DE102016123629A1 - Control system for a steering test bench - Google Patents

Abstract

The present invention relates to a method for controlling a test bench for testing a steering system of a vehicle, in which a mathematical state space model of the test stand is generated, and at least one actuator for generating a force required for the simulation of steering movements is controlled by means of a controller, the at least one control module comprising a proportional component and an integral component and a state feedback module (258), wherein the control module receives as inputs at least one state quantity (232, 236, 238) of the test bench determined by the state feedback module (258) on the basis of the state space model.

Description

The present invention relates to a method for controlling a test bench for testing a steering system of a vehicle and a test bench for testing a steering system of a vehicle.

In order to test newly developed vehicle components, such as, for example, control units, so-called "hardware in the loop" test stands are generally used, in which a respective test control unit is not tested in a real vehicle environment but in a test bench with a computer-aided simulation. To test electromechanical power steering systems steering test stands are used, simulating the expected forces during a ride by means of an actuator. The actuator is controlled by means of a control system which controls the actuator in dependence on determined by means of a vehicle dynamics model set forces and, as a result, imposes corresponding forces on a respective to be tested steering system.

As a rule, PID controllers are used to control an actuator of a test bench, i. H. Regulators having a proportional component which serves to increase a manipulated variable proportional to a control deviation. Furthermore, a PID controller comprises an integral component, by means of which signals are integrated in time and output as manipulated variable. A differential component of a PID controller responds to a rate of change of the control deviation.

Furthermore, methods are known in which a system behavior for the control by means of a state space representation is represented mathematically. All relationships of respective state variables, input variables and output variables of a system to be described are represented in the form of matrices and vectors. A corresponding state space model is described by two equations, namely a state differential equation of the first order and an output equation.

The state differential equation and the output equation describe a system behavior such as, for example, is based on a control loop of a PID controller.

In the European publication EP 1 435 082 B1 a motion system for a driving simulator is disclosed.

The German publication DE 109 22 037 A1 discloses a method for the electrical simulation of moments of inertia of a specimen clamped in a test bench, in which a movement of the specimen is calculated by means of a model.

A method for operating a vehicle roller dynamometer, in which a transient response of an actuator is taken into account by a simple low-pass filter of the first order, is in the German publication DE 198 06 755 A1 disclosed.

The German publication DE 197 30 851 B4 discloses a method for simulating inertial forces in a stationary vehicle inspection by means of a dynamometer, in which an estimator having a closed circular structure is used to determine a power output of the vehicle.

Against this background, it is an object of the present invention to provide a method for controlling an actuator of a steering test bed, which allows a more accurate transient response of the actuator compared to the prior art.

To achieve the object mentioned above, a method for controlling a test stand for testing a steering system of a vehicle is presented, in which a mathematical state space model of the test stand is generated, and at least one actuator for generating a force required for the simulation of steering movements is regulated by means of a controller, the at least one control module having a proportional component and an integral component and a state feedback module, wherein the control module is supplied as at least one input variable at least one state variable of the test bench determined by the state feedback module on the basis of the state space model.

Embodiments of the present invention will become apparent from the description and the dependent claims.

The proposed method is used in particular for controlling a steering test bench for testing an electromechanical steering system for a vehicle. For this purpose, it is provided according to the invention that a respective actuator of the steering test bed for generating a force to be impressed on a respective steering system is controlled by means of a regulator having at least one control module with a proportional component and an integral component and a state feedback module. This means that the regulator provided according to the invention has both a PI component and a state feedback module.

The controller provided according to the invention is capable of increasing a manipulated variable proportional to a control deviation via the control module. Furthermore, the controller according to the invention is able to aufintegrieren abruptly changing signals and output as a manipulated variable. Furthermore, the controller is configured to perform a state feedback by means of a state feedback module, so that a permanent control difference for constant control and interference signals is avoided. This means that due to the state feedback due to, for example, an influence of a disturbance parameter, constant state differences are avoided by a repeated regulation in an inner control loop of a state feedback module due to the state feedback.

