APPARATUS, ASSEMBLY AND METHOD FOR CONTROLLING AN ACTUATING SYSTEM OF AN AIRCRAFT IN AN OPEN-LOOP AND CLOSED-LOOP MANNER

Abstract

A device for control and closed-loop control of an actuating system of an aircraft is disclosed. The device has a first input interface, which is configured to receive first input data indicating a reference variable, a second input interface, which is configured to receive second input data indicating a controlled variable, and a control output, which is configured to output a control signal. The control signal indicates a manipulated variable for an actuating system of an aircraft, which is to be controlled by means of the actuating system. The reference variable indicates a target acceleration at a point of the aircraft that is to be controlled by means of the actuating system, and the controlled variable indicates an actual acceleration of the aircraft at the point. Taking into account the reference variable and the controlled variable, the device is configured to determine the manipulated variable, in particular from the difference between the reference variable and the controlled variable, and to output the control signal corresponding to the manipulated variable via the control output. Further, an arrangement for control and closed-loop control of an actuating system of an aircraft as well as a method are provided.

Claims

1. A device for control and closed-loop control of an actuating system of an aircraft, comprising a first input interface, which is configured to receive first input data indicating a reference variable; a second input interface, which is configured to receive second input data indicating a controlled variable; and a control output, which is configured to output a control signal that indicates a manipulated variable for an actuating system of an aircraft, which is to be controlled by means of the actuating system, wherein the reference variable indicates a target acceleration at a point of the aircraft that is to be controlled by means of the actuating system; the controlled variable indicates an actual acceleration of the aircraft at the point; and, taking into account the reference variable and the controlled variable, the device is configured to determine the manipulated variable, preferably from the difference between the reference variable and the controlled variable, and to output the control signal corresponding to the manipulated variable via the control output.

2. The device according to claim 1, comprising a third input interface, which is configured to receive third input data indicating an actuating system controlled variable, wherein the device is configured to determine an actuating system reference variable taking into account the reference variable and the controlled variable, and to determine the manipulated variable taking into account the actuating system reference variable and the actuating system controlled variable.

3. The device according to claim 2, wherein the actuating system reference variable is a target positioning speed of the actuating system, and the actuating system controlled variable is an actual positioning speed of the actuating system.

4. The device according to claim 1, wherein the device is configured to determine the manipulated variable without considering an actual actuator position of the actuating system, and without determining a target actuator position of the actuating system.

5. The device according to claim 1, comprising an additional input interface, which is configured to receive additional input data indicating an additional controlled variable, wherein the additional controlled variable indicates an actual acceleration of the aircraft at an additional point; and the device is configured to adjust the controlled variable taking into account the additional controlled variable, and subsequently determine the manipulated variable taking into account the reference variable and the controlled variable.

6. An arrangement for control and closed-loop control of an actuating system of an aircraft, comprising an aircraft, having an actuating system, which is configured to control the aircraft in at least one degree of freedom, and an acceleration sensor, which is arranged at one point of the aircraft; a flight control device with an output interface; and a device according to one of the preceding claims, wherein the flight control device is configured to calculate the reference variable indicating a target acceleration at the point of the aircraft from a flight status of the aircraft, and transmit the first input data indicating the reference variable to the first input interface of the device via the output interface; the acceleration sensor is configured to measure the local acceleration of the aircraft at the point, and transmit second input data indicating the controlled variable to the second input interface of the device, which indicate the local acceleration at the point; and the actuating system is configured to receive the manipulated variable from the control output of the device, and perform a positioning movement corresponding to the manipulated variable.

7. The arrangement according to claim 6, wherein the flight control device is configured to calculate the reference variable taking into account a actuating variable determined in a directly kinematic manner from a target trajectory of the aircraft.

8. The arrangement according to claim 6, wherein the actuating system is formed with an actuator that moves a flight control surface of a flight control surface assembly of the aircraft.

9. The arrangement according to claim 8, wherein the acceleration sensor is arranged on a part of the flight control surface assembly that is immovable relative to the aircraft.

10. The arrangement according to claim 6, comprising an additional acceleration sensor, which is arranged at an additional point of the aircraft, wherein the device is a device according to claim 5; and the additional acceleration sensor is configured to measure the local acceleration at the additional point, and transmit the additional input data indicating the additional controlled variable to the additional input interface of the device, which indicate the local acceleration at the additional point.

11. The arrangement according to claim 1, wherein the aircraft has an additional actuating system, which is configured to control the aircraft in the at least one degree of freedom or in at least one additional degree of freedom, and has an additional acceleration sensor, which is arranged at an additional point of the aircraft, wherein the flight control device is configured to also transmit the input data indicating the reference variable to the additional device via the output interface; the additional acceleration sensor is configured to measure the local acceleration of the aircraft at the additional point, and transmit second input data indicating an additional controlled variable to the additional device, which indicate the local acceleration at the additional point; and the additional actuating system is configured to receive the manipulated variable from the control output of the additional device, and perform a positioning movement corresponding to this manipulated variable.

12. The arrangement according to claim 6, wherein the aircraft is a highly flexible aircraft.

13. A method for control and closed-loop control of an actuating system of an aircraft, with the steps of providing a device for control and closed-loop control of an actuating system of an aircraft; generating first input data indicating a reference variable, wherein the reference variable indicates a target acceleration at a point of the vehicle that is to be controlled by means of the actuating system; generating second input data indicating a controlled variable, wherein the controlled variable indicates an actual acceleration of the vehicle at the point; receiving the first input data at a first input interface of the device; receiving the second input data at a second input interface of the device; determining a manipulated variable for an actuating system of the aircraft taking into account the reference variable and the controlled variable, preferably from the difference between the reference value and the controlled variable; and outputting a control signal indicating the manipulated variable via a control output of the device.

