matching equations by simultaneously establishing the open-loop dynamics model and the updated desired closed-loop dynamics structure, and to extract the nonlinear cooperative control law. Based on the set input configuration matrix, this module converts the control solution results into attitude control surface deflection commands, thrust vector commands, and tilt servo drive commands, and sends them to the corresponding execution physical components.

[0027] In existing tilt-plane control schemes, the attitude control loop and tilt control loop employ a decoupled control architecture. The gyroscopic torque and inertial coupling torque generated by the tilt mechanism's deflection motion on the airframe are defined as external disturbances to the attitude control system. The control system calculates and outputs a counter-torque command to counteract these torques based on a pre-established torque calculation model or disturbance observer. During the continuous tilt transition flight phase, the counter-torque command is superimposed on the basic attitude adjustment command, causing the control deflection amount allocated to the aerodynamic control surfaces or rotor actuators to reach the mechanical limits of the physical mechanism, resulting in the actuators entering the saturation region.

[0028] The cooperative control method and system provided in this embodiment are based on passive control theory, which combines the aircraft's main shaft and tilt mechanism into a unified port Hamiltonian system. The aircraft's attitude angular velocity and the tilt mechanism's angular velocity are mapped as energy interaction ports in this dynamic model.

[0029] In the open-loop dynamics model, the aerodynamic restoring torque generated by the external flow field on the aircraft body is converted into conservative potential energy within the system, i.e., an aerodynamic pseudo-potential energy field controlled by real-time dynamic pressure parameters. During the construction of the closed-loop control loop, the system does not generate reverse commands to directly counteract the tilting torque, but instead reconstructs the physical interconnect topology within the system through a configuration matrix.

[0030] The system calculates the specific elements of the aero-inertial two-port mapping matrix based on the real-time dynamic pressure parameters. When the tilting mechanism's deflection motion generates inertial coupled kinetic energy, the matrix establishes a non-diagonal transmission channel between the tilting mechanism subspace and the body attitude subspace. This channel imports the original inertial coupled energy into the body attitude port, relying on the system's original aerodynamic damping for energy dissipation and attitude maintenance.

[0031] Simultaneously, the dynamic damping reconstruction matrix monitors the difference between the control command value and the physical saturation boundary value of each attitude actuator. When the difference is less than a set threshold, the system increases the damping injection parameter of the corresponding channel of the tilting mechanism, changing the damping dissipation rate of the tilting actuator. Based on the energy gradient transfer rules, the above process adjusts the distribution ratio of inertial energy and aerodynamic potential energy within the system without adding additional actuator configurations, thereby controlling the actuation amplitude of the actuators.

[0032] See attached document Figure 1 When acquiring real-time generalized state data of the aircraft, the system first establishes a fixed Earth inertial coordinate system and a body coordinate system to describe the aircraft's motion. The system acquires the roll angle, pitch angle, and yaw angle of the body in the inertial coordinate system through the onboard inertial measurement unit, and acquires the deflection angle of the tilting component relative to the body coordinate system through the angle encoder configured on the tilting mechanism servo component.

[0033] Based on the acquired angle data, the system constructs a global generalized coordinate vector. The global generalized coordinate vector is decomposed into body attitude sub-vectors. With tilt mechanism sub-vector : ; in, , For roll angle, The pitch angle, Yaw angle; This represents the number of independent degrees of freedom in the tilting mechanism. , For the first The deflection angle of each tilting component. This indicates transpose. The total degrees of freedom of the system are calculated as follows: .

[0034] The system acquires the corresponding angular velocities of the airframe attitude and tilting components through an airborne rate gyroscope and servo speed sensors, and combines them into a generalized velocity vector. Subsequently, the system constructs a generalized momentum vector. : ; in, Let be the global symmetric positive definite inertia matrix of the system.

[0035] To fully describe the physical interaction between the body and the tilting components during motion, a globally symmetric positive definite inertia matrix is ​​used. A block matrix structure is used for representation. This matrix includes not only the body rotation inertia quantum matrix and the tilting mechanism's own inertia quantum matrix located on the diagonal, but also the cross-coupled inertia quantum matrix located off-diagonally. The elements in the cross-coupled inertia quantum matrix are related to the tilting component's deflection angle. The trigonometric functions are used to quantify in real time the displacement of the assembly's center of mass and the principal axis rotation angle of the moment of inertia tensor caused by tilting motion. Generalized momentum vector. The output is sent to the subsequent processing module to calculate the total kinetic energy of the open-loop system.

[0036] When acquiring real-time aerodynamic environment data, the system collects external flow field parameters through the airborne atmospheric data system. The system uses a Pitot tube probe located on the outside of the fuselage to acquire total air pressure data, uses a static pressure orifice to acquire atmospheric static pressure data, and uses a total temperature sensor to acquire total ambient temperature data.

[0037] The atmospheric data computer receives the measurements from the aforementioned sensors and calculates the real-time air density at the aircraft's altitude based on the gas state equation. The aircraft's current relative airspeed was calculated using Bernoulli's equation. The system calculates real-time dynamic pressure parameters based on the solution results. : ; During the tilt transition phase, the aircraft is affected by the rotor downwash and boundary layer separation on the wing surface, resulting in fluctuations in the measured air pressure data. To avoid directly introducing transient pressure pulses caused by airflow into the control feedback loop, the system calculates preliminary real-time dynamic pressure parameters. The signal is then input into a digital low-pass filter for processing. The system truncates pressure fluctuation signals above a preset cutoff frequency and outputs smoothed and filtered dynamic pressure parameters. These smoothed and filtered dynamic pressure parameters serve as a characterization variable of the external environmental state and are transmitted to the subsequent topology configuration module to trigger the weight update of the elements within the aero-inertial two-port mapping matrix.