In particular, it is provided that the controller provided according to the invention is formed by supplementing a state feedback module with a PI controller or the control module provided according to the invention and expanding a state vector of the state feedback module by a state variable for a control difference. Accordingly, a control variable to be controlled by the regulator provided according to the invention is composed of components determined by the control module and components determined by the state feedback module.

In the context of the present invention, an input quantity is to be understood as meaning a value or a number of values which are transferred from a first module of a controller to a second module of a controller, i. H. communicatively transmitted.

In one possible embodiment of the presented method, it is provided that values determined by the state feedback module of the at least one state variable of the test bench are fed back within the state feedback module and processed by means of a predetermined feedback vector. It is further provided that the values processed by means of the feedback vector are transmitted to the control module and used to control the at least one actuator.

By using a state feedback module, individual eigenvalues used in a feedback vector of the state feedback module can be arbitrarily specified in a closed control loop or within the state feedback module, so that a behavior of the state feedback module or of a respective value calculated by the state feedback module can be influenced. This means that an internal control loop is formed by the state feedback module provided according to the invention in an overall control loop which comprises the state feedback module and a control module. Since a behavior of the state feedback module can be adjusted by means of a modified feedback vector, a manipulated variable to be set by the overall control loop can be preprocessed within the regulator provided according to the invention, for example to influence a transient response of the manipulated variable. By using a feedback vector, for example, constants K _P and K _{I of} a PI controller can be calculated and used by the control module to control the actuator provided according to the invention.

In a further possible embodiment of the presented method, it is provided that at least one state variable of the following list of state variables is selected as the at least one state variable: a position encoder angle of the actuator, an angular velocity of the actuator, an output signal of a measuring amplifier of the test stand, a track rod path of a test object Steering system and a tie rod speed of the steering system.

On the basis of respective state variables of a respective test stand, which are determined, for example, by means of a sensor or a mathematical observer, a state vector can be formed which is matched with a feedback vector in the state feedback module provided according to the invention, ie multiplied by the feedback vector, for example Feedback vector predetermined behavior to impose the state variables or influence the state variables by the feedback vector so that they show a predetermined behavior. A result of the adjustment of the state vector with the feedback vector can be transmitted, for example, to the control module and represented there with a respective proportional part and / or an integral part of the controller provided according to the invention Constants are compared. The result of the adjustment of the result determined by the adjustment of the state vector with the feedback vector with the respective constants can be transmitted as manipulated variable to the at least one actuator provided according to the invention.

By means of a change of a respective feedback vector, the state feedback module provided according to the invention can be adapted to current requirements, so that a behavior of the controller provided according to the invention changes and, for example, a transient response of the controller or a transient response of a signal generated by the controller is adjusted.

In a further possible embodiment of the presented method, it is provided that a signal amplification of a measuring amplifier used for measuring the at least one state variable is selected as a function of a measuring range required for representing a force applied to the at least one actuator.

By adapting a signal amplification to a measuring range required for imaging a signal, it is possible to reduce the influence of interference signals which, for example, act on a respective controlled variable. A scale required for the representation and a corresponding signal amplification factor can be selected, for example, as a function of a minimum resolution predetermined by a user.

In a further possible embodiment of the presented method, it is provided that the at least one state variable is determined by means of at least one sensor.

In order to optimally image a state of a respective test stand or of a respective actuator to be controlled, signals detected by sensors, such as, for example, power take-offs, which are a current situation of the actuator and / or a current situation of a servomotor included in a respective steering system to be tested describe and correspondingly determined values, for example, into a state vector.

In a further possible embodiment of the presented method, it is provided that the at least one state variable is determined by means of a mathematical observer.

In order to detect a state variable by means of a sensor, it is generally necessary for a control device of a respective actuator or servomotor to be activated, so that corresponding sensor values are read out by the control device, for example by means of a measurement and calibration protocol via a CAN bus can. However, should it be necessary for test purposes to deactivate the control unit of the actuator and / or the servo motor of a steering system to be tested, no more sensor values can be read. For such a case, it is provided that the sensor values are determined by means of a mathematical observer. Such a mathematical observer may, for example, be a Lünberger observer or a Kalman filter.