14. The method according to claim 13, comprising receiving third input data that indicate an actuating system controlled variable at a third input interface of the device, wherein determining the manipulated variable taking into account the reference variable and the controlled variable comprises determining an actuating system reference variable taking into account the reference variable and the controlled variable, and determining the manipulated variable taking into account the actuating system reference variable and the actuating system controlled variable.

Description

DESCRIPTION OF EXEMPLARY EMBODIMENTS

[0055] Additional exemplary embodiments will be described in more detail below with reference to figures of a drawing. Shown here on:

[0056] FIG. 1 is a known arrangement for control and closed-loop control of an actuating system of an aircraft;

[0057] FIG. 2 is an arrangement for control and closed-loop control of an actuating system of an aircraft according to the disclosure;

[0058] FIG. 3 is an arrangement of an acceleration sensor on an elevator of an aircraft;

[0059] FIG. 4 is another arrangement for control and closed-loop control of an actuating system of an aircraft;

[0060] FIG. 5 is a schematic illustration of a concept for an acceleration-based roll position control of an aircraft;

[0061] FIG. 6 is a schematic illustration of a concept for an acceleration-based control of mechanical systems;

[0062] FIG. 7 is a schematic illustration of a concept for an acceleration-based control of elastic aircraft;

[0063] FIG. 8A-E is the overall system dynamics for a known as well as for a disclosed closed-loop control of an actuating system of an aircraft;

[0064] FIG. 9 is a Bode diagram for a previously known and for an embodiment according to the disclosure of a closed-loop control of an actuating system of an aircraft;

[0065] FIG. 10 is a schematic illustration of an arrangement on a flexible aircraft; and

[0066] FIG. 11 is a schematic illustration of an alternative arrangement on a flexible aircraft.

[0067] FIG. 1 shows an arrangement for controlling and regulating, i.e. for control and closed-loop control of, an actuating system of an aircraft according to a known approach. A flight control device 1 of the aircraft, which involves an automatic control system that can also be referred to as a flight control system, herein receives measured variables 2 for describing the movement state of the aircraft. Based on the movement state of the aircraft, the flight control device 1 determines a reference variable 3 for controlling the actuating system 4. Herein, the actuating system 4 is formed with a control element 4a and a force generator 4b, wherein the actuating system can comprise additional components. In particular, the force generator 4b can be a flight control surface of a flight control surface assembly, and possibly fixed components of the control assembly that participate in generating the force. For example, the force generator 4b can alternatively be a nozzle or propeller. The control element 4a is used to influence the force generator 4b in such a way that a desired force acts upon the controlled system, i.e., the aircraft. In configurations where the actuating system 4 is a flight control surface assembly, the control element 4a can be a servomotor, which swivels a flight control surface of a flight control surface assembly as a force generator 4b or part of the force generator relative to an immovable part of the flight control surface assembly. In alternative configurations, for example, the control element 4a can be a valve of a nozzle serving as the force generator 4a, or a drive motor of a propeller.

[0068] The reference variable 3 indicates a target actuator position of the control element 4a of the actuating system 4, for example a rotational position of a servomotor (corresponding to a flight control surface position), an opening state of a valve of a nozzle, or a drive position of a drive motor of a propeller, which results in a propeller speed, or a servomotor for blade angle adjustment.

[0069] A control device 5 of the arrangement receives the reference variable 3 via a corresponding input interface. In addition, the control device receives a controlled variable 6 by way of another input interface, which indicates the actual actuator position of the actuating system. The control device determines the control deviation as the difference between the target value of the actuator position according to the reference variable 3 and the actual value of the actuator position according to the controlled variable 6. An actuating system reference variable is determined from the control deviation through multiplication by a proportionality factor in the control device 5, and is compared to an actuating system controlled variable 7, so as to determine a manipulated variable 8 of the actuating system. For example, the manipulated variable 8 can be an actuator voltage or an actuator current.

[0070] The control element 4a effects a position of the force generator 4b based on the manipulated variable 8. As a result, a force and/or torque effect 9 acts upon the mechanical system 10 of the aircraft. While the aircraft as a mechanical system 10 is shown separately from the remaining components in FIGS. 1 and 2, the actuating system 4 and, in advantageous embodiments, the flight control device 1 and control device 5 also form part of the aircraft.

[0071] Apart from the desired force and/or torque effect 9, the mechanical system 10 of the aircraft is also exposed to disturbing forces and/or torques 11, which are caused by outside influences, for example wind exposure, in particular in the form of wind gusts. As can be discerned from the illustration in FIG. 1, the disturbing forces and/or torques 11 acting on the aircraft can only be compensated for by the flight control device 1 if measured variables 2 that include the influence of the disturbing forces and/or torques 11 are considered. As a consequence, this type of consideration takes place exclusively within the framework of flight control, which is usually slow by comparison to the actuating system control (servocontrol).

[0072] FIG. 2 now shows an arrangement for control and closed-loop control of an actuating system of an aircraft according to the disclosure. According to the disclosure, in comparison to the arrangement in FIG. 1, a device 12 for control and closed-loop control of the actuating system 4 is provided that is configured to receive a reference variable 13 at a first input interface, which reference variable 3 indicates a target acceleration at a point of the aircraft. To this end, the flight control device 1 is configured to determine the reference variable 13 indicating the target acceleration from the measured variables 2, and to transmit it to the device 12. At a second input interface, the device 12 receives a controlled variable 14 that indicates the actual acceleration at the point of the aircraft.

[0073] In particular, the acceleration of the aircraft can be a local acceleration at the actuating system 4. FIG. 3 exemplarily shows the arrangement of an acceleration sensor 15 on the immovable part of an elevator 16 of an aircraft. Herein, the elevator 16 is an actuating system 4 of the aircraft in which a movement of the flight control surface of a flight control surface assembly functioning as the force generator 4a by means of a servomotor as the control element 4a exerts a force on the aircraft, which leads to a pitching, and thereby causes the aircraft to rise or sink.