[0038] Obtaining the global generalized coordinate vector Generalized momentum vector and real-time dynamic pressure parameters after smoothing and filtering Subsequently, the system constructs an open-loop Hamiltonian energy function to describe the total energy state of the spacecraft. The energy function consists of the sum of the system's total kinetic energy, gravitational potential energy, and a predetermined aerodynamic pseudo-potential energy, and its specific expression is as follows: ; First item This represents the total kinetic energy of the system, including the body and tilting mechanism, during motion. (Second item) The system's gravitational potential energy is determined by pre-stored body mass data, mass data of each tilting component, and a global generalized coordinate vector. The displacement geometry is used to calculate the height of each mass center relative to the reference horizontal plane under the current state, and then the total gravitational potential energy is obtained.

[0039] Third item This refers to the set aerodynamic pseudo-potential energy. Existing passive control models only include mechanical energy; this invention internalizes the external flow field forces into the conservative field energy of the Hamiltonian system through this component. In specific implementation, the system extracts static aerodynamic moment coefficients from a pre-established aerodynamic database, including the longitudinal pitch static stability coefficient and the lateral yaw static stability coefficient. The system utilizes real-time dynamic pressure parameters... The spacecraft reference area and reference length are scaled using the aforementioned dimensionless coefficients to obtain the absolute aerodynamic restoring moment. Subsequently, the system modulates this absolute aerodynamic restoring moment with respect to the attitude angle vector. Perform integration to generate a pseudo-aerodynamic potential energy value corresponding to the current attitude deviation. The gradient of this pseudo-potential field is equal to the restoring torque exerted on the body by the external flow field, establishing a mapping relationship between aerodynamics and the port Hamiltonian energy topology.

[0040] After constructing the energy function, the system establishes the open-loop port Hamiltonian dynamic equations with dissipation terms: ; In this open-loop dynamic equation, the state variables are derived from the coordinate derivatives. and momentum derivative composition. and Represent the open-loop Hamiltonian energy function pairs respectively. and The vector of partial derivatives. and Both are zero matrices. for An identity matrix of order 1.

[0041] In the first term on the right-hand side of the matrix equation, the off-diagonal block constructs a lossless energy exchange structure between coordinates and momentum. The lower right block... The aerodynamic natural dissipation matrix is ​​used to characterize the energy decay caused by flow viscosity. The system extracts dynamic aerodynamic damping derivatives from an aerodynamic database, including roll damping derivatives, pitch damping derivatives, and yaw damping derivatives. The system then modulates these derivatives with real-time dynamic pressure parameters. Multiply and distribute the result to the aerodynamic natural dissipation matrix. The non-zero elements of this matrix are concentrated in the diagonal elements. Since aerodynamic damping mainly acts on the aircraft surface, the non-zero elements of this matrix are concentrated in the sub-vectors of the aircraft attitude. The corresponding rows and columns, and the tilt mechanism sub-vectors The corresponding area is set to zero or minimum damping value.

[0042] The second term on the right-hand side of the matrix equation represents the change in the system's momentum state caused by the external control input. The control input vector includes aerodynamic control surface deflection commands, rotor speed commands, and tilt servo motor drive torque commands issued by the system. To independently control the number of inputs. The input configuration matrix is ​​established based on the aircraft's mechanical hinge structure and actuator geometry. The system... The local mechanical torques generated by each actuator are transformed into system-level torque inputs in a generalized coordinate system. The output of this open-loop Hamiltonian dynamic equation will serve as the basic physical constraint model for subsequent system solutions to the desired control law.

[0043] The system receives flight trajectory commands from an external mission computer or manually operated equipment. From these commands, the system extracts the desired roll angle, desired pitch angle, and desired yaw angle, and from the flight state envelope planner, extracts the desired tilt angle corresponding to the current flight phase. The system combines these desired angles to form a desired generalized coordinate vector. Simultaneously, the system sets the desired generalized momentum vector at steady state to a zero vector containing multiple zero elements. The desired generalized coordinate vector and the zero vector together constitute the target's equilibrium state.

[0044] After obtaining the target equilibrium state, the system constructs a desired closed-loop Hamiltonian energy function to guide state convergence. This energy function consists of a desired kinetic energy term and a desired potential energy term to establish the total energy distribution rule of the system under closed-loop control.

[0045] For the desired kinetic energy term, the system configures a desired closed-loop inertia matrix. The system extracts the global symmetric positive definite inertia matrix from the open-loop dynamic model. To eliminate the nonlinear inertial disturbance caused by the rapid motion of the tilting mechanism on the body attitude in the open-loop state, the system restructures this global inertia matrix. Specifically, the system zeros or reduces the order of the off-diagonal block elements representing the strong cross-linking between the tilting rotor space and the attitude subspace in the original matrix to generate a diagonally dominant desired closed-loop inertia matrix. The system uses the configured desired closed-loop inertia matrix and the generalized momentum vector to calculate the desired kinetic energy of the system in the closed-loop state.

[0046] For the desired potential energy term, the system sets it as a quadratic mathematical model of the error between the current generalized coordinate vector and the desired generalized coordinate vector. To ensure the global asymptotic stability of the control system, the system performs a minimum condition verification. The system calculates the first partial derivative of the desired potential energy function with respect to the generalized coordinates (i.e., the potential energy gradient) and verifies that this gradient vector is equal to zero at the desired generalized coordinates. Subsequently, the system calculates the Hessian matrix formed by the second partial derivatives of the desired potential energy function with respect to the generalized coordinates and verifies that this Hessian matrix is ​​a positive definite matrix at the desired generalized coordinates. The desired potential energy function that satisfies the above mathematical verification conditions is added to the aforementioned desired kinetic energy term, so that the generated desired closed-loop Hamiltonian energy function has a globally isolated minimum characteristic at the target equilibrium point.