A mathematical observer is a calculation rule that has continuously a number of the last values of a respective variable to be observed as input variables, from which an approximation or estimation of the real state of the quantity or distance to be observed is calculated. The observed variable may, for example, be a state variable of an actuator for providing a tie rod force.

In a further possible embodiment of the presented method it is provided that the at least one state variable comprises a state variable for a control difference between a measured or determined by a mathematical observer actual value and a predetermined setpoint.

In order to map a current state of an actuator to be controlled, it is usually necessary that a deviation of the current state is taken into account by a target value in the control, as is typical for a closed loop.

In a further possible embodiment of the presented method, it is provided that variables of the state space model determined in the state feedback module are at least partially returned within the state feedback module and used as input variables of the state feedback module.

By a repeated return of values determined by the state feedback module provided according to the invention to the state feedback module, a behavior of the device can be determined by means of the State feedback module detected signal or a signal output by the inventively provided controller signal, for example, be adjusted depending on currently changing state variables. This means that the controller provided according to the invention reacts by the state feedback module to its own control steps.

A possible embodiment of the presented state space model corresponds to the equation system (1) shown below. $$\begin{array}{c}\left[\begin{array}{c}{\dot{z}}_1\\ {\dot{z}}_2\\ {\dot{z}}_5\end{array}\right]=\underset{{\mathbf{A}}_z}{\underset{\}}{\left(\begin{array}{ccc}0& 1& 0\\ \frac{c_1\cdot P^2}{{\left(2\pi \right)}^2\cdot {\eta }_1\cdot J_{Ges}}& -\frac{d_1\cdot P^2}{{\left(2\pi \right)}^2\cdot {\eta }_1\cdot J_{Ges}}-\frac{d_R}{J_{Ges}}& 0\\ \frac{K_{Filter}}{T_{Filter}}\cdot \frac{c_1\cdot P}{2\pi }& \frac{K_{Filter}}{T_{Filter}}\cdot \frac{d_1\cdot P}{2\pi }& -\frac{1}{T_{Filter}}\end{array}\right)}}\left[\begin{array}{c}{z}_1\\ z_2\\ z_5\end{array}\right]\\ \begin{array}{l}+\underset{{\overrightarrow{b}}_z}{\underset{\}}{\left[\begin{array}{c}0\\ \frac{P}{2\pi \cdot {\eta }_1\cdot J_{Ges}}\\ 0\end{array}\right]}}\cdot F_{sOll}\left(t\right)+\underset{{\overrightarrow{b}}_{Stör}}{\underset{\}}{\left[\begin{array}{c}0\\ 0\\ \frac{K_{Filter}}{T_{Filter}}\end{array}\right]}}\cdot u_{Stör}\left(t\right)\\ F_{eMC}\left(t\right)\underset{{\overrightarrow{c}}_z^T}{\underset{\}}{\left[\begin{array}{ccc}0& 0& -2000\end{array}\right]}}\left[\begin{array}{c}{z}_1\\ z_2\\ z_5\end{array}\right]\end{array}\end{array}$$

des Störfaktors der Sollkraft, ein Vektor ${\overrightarrow{b}}_{Stör}\text{‶}$

des Störfaktos der Störgröße und einem Ausgangsvektor ${\overrightarrow{c}}_z^T\text{‶}$

bezogen auf die Kraft an einem elektromechanischem ZylinderThe following applies: "J _GES " corresponds to a total moment of inertia, "P" to a spindle pitch, _" T _Filter " to a time constant of an input _filter , "K _Filter " to a proportionality constant of an input _filter , "c ₁ " to a spring stiffness, "d ₁ " to an attenuation constant, " d _R "of a friction constant," η ₁ "an efficiency in the conversion of rotational into translational motion," F _Soll (t) "of a desired force at time t," u _Sturge (t) "of a disturbance at time t," F _EMC (t) "of a force on an electromechanical cylinder at time t, and z ₁ , z ₂ correspond to respective state variables considered. z ₅ represents a state variable of a low-pass filter for filtering an output signal of a measuring amplifier of the test bench. In the equation system (1) symbolic assignments are made, which are then used in equation system (2): a data matrix "A _z ", a vector ${\overrightarrow{b}}_z\text{"}$