[0074] Alternatively, the acceleration can be an acceleration at another point of the aircraft, for example in a center of gravity of the aircraft. The acceleration can be measured directly with an acceleration sensor, or determined from one or several measured values, which can include accelerations at one or several other points or other variables than accelerations, for example vertical movements (changes in position) of the wings.

[0075] The control device 12 determines the control deviation as the difference between the target value for the acceleration according to the reference variable 13 and the actual value for the acceleration according to the controlled variable 14. An actuating system reference variable is determined from the control deviation in the control device 12 through multiplication by a proportionality factor, and compared with an actuating system controlled variable 7, so as to determine a manipulated variable 8 of the actuating system. Based on the manipulated variable 8, the control element 4a produces a positioning of the force generator 4b that leads to a force effect 9 on the mechanical system 10 of the aircraft.

[0076] In an alternative configuration, such a cascade structure can be replaced by a parallel feedback, in which the controlled variables 7 and 14 are fed back, and are herein each modified, in particular multiplied by an amplification factor and/or integrated. The reference variable 13 is modified according to the controlled variables 7 and 14 by means of a prefilter, after which the manipulated variable 8 is determined by adding the controlled variables 7, 14 and reference variable 13.

[0077] As may be seen in FIG. 2, the controlled variable 14 is a variable of the mechanical system 10 of the aircraft. The disturbing forces and/or torques 11 influence the acceleration of the aircraft, so that the acceleration of the aircraft indicated with the controlled variable 14 already contains these influences, at least in part. In comparison to the known concept illustrated in FIG. 1, the control concept illustrated in FIG. 2 thus already achieves a consideration of disturbing forces and/or torques 11 acting on the aircraft in the control of the actuating system 4 of the aircraft.

[0078] In particular, the manipulated variable 8 can be an actuator voltage or an actuator current. For example, the actuating system reference variable can be a target value for a positioning speed of the control element 4a, i.e., in particular of an actuator. In this case, the actuating system controlled variable 7 can be an actual positioning speed of the control element 4a. In an exemplary configuration, a target value for an actuator current is determined from the difference between the actuating system reference variable and the actuating system controlled variable. The target variable for the actuator current can be the manipulated variable 8. Alternatively, an additional inner control loop can be provided, in which the manipulated variable 8 is determined using the target value of the actuator current.

[0079] FIG. 4 shows such a configuration of an arrangement for control and closed-loop control of an actuating system of an aircraft, in which another inner control loop is provided. Exemplarily shown here is a pitching position control with a rotary, electromagnetic actuator of an elevator. As opposed to a known control with feedback of a flight control surface deflection n, a local acceleration b.sub.zH at the elevator is fed back, and a control deviation to a specified acceleration b.sub.zH,c is determined within the actuating system control 17. Applying the factor K.sub.bZH, a proportional positioning speed command {dot over (n)}.sub.c is determined from the above, which is set by the inner speed control loop. The commanded current flow I.sub.c (actuator reference variable) results proportionally (factor K.sub.n) from the speed error {dot over (n)}.sub.c−{dot over (n)}, the difference between the positioning speed command {dot over (n)}.sub.c (actuating system reference variable) and the actual positioning speed n, which is the actuating system controlled variable. Finally, the terminal voltage U of the motor forms the manipulated variable. It is set proportionally (factor K.sub.I) to the control error I.sub.c−I of the current control cascade. Herein, the actual current I constitutes the actuator controlled variable.

[0080] Within the framework of the physical processes within the actuator corresponding to a modeling as a DC shunt machine, the terminal voltage causes a change in the current flow in the motor windings that is anti-proportional to its inductivity L. However, consideration must be given to the voltage drop ΔU.sub.res=R.Math.I owing to the winding resistance R, as well as to the counter-voltage ΔU.sub.emf=K.sub.e.Math.{dot over (n)} induced by the rotational movement proportional to the motor constant K.sub.e, which diminish the terminal voltage. The current flow I arises through integrating the current change, and produces a drive torque M.sub.act proportional to the motor constant Kt.

[0081] With respect to the physical effect on the actuating system, in addition to the drive torque M.sub.act, the aerodynamic rudder hinge moment M.sub.aero acts on the flight control surface, which along comprises both components proportional to the deflection n with the factor C.sub.n,aero and damping components (factor C.sub.{dot over (n)},aero). In addition, the aerodynamic rudder hinge torque M.sub.aero is influenced by the direction of inflow (factor C.sub.α,aero). The resulting overall torque leads to a positioning acceleration {umlaut over (n)} that scales with the inverse 1/J of the rotational inertia.

[0082] Shown in the right part of FIG. 4 is a simplified view of the dynamics underlying the aircraft pitching movement 18. The pitching acceleration q is proportional to the pitching torque with the inverse pitching inertia 1/I.sub.yy, which arises from the pitching torque coefficient through denormalization with dynamic pressure q, wing area S and wing depth I.sub.μ. The latter essentially comprises influences of the elevator (C.sub.mn.Math.n), pitching rate (C.sub.mq.Math.I.sub.μ.Math.1/V.sub.A.Math.{dot over (q)}) and angle of attack (C.sub.mα.Math.α). Apart from the share of elongation Θ, the angle of attack α is determined by the influence γ of the path movement 19. In addition, it contains the main part of the disturbing influence (gusts) in the form of the wind adjustment angle α.sub.W. The local acceleration b.sub.zH at the elevator arises from the pitching acceleration {dot over (q)} with the lever r.sub.H, as well as from the vertical acceleration b.sub.Z of the aircraft center of gravity.