[0047] After setting the desired energy function, the system establishes the desired closed-loop dynamic structure framework. The system sets the generalized coordinate derivative and generalized momentum derivative in the closed-loop state to be generated by applying specific matrix transformations to the gradient vector of the desired energy function.

[0048] In this matrix transformation framework, the system initializes two structured matrix templates: the desired interconnection matrix and the desired damping matrix. The system defines the desired interconnection matrix as an antisymmetric matrix, used to define the physical transfer topology between different state variables within the system without energy loss. The system defines the desired damping matrix as a symmetric positive semi-definite matrix, used to define the energy decay rate when the system deviates from the target equilibrium state. In this step, the desired interconnection matrix and the desired damping matrix are established as a mathematical structure framework containing unassigned variables, and this framework is output to the matrix mapping and reconstruction module for subsequent calculation and filling of element values ​​based on real-time environmental parameters and actuator states.

[0049] See attached document Figure 1 The system receives the expected interconnection matrix framework containing unassigned variables from the topology configuration module, and the real-time dynamic pressure parameters processed by the state awareness module after smoothing filtering. The system divides the desired interconnect matrix into diagonal and off-diagonal blocks to construct an aero-inertial two-port mapping matrix. The system uses the following mathematical structure to assign values ​​to the specific elements of the mapping matrix: ; The system first constructs a cross-axis interconnection operator. The system configures non-zero antisymmetric elements in the off-diagonal block region of the operator, specifically at the intersection of the matrix rows corresponding to the body attitude sub-vector and the matrix columns corresponding to the tilt mechanism sub-vector, and at the intersection of the matrix rows corresponding to the tilt mechanism sub-vector and the matrix columns corresponding to the body attitude sub-vector. This distribution of non-zero elements establishes a lossless energy transfer channel between the inertial motion port of the tilt mechanism and the body attitude control port.

[0050] Subsequently, the system constructs a self-axis interconnection operator. The system restricts the non-zero elements of this operator to the diagonal block region corresponding to the tilt mechanism sub-vector, and sets all intersecting block region elements between the body attitude sub-vector and the tilt mechanism sub-vector to zero. The topology of this operator is used to map the transient inertial energy generated by the tilting motion back into the physical degree of freedom computation domain of the tilt mechanism itself.

[0051] The system sets the dynamic pressure normalized scheduling function. The weights of the two operators mentioned above are used to calculate the weights. The system extracts the aircraft aerodynamic data table pre-stored in the memory, calibrates the first aerodynamic pressure threshold required to maintain the stability of the basic airframe, and the second aerodynamic pressure threshold to enable the aerodynamic control surfaces to reach the linear control limit performance. The system then uses the currently acquired real-time dynamic pressure parameters... Input the scheduling function for comparison and calculation: when When the pressure is less than or equal to the first aerodynamic pressure threshold, the system calculates the output weight. ;when When the value is greater than or equal to the second aerodynamic pressure threshold, the system calculates the output weight. ;when When the value is between the first and second aerodynamic pressure thresholds, the system calculates and outputs a specific value between 0 and 1 based on a preset piecewise cubic polynomial interpolation algorithm.

[0052] Through the above function mapping, the system achieves dynamic updating of the specific values ​​of the aero-inertial two-port mapping matrix.

[0053] When the aircraft is hovering or flying at extremely low speeds, the measured real-time dynamic pressure parameters When it is in the low range, the system output It approaches 0. At this point... The internal structure is close to that of the self-axis interconnection operator. The system uses this matrix structure to block the off-diagonal channels through which nonlinear coupling energy is transmitted from the tilting mechanism to the aircraft's attitude. The gyroscopic torque and Coriolis torque induced by the tilting operation are transformed into local negative feedback variables that limit the tilting mechanism's own motion in the closed-loop dynamic equation calculation. This matrix configuration physically constrains the tilting angular acceleration of the tilting servo components under low dynamic pressure conditions, limiting the amplitude of its output command and preventing aircraft attitude instability caused by insufficient aerodynamic efficiency at high angular rate tilts.

[0054] As the aircraft's airspeed increases, the measured real-time dynamic pressure parameters When the pressure increases and approaches the second aerodynamic pressure threshold, the system output is It approaches 1. At this point... The internal structure is close to that of the cross-axis interconnection operator. The system opens an energy interconnection channel between the tilt rotor space and the attitude subspace through the non-zero elements of the off-diagonal matrix structure. Since the aerodynamic damping parameters provided by the external flow field are high at this time, the system introduces the transient energy generated by the rapid deflection of the tilt mechanism into the aircraft attitude port, allowing this energy to directly enter the aerodynamic pseudo-potential energy field set in the open-loop model. The energy introduced into the aircraft attitude port is dissipated through the deflection action of the aerodynamic control surfaces and aerodynamic damping. This operating state allows the tilt mechanism to smoothly transition energy using the external aerodynamic environment without relying on an independent counteracting torque when receiving high angular rate tilt commands.

[0055] The system receives the desired damping matrix framework output by the topology configuration module. To introduce physical constraints on the actuators in the underlying energy topology, the system constructs a dynamic damping reconstruction matrix with physical boundary mapping. The construction process of this matrix includes a benchmark damping setting stage and a boundary penalty damping superposition stage.