the disturbance factor of the desired force, a vector ${\overrightarrow{b}}_{Stör}\text{"}$

the disturbance factor of the disturbance variable and an output vector ${\overrightarrow{c}}_z^T\text{"}$

based on the force on an electromechanical cylinder

The state space model represents the physical behavior of the controlled system or the actuator of the test stand. For this purpose, various physical influences on the actuator, which can be described by means of linear equations, such as influences of friction, elasticity and damping, inertia, and manipulated variable delay and the Influence of a respective steering system to be tested, taking into account the state space model. It is particularly important to ensure that it can be provided that provided by an engine of the test bench drive energy is converted by a rotational movement in a translational movement.

In an embodiment, it is provided that the state space model is based on the approach that one on a respective actuator for transferring a rotational movement into a translational movement provided force "F _EMC " corresponds to the force that is to be provided by a respective tested steering system for a respective driving situation to be simulated.

$$u\left(t\right)=\underset{-{\overrightarrow{K}}_e}{\underbrace{\left[\begin{array}{cc}\underset{-{\overrightarrow{K}}_1^T}{\underbrace{\left(-{\overrightarrow{K}}_z^T-K_P\cdot {\overrightarrow{c}}_z^T\right)}}& \underset{-K_2}{\underbrace{K_I}}\end{array}\right]}}\left[\begin{array}{c}{z}_1\left(t\right)\\ z_2\left(t\right)\\ z_5\left(t\right)\\ e\left(t\right)\end{array}\right]+K_P\cdot w\left(t\right)$$

The inventively provided controller is characterized in particular by the fact that it has a control module which corresponds to a PI controller and comprises a state feedback module. In this case, a state vector can be extended by a state variable for a control difference between a setpoint and an actual value. In this case, a manipulated variable of the controller is composed of components of the state feedback module and of the control module, so that the following mathematical relationship applies to the state space model and the state feedback module: $$\begin{array}{l}\left[\begin{array}{c}\dot{\overrightarrow{z}}\left(t\right)\\ \dot{e}\left(t\right)\end{array}\right]=\underset{{\mathbf{A}}_e}{\underset{\}}{\left(\begin{array}{cc}{\mathbf{A}}_z& \overrightarrow{O}\\ -{\overrightarrow{c}}_z^T& 0\end{array}\right)}}\underset{{\overrightarrow{z}}_e\left(t\right)}{\underset{\}}{\left[\begin{array}{c}\overrightarrow{z}\left(t\right)\\ e\left(t\right)\end{array}\right]}}+\underset{{\overrightarrow{b}}_e}{\underset{\}}{\left[\begin{array}{c}{\overrightarrow{b}}_z\\ 0\end{array}\right]}}\cdot u\left(t\right)+\left[\begin{array}{c}\overrightarrow{O}\\ 1\end{array}\right]w\left(t\right)+\left[\begin{array}{c}{\overrightarrow{b}}_{Stör}\\ 0\end{array}\right]\cdot u_{Stör}\left(t\right)\\ y\left(t\right)=\underset{{\overrightarrow{c}}_e^T}{\underset{\}}{\left[{\overrightarrow{c}}_z^{T}0\right]}}\left[\begin{array}{c}\overrightarrow{z}\left(t\right)\\ e\left(t\right)\end{array}\right]\end{array}$$