[0083] According to the disclosure, the actuator position is not drawn upon as the controlled variable, for example as evident from FIG. 4. No force or torque measurement serves as the controlled variable either. In addition, the controlled variable is not measured in the drivetrain of the actuator or on the flight control surface, but rather on the assembly allocated to the flight control surface (the lift surface immovable relative to the aircraft) in the embodiment of FIG. 4. No measurement of the (rotational) acceleration {umlaut over (n)} of the actuator takes place that would be proportional to the positioning torque (drive torque of the actuator, M.sub.act). Rather, the local acceleration on the lift surface instead behaves proportionally to the flight control surface angle and the lifting force it generates, i.e., to a variable that is separated from the acceleration {umlaut over (n)} of the actuator by two integration steps, as evident in FIG. 4. The local acceleration is an output variable which to a substantial extent depends on the flight control surface angle n as a system state, and the feedback of which thus enables influencing the system dynamics in a similar manner. According to the embodiment shown in FIG. 4, the speed control loop of a classic servocontrol (middle cascade in FIG. 4) is to be retained, so that there still is a continued feedback of a number of linearly independent output variables corresponding to the system order. This can make it possible to configure the system dynamics as desired. As “rigid” a layout of the rudder angle dynamics as possible may herein be desired. In particular, reducing the actuator load or positioning effort might not be the goal; rather, it can be provided that the flight control surface be moved as quickly as possible into the position that compensates for the influence of gusts on the corresponding flight control surface assembly. This position is generally not identical to the resting position, into which the free rudder would be deflected with the setting torque held constant.

[0084] Local acceleration control can yield advantages over controlling the rudder hinge torque. The local acceleration measurement (as opposed to the flight control surface angle or rudder hinge torque) directly captures the added lift caused by the gust via the additional angle of attack aw. In elastic aircraft, the local accelerations directly reflect the structural dynamic vibration state. Feeding the acceleration back to the positioning speed of a flight control surface acting at the same location corresponds to a virtual dampening (similar to the so-called ILAF principle). Therefore, it can be suitable in particular for actively stabilizing highly elastic configurations. Furthermore, the local acceleration includes influences of various flight state variables (Θ, γ, q, see FIG. 4), which can also be compensated for by the control. These influences can become less important as compared to the highly dynamic feedback path via K.sub.bz,H, so that a significantly larger robustness can arise in relation to variable aerodynamic properties. In a direct, purely kinematic relation, the local acceleration can be determined from a planned path and attitude trajectory. This makes it possible to derive simple pilot control laws, which are independent of the properties of a specific aircraft. In this way, the high dynamics of the local acceleration control (which correspond to the classic position control loop of the servocontrol) can be taken advantage of not just for interference suppression, but also for guidance behavior. This can enable a significantly more agile path guidance.

[0085] Actuator control (servocontrol) and flight state control (flight control) represent traditionally separate research disciplines, which are covered in different expert circles. The feedback of a local acceleration measured on the aircraft structure in an inner control loop, which is traditionally part of the servo control, builds a bridge between the two areas. This requires a holistic examination of the entire controlled system, which interprets the aircraft and its control elements as a unit. Using the local acceleration as a default variable makes it possible to include parts of the flight dynamic in the controlled system of the servocontrol. It can become possible to simplify the controlled system of the flight control, and reduce dependencies on specific flight properties, so that classic flight control structures are no longer applicable.

[0086] According to illustration 4, the boundary for the actuating system control 17 is drawn at local acceleration b.sub.zH and flight control surface deflection n. Other illustrations are possible, in which the definitions of subsystems, in particular of the boundaries, are set differently (e.g., see FIG. 5), without this resulting in a change in the disclosed control principle.

[0087] The symbols used in FIGS. 5, 6 and 7 denote the following variables:

[0088] Scalars:

[0089] C.sub.Iβ: Sliding roll torque

[0090] C.sub.Iξ: Aileron effectiveness

[0091] C.sub.Ip: Roll damping

[0092] I: Actuator current

[0093] I.sub.yy: Rolling inertia torque

[0094] J: Torque of Inertia of the actuator

[0095] K.sub.t: Torque constant of the actuator

[0096] K . . . : Controller amplification of the . . . -control loop

[0097] S: Wing surface

[0098] V.sub.A: Flight speed

[0099] q: Dynamic pressure

[0100] b: Half span

[0101] p: Roll rate

[0102] β: Shift angle

[0103] β.sub.W: Wind shift angle

[0104] Ω: Angular velocity of the actuator

[0105] ξ: Aileron deflection

[0106] Vectors:

[0107] η: Modal amplitudes (structural dynamic degrees of freedom)

[0108] R: Position vector for the local acceleration measuring point

[0109] g: Generalized coordinates

[0110] u: Manipulated variables

[0111] x: Rigid body degrees of freedom

[0112] z: Disturbance variables

[0113] Matrices and Tensors:

[0114] BηPositioning influence on generalized forces of the structural dynamic degrees of freedom

[0115] Bχ: Positioning influence on generalized forces of the rigid body degrees of freedom

[0116] B: Positioning influence on generalized forces

[0117] C: Generalized rigidity matrix

[0118] D: Generalized damping matrix

[0119] E.sub.η: Disturbance influence on generalized forces of the structural dynamic degrees of freedom

[0120] E.sub.x: Disturbance influence on generalized forces of the rigid body degrees of freedom

[0121] E: Disturbance influence on generalized forces

[0122] E.sub.η.sup.ext, E.sub.η.sup.ext: Influence of the structural deformation-induced aerodynamic forces on rigid body movement

[0123] K . . . : Amplification matrix of the . . . -control loop

[0124] L: Kinematic translation ratios between generalized rigid body degrees of freedom and position of the local acceleration measuring points M: Generalized inertia matrix

[0125] Q.sub.η, Q.sub.η: Influence of the structure deformation-induced aerodynamic forces on structural dynamics

[0126] Q.sub.x, Q.sub.x: Influence of the rigid body movement-dependent aerodynamic forces on structural dynamics

[0127] Δ: Eigenforms (eigenvectors) of the structural dynamics

[0128] β: Generalized structural damping factors

[0129] γ: Generalized rigidity matrix

[0130] μ: Modal mass matrix

[0131] Indices:

[0132] c: Command size, default value, target value

[0133] In classic flight control, the command corresponds to the position (angle) of the aerodynamic flight control surface. A highly dynamic (rigid) positional control of the actuator ensures that the actual flight control surface position precisely follows the positioning command. The control structure corresponds to a cascade control with an inner control loop, the actuator control (ACL), and an outer control loop, the flight control (FCL). A feedback of position angles, rotation rates and speeds takes place. As a rule, acceleration measurements are only used for observation or as a replacement for poorly measurable states.