[0056] In the reference damping setting stage, the system uses the acquired real-time dynamic pressure parameters. Calculate the baseline positive definite desired damping matrix that matches the current aerodynamic environment. Based on the pre-determined aerodynamic dissipation characteristics that vary with dynamic pressure, the system arranges the non-zero elements of the benchmark positive definite expected damping matrix on the main diagonal blocks corresponding to the body attitude, which is used to characterize the basic energy dissipation rate of the system before it approaches mechanical saturation.

[0057] In the boundary penalty damping superposition stage, the system reads the actual control input vector from the previous control cycle through the feedback loop. This vector contains the instruction values ​​sent by the system to each independent execution unit, and its components are represented as follows: The system extracts the pre-calibrated physical saturation boundary vector of the actuator from the hardware configuration database. Its boundary vector Corresponding to the The absolute physical actuation extreme values ​​of an actuator, such as the maximum mechanical deflection angle of an aerodynamic control surface or the highest steady-state speed of a rotor motor.

[0058] For each execution channel, the system calculates the absolute margin value between the current actual control input command and the physical saturation boundary, i.e. Subsequently, the system constructs the boundary penalty expected damping matrix based on this absolute margin value. This matrix is ​​set as a diagonal matrix, and the first element on its main diagonal is... element The calculation model is as follows: ; in, In order to target the Each execution component has a preset positive penalty gain coefficient, which is used to adjust the gradient of damping growth; It is a preset strictly positive minimum constant used to prevent division errors where the denominator is zero when the control command value is equal to the physical saturation boundary value.

[0059] After calculating the boundary penalty expected damping matrix, the system adds it to the reference positive definite expected damping matrix to generate the complete dynamic damping reconstruction matrix. .

[0060] When filling in the elements of the desired closed-loop dynamic structure, the system performs a specific cross-mapping operation. The system maps and superimposes the penalized damping elements, calculated based on the physical margin of the attitude actuators (such as elevators or ailerons), onto the diagonal block region in the dynamic damping reconstruction matrix corresponding to the degrees of freedom of the tilting mechanism.

[0061] Through the above cross-mapping configuration, when the control command of the aircraft attitude channel... Approaching its physical extreme value When the denominator term in the formula approaches the preset constant, This leads to the penalty damping element in the calculated output. The damping increases non-linearly. This increased damping value directly affects the region of mathematical equations containing the derivative of the generalized momentum of the tilting mechanism. Due to the increase in the damping parameter, the system limits the rate of change of the generalized momentum corresponding to the tilting channel during the subsequent solution of the partial differential algebraic matching equations. This calculation process directly outputs tilting servo drive commands with reduced amplitude, reducing the physical deflection angular velocity of the tilting components, thereby reducing the inertial coupling kinetic energy input from the tilting motion to the body attitude channel, and preventing the attitude actuator from entering the saturation runaway region due to continuous receipt of over-limit commands. The system uses the actual control input vector of the previous control cycle for damping reconstruction, avoiding an algebraic loop unsolvable state between the calculation of the desired damping matrix and the solution of the control command in the current cycle.

[0062] After obtaining the open-loop dynamic model, the desired closed-loop dynamic structure, the aero-inertial two-port mapping matrix, and the dynamic damping reconstruction matrix, the system executes an analytical solution program for the nonlinear cooperative control law.

[0063] The system extracts the generalized momentum derivative equation from the open-loop dynamic model and the desired generalized momentum derivative equation from the desired closed-loop dynamic structure. The system then combines the open-loop generalized momentum derivative expression (including external control input) with the closed-loop generalized momentum derivative expression (including the aerodynamic inertia two-port mapping matrix and the dynamic damping reconstruction matrix) into a simultaneous equation to construct a partial differential algebraic matching equation: ; In this matching equation, the left side of the equals sign is the input configuration matrix. With nonlinear cooperative control law vector The product of the terms. The first term on the right side of the equation. The first term is the partial derivative of the open-loop Hamiltonian energy function with respect to the generalized coordinates; the second term is the desired closed-loop inertia matrix. Inverse of the open-loop inertia matrix and the partial derivative of the expected Hamiltonian energy function The product; the third term is a matrix containing the real-time dynamic pressure and physical boundary mapping features. with vector The product of.

[0064] Because the aircraft system includes multiple aerodynamic control surfaces, rotors, and tilt motors, the number of independent actuators is greater than the system's degrees of freedom. Therefore, the input configuration matrix... The matrix is ​​not square. The system calls a weighted pseudo-inverse algorithm to solve the equation. The system extracts the actuator performance weight matrix from the storage module. The elements of this diagonal matrix correspond to the currently available performance coefficients of each physical execution unit. The system uses the formula... The weighted pseudo-inverse matrix of the input configuration matrix is ​​calculated. .

[0065] The system simultaneously left-multiplies both sides of the partial differential algebra matching equation by a weighted pseudo-inverse matrix. The nonlinear cooperative control law vector is extracted analytically. : ; To eliminate the dynamic damping reconstruction matrix To address the algebraic loop problem arising from the introduction of the control input's own variables during the calculation process, the system incorporates a first-order hysteresis filter buffer within the arithmetic unit. The system then processes the calculation results from the previous digital control cycle. The input vector is latched into a buffer and substituted as a known parameter into the right-hand side of the equation for the current computation cycle. The target control input vector for the current cycle is then output through a one-way computation process. .

[0066] In obtaining the target control input vector Subsequently, the system transmits this data to the control distribution module via its internal data bus. The control distribution module, based on a pre-defined proportional coefficient mapping table, assigns the target control input vector, which includes a mixture of body torque and tilt drive torque commands, to the target control input vector. It is parsed as an independent physical channel instruction.