$$u\left(t\right)=\underset{-{\overrightarrow{K}}_e}{\underset{\}}{\left[\begin{array}{cc}\underset{-{\overrightarrow{K}}_1^T}{\underset{\}}{\left(-{\overrightarrow{K}}_z^T-K_P\cdot {\overrightarrow{c}}_z^T\right)}}& \underset{-K_2}{\underset{\}}{K_I}}\end{array}\right]}}\left[\begin{array}{c}{z}_1\left(t\right)\\ z_2\left(t\right)\\ z_5\left(t\right)\\ e\left(t\right)\end{array}\right]+K_P\cdot w\left(t\right)$$

entspricht einem Zustandsvektor, „e(t)“ einer Zustandsvariable für die Regeldifferenz, „A_e“ einer Dynamikmatrix, ${\overrightarrow{K}}_e^T\text{″}$

einem Rückführungsvektor, „A_z“ einer Datenmatrix, ${\overrightarrow{b}}_{Stör}\text{″}$

einem Störvektor, „y(t)“ einem Ausgangswert zum Zeitpunkt t, „u(t)“ einer Stellgröße zum Zeitpunkt t, „w(t)“ entspricht einer neuen Eingangsvariablen zum Zeitpunkt t. „K_P“ entspricht einer Konstanten eines Proportionalteils des Regelmoduls und „K_I“ einer Konstanten eines Integralteils des Regelmoduls. ${\overrightarrow{K}}_z^T\text{″}$

entspricht jeweiligen Parametern einer Zustandsrückführung und ${\overrightarrow{c}}_e^T\text{″}$

einem transponierten Ausgangsvektor eines Beobachters. ${\overrightarrow{K}}_1^T\text{″}$

steht für alle Einträge im Rückführungsvektor ${\overrightarrow{K}}_e^T\text{″}\text{,}$

außer dem letzten Eintrag, der als K₂ bezeichnet ist.Where: $\overrightarrow{z}\left(t\right)\text{"}$

corresponds to a state vector, "e (t)" of a state variable for the control difference, "A _e " of a dynamic matrix, ${\overrightarrow{K}}_e^T\text{"}$

a feedback vector, "A _z " of a data matrix, ${\overrightarrow{b}}_{Stör}\text{"}$

an interference vector, "y (t)" an output value at time t, "u (t)" a manipulated variable at time t, "w (t)" corresponds to a new input variable at time t. "K _P " corresponds to a constant of a proportional part of the control module and "K _I " to a constant of an integral part of the control module. ${\overrightarrow{K}}_z^T\text{"}$

corresponds to respective parameters of a state feedback and ${\overrightarrow{c}}_e^T\text{"}$

a transposed output vector of an observer. ${\overrightarrow{K}}_1^T\text{"}$

stands for all entries in the feedback vector ${\overrightarrow{K}}_e^T\text{"}\text{.}$

except the last entry, called K ₂ .

kann über den mathematischen Zusammenhang gemäß Formel (4) hergeleitet werden. $$u\left(t\right)=-{\overrightarrow{K}}_e^T\cdot \overrightarrow{z}\left(t\right)=-\left[{\overrightarrow{K}}_1^T{K}_2\right]\cdot \overrightarrow{z}\left(t\right)$$

The feedback vector ${\overrightarrow{K}}_e^T\text{"}$

can be derived from the mathematical relationship according to formula (4). $$u\left(t\right)=-{\overrightarrow{K}}_e^T\cdot \overrightarrow{z}\left(t\right)=-\left[{\overrightarrow{K}}_1^T{K}_2\right]\cdot \overrightarrow{z}\left(t\right)$$

$$-K_1=K_2$$

The constants "K _P " and "K _I " can be derived by the mathematical relationship according to formulas (5) and (6). $${\overrightarrow{K}}_z^T+K_P\cdot {\overrightarrow{c}}_z^T={\overrightarrow{K}}_1$$

$$-K_1=K_2$$

jeweilige Eigenwerte eines geschlossenen Regelkreises zum Berechnen eines Sollwerts durch ein Zustandsrückführungsmodul vorzugeben und somit ein Verhalten des Zustandsrückführungsmoduls bzw. des Sollwerts zu beeinflussen.By substituting equation (3) into equation (2), it is possible to use the feedback vector ${\overrightarrow{K}}_e^T.$

To specify respective eigenvalues of a closed loop for calculating a setpoint by a state feedback module and thus to influence a behavior of the state feedback module or the setpoint.