[0134] Also known is a rudder hinge torque-based flight control. The command for the FCL corresponds to a torque specification, meaning a direct current specification, for the actuator. In a state of equilibrium, the torque specification corresponds to the aerodynamic rudder hinge torque. The concept is similar to the force-oriented control behavior of the pilot during manual control. This type of control is supposed to offer advantages with respect to flight silence and load reduction, since the control surface deviates owing to an altered hinge torque of the gust. This is intended to reduce an actuator load and force fight in the case of redundant actuators.

[0135] A local linearization and inversion of the system dynamics takes place in the likewise previously known incremental nonlinear inversion (INDI). Incremental growths in the positioning command are calculated. The method is based on measured and commanded (rotational) accelerations, and reduces the influence of the (aerodynamic) model accuracy and center of gravity for elevated robustness. The positioning law is herein based upon the comparison between planned and actual changes (and thus, derivations) of the state variables, which are calculated or observed based on rotatory and translatory acceleration measurements. As opposed to the concepts disclosed herein, a direct use of this change in positional variable in an inner cascade of the servocontrol or an expansion of the INDI approach to the actuator dynamics is not known for this approach. In a proposed approach, the actuator current serves as a given variable, and a positioning law modified for this purpose is derived. As opposed to the approach according to the present disclosure, the quasi-stationary dependence of the actuator current on the rudder hinge torque is taken as the basis, so that the dynamics of the actuating system themselves remain unregulated.

[0136] Feeding back acceleration measurements or modal degrees of freedom is known for an active flutter control and load reduction. Herein, the command corresponds to the flight control surface position. Alternatively, additional forces are applied by vibration actuators. This often does not take place in terms of closed-loop control, but specifically to compensate for individual resonance frequencies.

[0137] In the known systems, the dynamics (bandwidth) of the flight controller to a large extent determine the precision of path and position maintenance (interference suppression), flight silence (interference suppression), and agility of path guidance (guidance behavior). The maximum bandwidth is limited by the dynamics of the independently configured actuator control (inner control loop), and possibly also by the dynamics of the mechanical transmission path between the actuator and flight control surface, the structural dynamics of an elastic aircraft, and the transient aerodynamics. A precise aerodynamic model is required for an optimal FCL configuration. This is costly and can be associated with a lack of robustness. The inner control loops, at least the position control, must be individually designed for each aircraft type. A precise aeroelastic model is required to preclude excitations of the structural dynamics. Having the flap deflection ilk act directly on the vertical load multiple (i.e., the load acceleration) n.sub.Z complicates the design of a gust load control. Abatement potential is limited without the provision of a pilot control, which is accompanied by a complex angle of attack measurement.

[0138] FIG. 5 shows a schematic illustration of a concept for an acceleration-based rolling position control of an aircraft. In comparison to the known system, the position control of the actuators is replaced by the feedback of an acceleration measurement, which determines the aerodynamic force effect of the flight control surface (for example, local acceleration at the flight control surface or rotational acceleration of the aircraft). The classic division between actuator control and flight control is altered herein. The interface between FCL and actuator control slides inwardly by one cascade. The state feedback of the actuator deflection is replaced by an output feedback of the acceleration proportional thereto, which additionally contains the interference influence (gusts). The command of the FCL then corresponds to the positioning rate (angular velocity) of the flight control surface. A measurement of the flight control surface position is only required to consider the positional limit.

[0139] In the case of an aircraft, the controlled system of the FCL, the rolling torque coefficient C.sub.I is proportional to the aileron deflection ξ, which in known systems constitutes the manipulated variable, with the factor C.sub.Iξ. The rolling torque coefficient C.sub.I β is proportional, with the factor C.sub.I, to the shift angle β, which is construed as a disturbance variable for pure rolling control, and in particular incorporates the wind influence β.sub.W. The rolling torque coefficient C.sub.I is proportional, with the factor C.sub.Ip, to the dimensionless rolling rate p*=p.Math.b/VA. The rolling torque follows from the coefficient C.sub.I through multiplication by the reference variables (q, S, b), with the rolling acceleration also being proportional to the inverse rolling inertia (1/I.sub.yy). The rolling rate p and rolling angle Φ arise through integration from the rolling acceleration. Known flight control comprises the complete feedback of the states “rolling rate (p)” and “suspension angle (Φ)”. Herein, the system is set up in the form of a cascade control, in which the outer control loop comprises the rolling position with the reference variable “rolling command (Φ.sub.c)” and manipulated variable “rolling rate command (p.sub.c)”, which with amplification K.sub.Φ is proportional to the control error Φ.sub.c-Φ. The inner control loop then relates to the rolling rate with the reference variable “rolling rate command (p.sub.c)” and manipulated variable “aileron command (ξc)”, which with the amplification K.sub.p is proportional to the control error p.sub.c-p.