[0067] The system converts the parsed commands into formats, generating deflection angle signals to drive aerodynamic control surfaces, pulse width modulation signals to adjust the output torque of the rotor motors, and target displacement or torque signals to drive the servo motors of the tilt mechanism. The system synchronously transmits these signals to each physical actuator of the aircraft via the onboard aviation data bus. Each physical actuator generates corresponding mechanical displacement or torque outputs based on the received signals, changing the aerodynamic shape of the aircraft and adjusting the direction of the power vector. External sensors continuously collect new state data and flow field data generated after the actuators move, feeding this data back to the state perception module to initiate the next cycle of system closed-loop calculation.

[0068] This embodiment also includes a system device for executing the aforementioned cooperative control method, which is composed of multiple hardware circuit units and embedded software modules.

[0069] The state awareness module includes an inertial measurement unit (IMU), an angle encoder array, an atmospheric data sensing unit, and a signal preprocessing circuit. The IMU is fixedly mounted at the aircraft's fuselage structural reference center and outputs three-axis attitude angles and angular velocities. The angle encoder array is mounted at the mechanical rotary bearings of each tilting mechanism and outputs the physical deflection angle of the tilting component relative to the aircraft's coordinate system. The atmospheric data sensing unit includes a pitot tube probe and a static pressure port external to the aircraft. These components are connected to the signal preprocessing circuit via an onboard aviation data bus. The signal preprocessing circuit internally includes an analog-to-digital converter and a digital low-pass filter. It converts the analog air pressure signal output from the atmospheric data sensing unit into digital dynamic pressure parameters and performs high-frequency pressure pulse truncation filtering to output smooth digital environmental parameters.

[0070] The global modeling and topology configuration module is connected to the data output of the state awareness module. This module consists of a non-volatile memory and a first processing unit. The non-volatile memory is pre-programmed with an organic mass distribution parameter table, a matrix of geometric installation positions of the actuators, and an aerodynamic parameter lookup table. Based on the received attitude data and mass distribution parameters, the first processing unit calculates the mechanical kinetic energy and gravitational potential energy of the system in its current state. Simultaneously, based on the digital dynamic pressure parameters, the first processing unit retrieves the aerodynamic restoring torque coefficient from the aerodynamic parameter lookup table, performs spatial numerical integration, and calculates and generates aerodynamic pseudo-potential energy values. This module includes an associated register for caching the completed open-loop dynamic model matrix and the desired closed-loop structure template containing unknown variables.

[0071] The matrix mapping and reconstruction module is the core data processing node of the system, enabling bidirectional data read / write between the system's high-speed bus and the global modeling and topology configuration module. This module is internally divided into a scheduling function calculation logic unit and a boundary damping evaluation logic unit. The scheduling function calculation logic unit reads digital dynamic pressure parameters, performs polynomial interpolation operations using a hardware multiplier, and calculates normalized weights between 0 and 1. Based on these weights, this logic unit performs a weighted summation of the cross-axis interconnection operator data and the self-axis interconnection operator data to generate an aero-inertial two-port mapping matrix, which is then written to a specified memory address.

[0072] The boundary damping evaluation logic unit includes a division circuit and a comparator. This logic unit extracts the absolute physical extreme value data of each actuator from the hardware configuration database and reads the actual control input commands. The comparator performs a subtraction operation to obtain the absolute margin value, and the division circuit then calculates the penalty damping element with the absolute margin value as the denominator. This logic unit performs a cross-mapping write operation, writing the penalty damping element calculated based on the body attitude channel margin into the storage block corresponding to the tilt mechanism subspace in the dynamic damping reconstruction matrix.

[0073] The analysis and control allocation module connects to the output ports of the matrix mapping and reconstruction module and the global modeling and topology configuration module, directly driving the aircraft's physical actuators. This module includes a partial differential algebra solver, a first-order hysteresis filter buffer, and a signal modulation conversion interface. The partial differential algebra solver includes a matrix multiplication and addition arithmetic logic unit and a matrix inversion hardware accelerator. This solver extracts the numerical matrices generated by each module, calls a weighted pseudo-inverse algorithm to perform parallel calculations of the multidimensional matrix equations, and outputs a nonlinear cooperative control command vector containing numerical information.

[0074] When triggered by the system clock signal, the first-order hysteresis filter buffer latches the nonlinear cooperative control command vector of the current control cycle and feeds it back to the boundary damping evaluation logic unit after a delay of one clock cycle. This is used to block the algebraic loop physical unsolvable state in the aforementioned damping reconstruction calculation loop.

[0075] The signal modulation and conversion interface receives the digital command vector output by the solver and, according to the actuator allocation matrix rules, splits it into independent channel commands. The pulse width modulation circuit inside the interface converts the rotor speed digital command into a PWM square wave signal with a specific duty cycle and sends it to the rotor electronic speed controller; the digital-to-analog converter converts the angle digital command into an analog drive voltage signal, which is sent to the aerodynamic control surface servo actuator and the tilt mechanism servo motor along independent electrical cables to complete the closed-loop physical drive.

[0076] This embodiment illustrates the closed-loop collaborative operation mechanism between the configuration matrices and actuators within the system by describing the specific physical response process during the low dynamic pressure hovering start-up phase and the high dynamic pressure envelope boundary phase.