Furthermore, the present invention relates to a test bench for testing a steering system of a vehicle, having at least one actuator for moving the steering system and a control unit, wherein the control unit is configured to generate a mathematical state space model of the test bench, and the at least one actuator by means of a controller at least one control module having a proportional component and an integral component and a state feedback module, wherein the control device is further configured to transmit to the control module as at least one input variable at least one state variable of the test bench determined by the state feedback module on the basis of the state space model.

The presented test stand is used in particular for carrying out the presented method.

Further advantages and embodiments of the invention will become apparent from the description and the accompanying drawings.

It is understood that the features mentioned above and those yet to be explained below can be used not only in the particular combination indicated, but also in other combinations or in isolation, without departing from the scope of the present invention.

list of figures

• 1 shows a physical closed loop model of a possible embodiment of the test stand according to the invention.
• 2 shows a schematic representation of a possible embodiment of a regulator provided according to the invention underlying control loop.

In 1 is a physical control loop model of a test bench 180 for checking a steering system 130 shown. An effect of the test bench 180 on the steering system 130 can through the equation system ( 1 ) are described mathematically. A control unit 132 of the steering system 130 regulates the steering system 130 by means of a steering loop 100 , The steering system 130 includes one between bodies 138 arranged electric motor 134 , by means of which a force on a tie rod 142 is applied, so that the tie rod 142 moved, as indicated by a double arrow 141. By a movement of the tie rod 142 a steering movement of a vehicle is simulated.

Via a CAN bus 156 be from the controller 132 used to control the engine 134 settings or values to a CAN card 126 ie a communication interface of the test bench 180 transmitted. To the test bench 180 or an actuator in the form of a servomotor 104 the test bench 180 to control, determines a real-time computer 114 by means of the CAN card 126 received values a reference variable, indicated by arrow 118 a control loop 190 for controlling the electric servomotor 104 the test bench 180 , The real-time computer calculates this 114 a vehicle dynamics model, as for example by equation system ( 1 ) and a controller with a predetermined increment of one millisecond, as a fixed to the real-time computer 114 given size, indicated by reference numerals 116 ,

By means of the controller determines the real-time computer 114 the command, as indicated by arrow 118 is indicated by the control unit 132 transmitted values are compared in a control module with a proportional component and an integration component with a predetermined setpoint. In this case, the from the control unit 132 transmitted values, if necessary, in a state feedback module of the controller, the one inner loop of the controller forms, be repeatedly returned in a loop. In the state feedback module, those of the controller 132 transmitted values are compared with a feedback vector, which in particular comprises predetermined eigenvalues. These can be from the control unit 132 transmitted values, for example, be multiplied by the feedback vector. By balancing the of the control unit 132 transmitted values with the feedback vector can be acted upon behavior of the controller, in particular with respect to a transient response of the controller or a determined by the controller desired signal. Respective values output from the state feedback module may be supplied to the regulator module of the regulator.

By means of an ethernet card 120 will be the one from the real-time calculator 114 Reference value calculated by the controller 118 to a system 128 for generating a simulated tie rod force acting on the steering system 130 acts, transmits. The system 128 includes a frequency converter or a torque controller 102 , a servomotor 104 and an electromechanical cylinder 108 , by means of which a rotational movement of the servomotor 104 is converted into a translational movement. Furthermore, the system includes 128 a force transducer 110 for detecting a by the servomotor 104 or the electromechanical cylinder 108 provided strength.

The reference variable, as indicated by arrow 118 is indicated by means of the Ethernet card to the frequency converter or the torque controller 102 transmitted over a TCP / IP connection, as indicated by arrow 158 indicated. Furthermore, one by the real-time computer 114 determined manipulated variable 160 via a digital-to-analog converter 122 and a connection 146 to the frequency converter or torque controller 102 transfer. The manipulated variable 160 fluctuates between, for example, a voltage of +5 volts and -5 volts, where the manipulated variable 160 on the connection 146 with an electromagnetic interference signal 150 is charged.