[0140] For the actuator, the control system ACL, the actual current flow I corresponds to the current command I.sub.c when disregarding the electrical time constant. The torque with torque constant K.sub.t is proportional to the current flow. Additional torque coefficients (friction, aerodynamic rudder hinge torque, etc.) are disregarded. The rotational acceleration (ω.sup.−) of the downforce follows from the conservation of angular momentum as the torque/inertia torque (J). The positioning speed (ω) and downdraft angle (which corresponds to the aileron deflection ξ) follow through integration of the rotational acceleration. Known actuator control then involves the complete feedback of the states “positioning speed (ω)” and “actuator position (ξ)”. The system is set up in the form of a cascade control, in which the outer control loop comprises the actuator position with the reference variable “aileron command (ξ.sub.c)” and manipulated variable “setting rate command (ω.sub.c)”, which with the amplification K.sub.ξ is proportional to the control error ξ.sub.c-ξ. The inner control loop then relates to the positioning speed with the reference variable “setting rate command (ω.sub.c)” and manipulated variable “current command (Ic)”, which with the amplification K.sub.ω is proportional to the control error ω.sub.c-ω.

[0141] By comparison to the above, FIG. 5 shows an acceleration-controlled concept with the controlled system aircraft 20, the actuator 21 and the controlled system actuator 22. The measured rolling acceleration ({dot over (p)}) is fed back instead of the aileron deflection (ξ) proportional thereto. The command ω.sub.c of the FCL corresponds to the positioning rate ξ. Apart from the flight control surface position ξ, the disclosed feedback also directly acquires the interference influence through β or β.sub.W. As a consequence, the disturbance is already compensated for one control loop further in than for known flight control. Given a highly dynamic configuration of this acceleration control loop ({dot over (p)}-feedback), significantly better interference suppression can be achieved. The default value for the position control loops (traditional “inner loops” of the FCL) corresponds directly to the rate acceleration (second derivation of the controlled variable). Given a highly dynamic configuration of acceleration control, the aircraft directly follows the specified rate acceleration ({dot over (p)}.sub.c≈{dot over (p)}). This yields a simple, linear behavior independent of aircraft-specific parameters. The configuration of position control can be standardized, and can take place independently of the aircraft type and flight status. Given a highly dynamic configuration, the acceleration feedback via K.sub.{dot over (p)}.Math.C.sub.Iξ becomes dominant relative to the remaining aerodynamic influences (via C.sub.Iβ and C.sub.Ip), whereby the influence of aerodynamic parameters (other than C.sub.Iξ) on the control circuit is reduced. This makes it possible to forego a precise aerodynamic model for the FCL configuration. Only the rudder effectiveness C.sub.Izi and dynamic pressure remain relevant. Preventing the positioning command from acting directly on the acceleration measurement simplifies the configuration and elevates the potential of control-based gust load reduction.

[0142] For rigid aircraft, the principle can be transferred to the pitching degree of freedom (measured variable: pitching acceleration, primary manipulated variable: elevator), the yaw degree of freedom (measured variable: yaw acceleration, primary manipulated variable: rudder), lift degree of freedom (measured variable: vertical acceleration n.sub.z, primary manipulated variable: flap), longitudinal degree of freedom (measured variable: longitudinal acceleration n.sub.x, primary manipulated variable: spoiler), as well as transverse degree of freedom (only for lateral force control). The acceleration component is ideally fed back not just to the primary manipulated variable, but to all manipulated variables that influence the respective degree of freedom, for example via the aileron rolling torque, rudder yaw torque, elevator lift or flap pitching torque. The degrees of freedom can be completely decoupled by suitable selection of the amplification matrix. The described degrees of freedom, the acceleration of which is measured, can be chosen as desired. For example, the rotation and translation of the center of gravity is named in aircraft-fixed coordinates. Likewise conceivable are other coordinate systems, as well as other (possibly even several) reference points of the rigid body, for example the vertical position of both wing tips instead of the rolling angle. Any combination of independent degrees of freedom that clearly describes the system is possible. The latter constitutes a valid set of generalized coordinates (q) in the sense of Lagrange formalism.

[0143] Based on a schematic illustration of a concept for an acceleration-based control, FIG. 6 shows the transferability of the disclosed control concept to general mechanical systems. The principle can be transferred to any mechanical system 23 with n degrees of freedom, which can be clearly described by generalized coordinates in terms of Lagrange formalism, which is controlled by one or several manipulated variables that exert a direct force or torque influence on the system 23, and the manipulated variables of which are operated by a controlled actuator 24 with at least simple integrating behavior (all mechanical actuators).

[0144] Given an actuator position-controlled approach, as opposed to the system according to FIG. 6, the manipulated variable u.sub.i has a force influence that can be described by generalized forces Q.sub.i=B.sub.ij({dot over (q)},q).Math.u.sub.i. The same holds true for force or torque disturbances z.sub.i with impact factors E.sub.ij({dot over (q)},q). The acceleration {umlaut over (q)}.sub.i of the generalized coordinate is proportional with M.sub.ij.sup.−1(q) to the generalized force Q.sub.i. Generalized speeds {dot over (q)} and coordinates q follow through integration. Generalized speeds {dot over (q)} produce non-conservative “damping forces” D({dot over (q)},q).Math.{dot over (q)} in dissipative systems. Conservative forces are proportional with C({dot over (q)}) to generalized coordinates q. In known control concepts, all states are completely fed back, represented by the generalized coordinates (q) and speeds ({dot over (q)}). Buildup takes places in the form of a cascade control, wherein the outer control loop relates to generalized coordinates. Their reference variable comprises the target values of the generalized coordinates (q.sub.c). The manipulated variable consists of the target values for the generalized speeds ({dot over (q)}.sub.c), which are proportional with the amplification matrix (K.sub.g) to the control error q.sub.c-q. The inner control loop relates to generalized speeds, wherein the reference variable comprises target values for the generalized speeds ({dot over (q)}.sub.c), and the manipulated variable comprises positioning commands (u.sub.c), which are proportional to the control error {dot over (q)}.sub.c-{dot over (q)} with the amplification matrix K.sub.q.