[0077] When the aircraft is in the initial phase of transitioning from hovering to low-speed forward flight, the real-time dynamic pressure parameters collected by the atmospheric data sensing components are lower than the first aerodynamic pressure threshold set by the system. Under these physical conditions, the aerodynamic restoring torque generated by the flow field acting on the aerodynamic control surfaces is insufficient to maintain the basic attitude balance of the main axis of the aircraft. When the control system receives an external trajectory command to drive the tilt mechanism to deflect, the scheduling function calculation logic unit in the matrix mapping and reconstruction module performs interpolation calculations based on the aforementioned low dynamic pressure parameters, outputting normalized weights that approach zero. Based on these weights, the aero-inertial two-port mapping matrix generated by the system approximates the self-axis interconnection operator in its internal structure.

[0078] In the subsequent closed-loop dynamics solution process, the self-axis interconnection operator sets all cross-axis off-diagonal block elements in the matrix to zero. This numerical operation blocks the path of the gyroscopic torque and Coriolis torque generated by the tilt mechanism during deflection to the body attitude motion equation. The state parameters corresponding to the aforementioned inertial coupling torque are restricted to the diagonal block operation domain associated with the tilt mechanism subspace and participate in the iterative calculation as local negative feedback when solving the partial differential algebra matching equation. The result calculated by the solver based on this equation causes a reduction in the magnitude of the target displacement or torque command output from the analytical and control allocation module to the tilt servo motor. This operating mechanism limits the physical deflection angular velocity of the tilt component from the bottom layer of the control algorithm, preventing the aircraft from becoming unstable due to the execution of large angular rate tilt commands under conditions of insufficient aerodynamic damping.

[0079] During the forward acceleration transition phase of the aircraft, the real-time dynamic pressure parameters continuously increase and exceed the set second aerodynamic pressure threshold. The scheduling function's computational logic unit outputs normalized weights approaching 1, and the internal structure of the aero-inertial two-port mapping matrix is ​​updated to a cross-axis interconnection operator. The system utilizes the non-zero elements within the cross-axis off-diagonal blocks to guide the inertial coupling state parameters generated by the tilt motion to the airframe attitude subspace, where they are smoothly dissipated by the high aerodynamic damping parameters set in the open-loop model.

[0080] During this high dynamic pressure flight phase, if the aircraft receives superimposed large attitude maneuver commands, the deflection command values ​​received by specific actuators (such as elevators or ailerons) in the airframe attitude channel will increase significantly. The boundary damping evaluation logic unit reads the actual control surface input command from the previous digital control cycle through a hysteresis filter buffer and compares it with the absolute physical actuation extreme value of that specific actuator stored in the hardware configuration database. When the actual control surface input command approaches the absolute physical actuation extreme value, the absolute margin between the two decreases. The division circuit within the boundary damping evaluation logic unit uses this decreased absolute margin value as the denominator to calculate and output a non-linearly increasing penalty damping value.

[0081] The system then executes a cross-mapping procedure, superimposing the penalty damping value, which surged due to the attitude channel command approaching saturation, onto the diagonal block of the dynamic damping reconstruction matrix associated with the tilt mechanism subspace. During the partial differential algebra solver calculations in the current control cycle, the energy dissipation coefficient in the tilt motion equation region increases due to the injection of this damping value. Based on the updated damping reconstruction matrix, the solver outputs a tilt servo drive command with a decreased amplitude. Upon receiving this command, the tilt servo mechanism reduces its mechanical rotation rate.

[0082] The reduction in the mechanical rotation rate of the tilting components simultaneously decreases the inertial disturbance parameters transmitted to the body attitude subspace via the cross-axis interconnection operator. The analysis and control allocation module performs a new round of control calculations based on the attitude subspace equations after the disturbance reduction, resulting in a decrease in the deflection command values ​​output to the aforementioned specific actuators. This closed-loop process, through the numerical reconstruction and cross-mapping mechanism of the system's internal configuration matrix, allows the aerodynamic control surface commands approaching mechanical extremities to revert to the linear actuation range, achieving coordinated operation of the tilting servo mechanism and the body attitude actuators, both located on the same physical body, under physical boundary constraints.

[0083] Specific application examples: This embodiment uses a 250 kg-class tiltrotor aircraft (eVTOL) as the verification platform. The aircraft has a tiltrotor mechanism mounted on each wingtip that can deflect around a lateral axis. Elevators for adjusting the aircraft's pitch attitude are located on the wingtips.

[0084] The test scenario was set as a pitch maneuver command encountered during the high dynamic pressure transmodal acceleration transition. In this scenario, the aircraft was in the acceleration phase of transitioning from helicopter mode to fixed-wing mode (current tilt angle 45 degrees), with high external air velocity and real-time dynamic pressure parameters... The second aerodynamic pressure threshold has been exceeded. At this point, the flight control system receives a large step pitch angle climb command.

[0085] Upon receiving the large step pitch command, the analysis and control allocation module calculates the elevator deflection command. Rapidly increasing, approaching the mechanical deflection limit of the elevator (setting the physical saturation boundary value) ).

[0086] According to the present invention, the boundary damping evaluation logic unit within the matrix mapping and reconstruction module reads the actual control surface commands from the previous control cycle in real time and calculates the absolute margin. When this margin shrinks rapidly, the system invokes the formula: ; The penalty damping element for nonlinear surge was calculated. .

[0087] Subsequently, the system performs a critical cross-mapping operation: the damping element triggered by the elevator (attitude channel) margin is... Write to the dynamic damping reconstruction matrix The middle corresponds to the diagonal block of the tilting mechanism subspace.

[0088] Meanwhile, due to the high dynamic pressure state, the expression for the aerodynamic inertial two-port mapping matrix is: ; at this time Approaching 1, the system behaves as a cross-axis interconnection operator. Dominantly, there is a strong energy interconnection between the body posture and the tilting mechanism.