Based on the manipulated variable 160 and the command, as indicated by arrow 118 is indicated, the frequency converter or torque controller controls 102 the servomotor 104 , which in turn generates an engine torque, that over a wave 152 to a clutch 106 and finally to the electromechanical cylinder 108 that on structures 136 is arranged, is transmitted. The electromechanical cylinder 108 sets this from the servomotor 104 provided engine torque in a translational or linear movement of a cylinder rod 140 like a double-headed arrow 154 indicated. Both through the electromechanical cylinder 108 provided force as well as one by the electric motor 134 of the steering system 130 Power provided by a force transducer 110 and to a measuring amplifier or transmitter 112 reported. The measuring amplifier or transmitter 112 generates one of which by the force transducer 110 reported signal dependent signal, which is a controlled variable 148 which may vary in a range of +10 volts to -10 volts. The controlled variable 148 is over a line 144 to an analog-to-digital converter 124 and from this to the real-time computer 114 transfer. In this case, the controlled variable 148 during transmission to the analog-to-digital converter 124 with an electromagnetic interference signal 150 applied.

In 1 set the according to arrow 162 dotted represent connections digital signals. On the other hand, according to arrow 164 continuous drawn connections represent analog signals.

In 2 is a regulator 280 for controlling an actuator of a test stand. Starting from a means of a sensor, such as by means of the force transducer 110 out 1 detected tie rod target force, the signal 270 first in one step 272 inverted. The step 272 is necessary because a negative control signal provides a positive force signal, and in the following a direction of the forces to be compared is to be considered mathematically. The signal 270 corresponds to, for example, the input variable w (t) from equation (3) and can be used according to equation (3) for calculating a manipulated variable u (t).

For determining the manipulated variable u (t) by the controller 280 becomes the signal 270 after inversion with a through the regulator 280 determined setpoint 282 matched in an Additor 283. The desired value can be determined as a function of an output value y (t) to be determined in accordance with equation (2), for example a traction rod force according to equation (2). Furthermore, a result 274 of matching, denoted e * (t), to a PI prefilter 254 transmit, as by arrow 214 indicated. A signal filtered accordingly by the PI prefilter 216 gets to another additor 284 transferred as by arrow 276 indicated and there with a detected from a state feedback module 258 signal 222 adjusted. The result 278 of the comparison ultimately corresponds to the manipulated variable u (t), as can be determined according to equation (3). The manipulated variable u (t) determined by the comparison or the corresponding processed signal 278 becomes the first state variable 244 to a Lünberger observer 262 transmitted. Furthermore, the result 278 of the comparison according to arrow 218 to a digital-to-analog converter 256 transmitted, the one the result 278 corresponding signal in a voltage 220 converted and these according to arrow 202 to a route model 250 transfers. Based on the tension 220 is determined by means of the route model 250 a tie rod force 204 determined according to arrow 210 to a measuring amplifier 252 is transmitted. Based on the tie rod force 204 can the setpoint 282 be determined.

Furthermore, by means of the route model 250 starting from the tension 220 a second state variable 206 , which, for example, according to the equation system ( 1 ) indicates an angular velocity, and a third state variable 208 which indicates, for example, a track rod path for setting a respective steering angle.

The measuring amplifier amplifies this from the system model 250 Tie rod force received signal 204 and transmits a correspondingly amplified signal 212 to an average filter 264 , At the same time the signal becomes 212 with an electromagnetic interference signal 268 and / or other sources of interference 266 charged as by an additor 248 indicated.

One from the mean value filter 264 based on the signal 212 determined mean value signal 246 is connected to an analog-to-digital converter 260 transmitted as indicated by arrow 234. The analog-to-digital converter 260 converts the mean value signal 246 into a digital signal and generates, conditionally, a the setpoint 282 corresponding signal. Furthermore, the analog-to-digital converter transmits 260 a voltage corresponding to the mean value signal 246 as the fourth state variable 232 , such as in the equation system ( 1 ) is provided as a signal of a low-pass filter for filtering an output signal of a measuring amplifier of the test bench z ₅ , to the state feedback module 258 as by arrow 224 indicated.