[0145] The actuator has an arbitrary transfer behavior G(s) between the commanded and actual change in the manipulated variable {dot over (u)}, but at least one integration stage. Actuator control comprises the feedback of (at least) manipulated variables u, wherein the reference variable is the target value for the system manipulated variables u.sub.c. The actuator manipulated variable is the target value for the system positioning rates {dot over (u)}.sub.c, which is proportional to the control error u.sub.c-u with amplifications K.sub.u.

[0146] By comparison to known controls, the measured, generalized accelerations {umlaut over (q)} are fed back in the acceleration-controlled approach according to FIG. 6 instead of the manipulated variables u proportional thereto. The control command corresponds to the setting rate {dot over (u)}.

[0147] FIG. 7 illustrates a concept for an acceleration-based control of elastic aircraft. FIG. 7 here shows the general case of a complete state feedback. An elastic aircraft constitutes a special case of the disclosed concept explained in connection with FIG. 6, since it can be described by Lagrange formalism. In principle, generalized coordinates can be selected as desired. One possibility involves individual positions of acceleration sensors distributed over the aircraft. The measured acceleration herein corresponds directly to the generalized acceleration {umlaut over (q)}. This requires at least six sensors for acquiring the rigid body movement. The number of additional sensors determines how many elastic modes can be acquired. An alternative option involves a separation of rigid body movement (mean axes) and structural dynamics. Herein, a division takes places into rigid body degrees of freedom (x=[x, y, z, Φ, Θ, Ψ].sup.T) and amplitudes (η=[η.sub.1, η.sub.2, . . . η.sub.n].sup.T) of the elastic modes, so that q=[x.sub.i, η.sub.i].sup.T. The movement equations for the rigid body movement and structural dynamics are inertially decoupled, but a coupling by way of outside forces (aerodynamics) does exist.

[0148] The controlled system comprises the rigid body dynamics (below in FIG. 7), which is built up similarly to the system in FIG. 6, wherein the correlations q=x, B=B.sub.x, and E=E.sub.x apply. With respect to structural dynamics (above in FIG. 7), the manipulated variable u.sub.i has a force influence that can be described by generalized forces Q.sub.j.=B.sub.η, ij({dot over (q)}, q). The same applies to force and torque disturbances (z.sub.i) with the impact factors E.sub.η, ij({dot over (q)}, q). The second derivation of the modal amplitude {umlaut over (η)}.sub.i is proportional to the generalized force Q.sub.j with the inverse modal mass matrix μ.sup.−1(ij). The structural damping produces damping forces, which are proportional to the rate of change in the modal amplitudes {dot over (n)}.sub.i with damping factors β. The structural elasticity produces spring forces that are proportional to the modal amplitudes with the generalized rigidity matrix y.

[0149] Outside forces produce a coupling between the structure movement and rigid body movement. Herein, the outside forces (aerodynamic forces/torques) depend on rigid body states {dot over (x)} and {hacek over (x)} and structural dynamic states {dot over (n)} and n. Outside forces influence both the rigid body movement ({umlaut over (x)}) and the structural dynamics ({dot over (η)}). The portion of the forces on the rigid body movement dependent on rigid body movement was already considered by D, C. The influence of the structural deformation-induced portion of forces on the structural dynamics (Q.sub.η and Q.sub.{dot over (η)}) is (by contrast) not already contained in β, y. The dependence of forces on rigid body movement yields an influence on the structural dynamics: Q.sub.x, Q.sub.{dot over (x)}. The outer control loops relate to the generalized coordinates (q=[x.sub.i, η.sub.i].sup.T), the inner control loops to the local degrees of freedom (R.sub.f). The system behavior depends on the description form (transformation between various degrees of freedom systems/state illustrations).

[0150] A specific case of the concept according to FIG. 7 relates to a local acceleration feedback. Herein, the acceleration is measured directly at the location of the flight control surface. The feedback matrix K.sub.R is only diagonally occupied, meaning that the acceleration acts exclusively on the flight control surface where the measurement takes place. The system dynamics can be freely specified, provided the number of setting/measuring positions corresponds to the number of (considered) degrees of freedom (and have been suitably selected, i.e., are linearly independent; this leads to controllability and observability). By contrast, providing a complete eigenstructure is not possible. The system remains coupled. Assuming that the positioning rate command is converted without any delay (disregarding the actuator dynamics), the feedback of acceleration to the positioning rate is equivalent to the feedback of the speed to the actuator position. Assuming that the aerodynamic force generation by the flight control surface takes place without any delay (disregarding transient aerodynamics), a speed-proportional counterforce is generated. The acceleration feedback thus corresponds to a virtual, viscous damper that acts at the location of the flight control surface. Introducing an integrating part or feeding back a local speed measurement would similarly allow introducing a virtual spring element, thereby yielding a more or less rigid clamping of the wing at the location of the flight control surface. Because the damping force always counteracts the direction of movement, an energy supply, and hence a destabilization of the structural dynamic modes is precluded. However, this only applies for as long as the assumptions are justified, i.e., for all structural modes that are clearly lower frequency than the actuator dynamics/transient aerodynamics. This limits the maximum realizable dynamics for acceleration feedback.

[0151] By contrast, the risk of an excitation exists in an acceleration measurement that is locally separate from the flight control surface (e.g., IMU in the cockpit), since the acceleration signal only reacts to the force generated on the flight control surface after a delay caused by the structural dynamics. Expressed differently, an eigenform can possibly exist the vibration antinode of which, at the measurement location, has an opposite sign compared to the location of the force generation (flight control surface). A negative, destabilizing damping force thus arises in the frequency range of this eigenmode. This is precluded if the measurement takes place directly at the location of force generation.