[0089] When finally solving the partial differential algebra matching equations simultaneously: ; Because of the right side of the equation The significant increase in the damping term of the tilt channel automatically reduces the amplitude of the tilt servo drive command analyzed by the solver. This slows down the yaw rate of the tilt mechanism, reducing the gyroscopic and Coriolis torques transmitted to the aircraft's pitch channel. The reduction in external inertial disturbances results in a decrease in the final yaw command allocated to the elevator. The temperature dropped back to the safe linear range, preventing the mechanism from getting stuck at the saturation boundary.

[0090] To visually demonstrate the effects of the aforementioned dynamic collaborative working mechanism, the collaborative control method provided by this invention is compared and tested with the traditional decoupling control method that uses independent loop allocation.

[0091] See attached document Figure 3 , attached Figure 3 The horizontal axis represents simulation time, and the vertical axis represents elevator deflection angle. (Appendix) Figure 3 The horizontal dashed line represents the absolute physical actuation limit of the elevator (25 degrees).

[0092] When the aircraft receives the high-maneuver command at the 2nd second, the traditional control method (dashed curve) lacks a cross-channel energy compensation mechanism, causing the elevator command to rapidly hit the 25-degree physical limit and enter a deep saturation state. However, using the cooperative control method of this invention (solid curve), when the control surface deflection angle approaches 23 degrees, the dynamic damping reconstruction mechanism is triggered. The tilt mechanism actively decelerates to share the attitude control pressure, allowing the elevator deflection angle to smoothly transition near the saturation boundary and stabilize within the physical limit in the subsequent time, maintaining the linear adjustment margin of the control surface.

[0093] See attached document Figure 4 , attached Figure 4 The horizontal axis represents simulation time, and the vertical axis represents the aircraft's pitch angle. (Appendix) Figure 4 The midpoint line represents the target desired pitch angle command.

[0094] Under the traditional decoupling control method (dashed curve), the elevator becomes saturated and loses its ability to resist the strong inertial coupling torque continuously applied to the tilt mechanism. This leads to severe runaway overshoot and oscillation of the aircraft pitch angle during the 3-6 second period, seriously endangering flight safety. In contrast, the cooperative control method of this invention (solid curve) achieves dynamic energy redistribution under physical constraints through the underlying partial differential algebraic equations. As a result, the aircraft's actual pitch angle can smoothly and accurately track target commands, exhibiting extremely strong robustness and resistance to saturation interference globally.

Claims

1. A method for coordinated control of the attitude and tilt mechanism of an aircraft, characterized in that, Includes the following steps: Acquire real-time generalized state data of the aircraft and real-time aerodynamic environment data including real-time dynamic pressure parameters; Based on the real-time generalized state data and the real-time aerodynamic environment data, an open-loop dynamic model is established, which includes an aerodynamic pseudo-potential energy field and an aerodynamic natural dissipation matrix that are scheduled by the real-time dynamic pressure parameters. Based on the received target equilibrium state data, a desired closed-loop dynamic structure including the desired interconnection matrix and the desired damping matrix is ​​established; A two-port aerodynamic inertia mapping matrix is ​​calculated based on the real-time dynamic pressure parameters, and a dynamic damping reconstruction matrix is ​​calculated based on the current control command value of the actuator and the physical saturation boundary value, which is used to update the desired closed-loop dynamic structure. By combining the open-loop dynamics model and the updated desired closed-loop dynamics structure, the partial differential algebraic matching equations are solved to extract the nonlinear cooperative control law, and the corresponding control commands are output to the actuators of the aircraft.

2. The method for coordinated control of attitude and tilt mechanism of an aircraft according to claim 1, characterized in that, The steps for establishing an open-loop dynamic model that includes an aerodynamic pseudo-potential energy field and an aerodynamic natural dissipation matrix include: Extract the static aerodynamic moment coefficients from the pre-established aerodynamic database, and scale the static aerodynamic moment coefficients using the real-time dynamic pressure parameters to obtain the absolute aerodynamic restoring torque; The aerodynamic pseudo-potential energy field is generated by integrating the absolute aerodynamic restoring torque with respect to the attitude angle vector. Extract the dynamic aerodynamic damping derivative, multiply the dynamic aerodynamic damping derivative by the real-time dynamic pressure parameter, and distribute the calculation result to the diagonal elements of the aerodynamic natural dissipation matrix.

3. The method for coordinated control of attitude and tilt mechanism of an aircraft according to claim 1, characterized in that, The steps of setting the target equilibrium state and the desired closed-loop Hamiltonian energy function, and establishing the desired closed-loop dynamic structure including the desired interconnect matrix and the desired damping matrix, include: Extract the global symmetric positive definite inertia matrix from the open-loop dynamic model, and set the off-diagonal block elements in the global symmetric positive definite inertia matrix that represent the strong cross-linking between the tilt rotor space and the attitude subspace to zero to generate the desired diagonally dominant closed-loop inertia matrix. The current generalized coordinate vector is constructed based on the real-time generalized state data, and the desired generalized coordinate vector is constructed based on the target equilibrium state. The quadratic mathematical model of the error between the current generalized coordinate vector and the desired generalized coordinate vector is set as the desired potential energy function, and it is verified that the first-order partial derivative of the desired potential energy function with respect to the generalized coordinates is equal to zero, and the Hessian matrix formed by the second-order partial derivatives of the desired potential energy function with respect to the generalized coordinates is a positive definite matrix. The desired interconnect matrix is ​​defined as an antisymmetric matrix, and the desired damping matrix is ​​defined as a symmetric positive semi-definite matrix.

4. The method for coordinated control of attitude and tilt mechanism of an aircraft according to claim 1, characterized in that, The step of calculating and generating the aerodynamic inertial two-port mapping matrix based on the input real-time dynamic pressure parameters includes: Construct a cross-axis interconnection operator, and configure non-zero antisymmetric elements in the off-diagonal block region of the cross-axis interconnection operator; Construct a self-axis interconnection operator, in which non-zero elements are configured in the diagonal block region corresponding to the tilt mechanism sub-vector, and the elements in the cross block region are configured to zero values; A normalized weight between 0 and 1 is calculated and generated. Based on the normalized weight, the cross-axis interconnection operator data and the self-axis interconnection operator data are weighted and summed to generate the aero-inertial dual-port mapping matrix.

5. The method for coordinated control of attitude and tilt mechanism of an aircraft according to claim 4, characterized in that, The step of calculating and generating normalized weights between 0 and 1 includes: When the real-time dynamic pressure parameter is less than or equal to the set first pneumatic dynamic pressure threshold, the normalized weight value that approaches zero is calculated and output. When the real-time dynamic pressure parameter is greater than or equal to the set second aerodynamic pressure threshold, the normalized weight value that approaches 1 is calculated and output. When the real-time dynamic pressure parameter is between the first pneumatic dynamic pressure threshold and the second pneumatic dynamic pressure threshold, the normalized weight is calculated and output using a preset piecewise cubic polynomial interpolation algorithm.

6. The method for coordinated control of attitude and tilt mechanism of an aircraft according to claim 1, characterized in that, The step of calculating and generating the dynamic damping reconstruction matrix based on the actual control input command of the actuator and the preset physical saturation boundary value includes: Calculate the reference positive definite desired damping matrix that matches the current aerodynamic environment based on the real-time dynamic pressure parameters; The actual control input command of the previous control cycle is read as the actual control input command of the feedback actuator, and the absolute physical actuation extreme value of each actuator is extracted as the physical saturation boundary value. Calculate the absolute margin between the actual control input command and the physical saturation boundary value; The sum of the absolute margin value and the preset strict positive minimum constant is used as the denominator, and the preset positive penalty gain coefficient is used as the numerator to calculate the penalty damping element. The penalty damping element is then used to construct the boundary penalty expected damping matrix in diagonal matrix form. The dynamic damping reconstruction matrix is ​​generated by adding the reference positive definite expected damping matrix to the boundary penalty expected damping matrix.

7. The method for coordinated control of attitude and tilt mechanism of an aircraft according to claim 6, characterized in that, The steps for performing the cross-mapping operation during the generation of the dynamic damping reconstruction matrix include: The aircraft's actuators include attitude actuators and tilting mechanisms; The penalty damping element, calculated based on the absolute margin value of the attitude actuator, is mapped and superimposed onto the diagonal block region in the dynamic damping reconstruction matrix corresponding to the motion degree of freedom of the tilting mechanism.

8. The method for coordinated control of attitude and tilt mechanism of an aircraft according to claim 1, characterized in that, The steps for solving the partial differential algebraic matching equations and extracting the nonlinear cooperative control law by simultaneously establishing the open-loop dynamics model and the updated desired closed-loop dynamics structure include: Extract the input configuration matrix from the open-loop dynamics model and calculate the weighted pseudo-inverse matrix of the input configuration matrix; Construct the partial differential algebra matching equation, and multiply both sides of the partial differential algebra matching equation by the weighted pseudo-inverse matrix on the left. The nonlinear cooperative control law vector containing the mixed commands of body torque and tilt drive torque is extracted by analysis.

9. The method for coordinated control of attitude and tilt mechanism of an aircraft according to claim 8, characterized in that, When solving the partial differential algebra matching equation, the calculation result of the previous digital control cycle is latched by an internal first-order hysteresis filter buffer, and the calculation result is delayed by one clock cycle and then inserted into the expression of the current operation cycle. This is used to block the physical unsolvable state of the algebraic loop in the damped reconstruction calculation loop.

10. A coordinated control system for the attitude and tilt mechanism of an aircraft, characterized in that, A method for coordinated control of attitude and tilt mechanisms of an aircraft according to any one of claims 1-9 includes: The state perception module is used to acquire real-time generalized state data and real-time aerodynamic environment data of the aircraft. The real-time generalized state data includes the airframe attitude angle, airframe attitude angular velocity, tilt mechanism deflection angle and tilt mechanism deflection angular velocity. The real-time aerodynamic environment data includes real-time dynamic pressure parameters. The global modeling module is used to establish an open-loop dynamic model containing an aerodynamic pseudo-potential energy field and an aerodynamic natural dissipation matrix based on the real-time generalized state data and the real-time aerodynamic environment data, wherein the aerodynamic pseudo-potential energy field is scheduled by the real-time dynamic pressure parameters. The topology configuration module is used to receive the set target equilibrium state data, construct the corresponding desired closed-loop Hamiltonian energy function, and establish the desired closed-loop dynamic structure containing the desired interconnect matrix and the desired damping matrix. The matrix mapping and reconstruction module is used to calculate and generate an aerodynamic inertial two-port mapping matrix based on the input real-time dynamic pressure parameters, and to calculate and generate a dynamic damping reconstruction matrix based on the feedback current control command value of the actuator and the preset physical saturation boundary value, in order to update the desired closed-loop dynamic structure. The analysis and control allocation module is used to solve the partial differential algebra matching equation by combining the open-loop dynamic model and the updated desired closed-loop dynamic structure to extract the nonlinear cooperative control law, and output the corresponding control commands to the actuators of the aircraft according to the set input configuration matrix.

Images