[0152] In classic flight control, the bandwidth limitation arises from the frequency separation to the structural dynamics. By contrast, a bandwidth limitation arises for the local acceleration feedback from the frequency separation to transient aerodynamics and actuator dynamics.

[0153] In particular, advantages of the acceleration-controlled concept can lie in the achievability of a higher dynamic for flight control, and hence improved interference suppression, flight silence and higher agility, especially in the case of highly elastic aircraft, wherein no frequency separation to the structural dynamics or filtering of elastic modes is required. An automatic damping of all elastic modes below the actuator dynamics and in particular the aerodynamics can be enabled, independently of the concrete elastic properties of the aircraft. Interference influences (local gusts) are balanced out directly at the attack site, without exciting the structural dynamics (similar to a bird that only locally spreads feathers to yield to a gust). Structural loads are reduced. The acceleration measurement and control can be integrated into an actuator control (smart actuator). This enables a decentralized system structure. The acceleration-regulated concept permits new redundancy concepts and allows a simple adaption of control laws in the event of a flight control surface failure. Given a sufficient frequency separation between the acceleration control (decentralized in the actuator) and position control (centrally in the FCC), the position control and all higher-level control loops can remain unchanged, since the reduced dynamics of acceleration control are still fast enough. This reflects the fact that, in classic flight control, the failure of a redundant actuator for the same flight control surface as a rule requires no adaption of the FCL.

[0154] FIGS. 8A to 8E show the overall system dynamics in the complex plane of an exemplary embodiment of the arrangement, which can arise from transferring the structure illustrated in FIG. 5 to the pitch degree of freedom. FIG. 8B here shows a magnified cutout of FIG. 8A, FIG. 8C a magnified section of FIG. 8B, FIG. 8D a magnified section of FIG. 8C, and FIG. 8E a magnified section of FIG. 8D. The illustration in FIGS. 8A to 8E is based upon a device for controlling the elevator as per the present disclosure. The longitudinal position θ and pitch rate q are fed back. The pole and zero position distribution is shown, which for the depicted embodiment arises for various feedback amplifications (+-shaped markings). Also shown for comparison is the pole and zero position distribution that arises for a previously known control, in which the adjustment angle of the elevator is controlled by the actuating system control (x-shaped markings). The magnified x- or +-markings denote the pole positions that arise when both feedback amplifications (for q and θ) assume the value zero. For the previously known control (x), this pole distribution corresponds to a pattern known for uncontrolled aircraft, in which one conjugated complex pole pair is to be allocated to the phygoid movement and another to the angle of attack vibration. Feeding back the local acceleration within the framework of the disclosed embodiment, the angle of attack vibration is strongly dampened and split into two real poles. By contrast, this has hardly any influence on the phygoid poles.

[0155] The depicted smaller markings connected by lines denote pole positions that can arise given a simultaneous increase in feedback amplifications for the pitch rate and longitudinal position in a constant ratio. The star-shaped markings denote pole positions that can arise for an advantageous selection of feedback amplifications when a high bandwidth for the control circuit is desired at a damping level that does not drop below the value 0.7.

[0156] FIG. 9 shows a Bode diagram for a previously known 25 and for the disclosed 26 embodiment according to FIGS. 8A to 8E, which can arise for a respectively advantageous selection of feedback amplifications. Shown is the frequency response of an interference transmission function of the vertical wind speed w.sub.Wg (gust) on the longitudinal position θ. In a frequency range below the dynamic of the actuating system control, the disclosed embodiment 26 (solid line) reveals an improved interference behavior, since the transmission function of the previously known arrangement 25 (broken line) has an additional zero point, which nearly compensates for the pole associated with the position control loop of the servocontrol. This zero point is eliminated by the acceleration feedback.

[0157] FIG. 10 illustrates an exemplary embodiment for a flexible aircraft. Without placing any limitation on generality, the view in FIG. 11 is herein limited to the flexibility of the main wing in relation to a bending around the longitudinal axis, a torsion around the transverse axis, as well as the resultant local vertical movements. In addition to the six degrees of freedom of a rigid body, a flexible aircraft has additional degrees of freedom, which describe the deformation state. Apart from the rigid body degrees of freedom of an undeformed reference configuration 27, a conventional presentation form comprises the amplitudes of superposed eigenforms of the elastic modes, which describe characteristic deformation patterns as compared to the reference configuration. Such an eigenform is exemplarily depicted in FIG. 10. Herein, the eigenform can have local extreme points 28, at which the deviations from the reference configuration are greatest, and node points 29, at which the deviations from the reference configuration disappear. In order to control and stabilize the deformation degrees of freedom, a flexible aircraft can have several actuating systems (flight control surface assemblies), which can be comprised of flight control surfaces 30 and possibly fins 31. A device according to the present disclosure can use one or several actual accelerations at any points 32 of the aircraft for control purposes. In an advantageous embodiment, for example, these can be extreme points or node points of one or several eigenforms, but also points at which none of the eigenforms relevant for control purposes has a node point. In particular, the number of used accelerations can correspond to a number of eigenvalues that are to be stabilized or influenced by the control. For example, all accelerations can herein be received by each of the provided devices. Alternatively, a device can only receive those accelerations that can be influenced by an adjustment of the actuating system controlled by the device.

[0158] Shown in FIG. 11 is a design of an arrangement for a flexible aircraft, in which the points 32 of the aircraft where the local acceleration is used for controlling the actuating systems lie in proximity to the flight control surfaces 30. For example, all accelerations can herein once again be received by all disclosed devices. Alternatively, each device can receive only the respective acceleration that is present in proximity to the actuating system controlled by this device.

[0159] The features disclosed in the above specification, the claims and the drawing can be important both individually and in any combination for implementing the different embodiments.