That the setpoint 282 appropriate signal is in addition to the Lünberger observer 262 transfer. It will be out of the setpoint 282 corresponding signal a currently generated by an electric motor of a steering system to be tested determined force, as indicated by arrow 242 indicated. Furthermore, from the setpoint 282 corresponding signal determines a force currently generated by a servomotor of the test stand, as indicated by arrow 240 indicated.

Based on the force generated by the electric motor of the steering system being tested and the force generated by the servomotor of the test bench in conjunction with the first state quantity 244 determined the Lünberger observer 262 a fifth state variable 236 and a sixth state quantity 238 , In this case, the first state variable, for example, as z ₁ , according to equation system ( 1 ) indicates a steering angle and the sixth state variable, which, for example, as z ₂ according to equation system ( 1 ) indicates an angular velocity used for adjusting the steering angle to the state feedback module 258 be transmitted as by arrows 226 and 228 indicated. Starting from the fourth state variable 232 , the fifth state quantity 236 and the sixth state quantity 238 The state feedback module determines that to be matched with the prefiltered signal 216 provided signal 222 , This means that the prefiltered signal 216 the signal detected by the sensor 270 is matched with a signal based on the state variables 232 . 236 and 238 was determined. This will be the setpoint 282 corresponding signal for determining the state variables 232 . 236 and 238 possibly repeated to the Lünberger observer 262 and according to the state feedback module 258 supplied, so that the set value 282 corresponding signal by the state feedback module 258 in response to current values of the state variables 232 . 236 and 238 can be adjusted. This can be the setpoint 282 corresponding signal, for example, with predetermined eigenvalues, the behavior of the state feedback module 258 determine, be matched.

Through the state feedback module 258 or by respective ones of the state feedback module for changing the setpoint 282 corresponding eigenvalues used, a change behavior of the setpoint 282 corresponding signal and, as a result, a transient response of the controller 280 to be influenced. For this purpose, the eigenvalues may, for example, be specified in a feedback vector by a user or a technician.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• EP 1435082 B1 [0006]
• DE 10922037 A1 [0007]
• DE 19806755 A1 [0008]
• DE 19730851 B4 [0009]

Claims (11)

A method for controlling a test bench (180) for testing a steering system (130) of a vehicle in which a mathematical state space model of the test bench is generated, and at least one actuator (104) for generating a required force for simulating steering movements by means of a regulator, the at least one control module having a proportional component and an integral component and a state feedback module (258), wherein at least one state variable (232, 236, 238) of the test bench, which is determined by the state feedback module (258) on the basis of the state space model, is supplied to the control module as at least one input variable , Method according to Claim 1 in that values determined by the state feedback module (258) are returned to the at least one state variable of the test bench within the state feedback module and processed by means of a feedback vector having predetermined eigenvalues, and wherein the values processed by the feedback vector are transmitted to the control module and to control the at least one actuator (104). Method according to Claim 1 or 2 wherein the at least one state quantity of the following list of state variables is selected as the at least one state quantity (232, 236, 238): a positioner angle of the actuator, an angular velocity of the actuator, an output of a test bench amplifier, a steering rack track and a tie rod speed of the steering system. Method according to Claim 3 in which a signal amplification of a measuring amplifier used for measuring the at least one state variable is selected as a function of a measuring range required for representing a force applied to the at least one actuator (104). Method according to one of the preceding claims, in which the at least one state variable (232, 236, 238) is determined by means of at least one sensor (110). Method according to one of Claims 1 to 4 in which the at least one state variable (232, 236, 238) is determined by means of a mathematical observer (262). Method according to one of the preceding claims, wherein the at least one state variable (232, 236, 238) comprises a state variable for a control difference between a measured or determined by a mathematical observer actual value and a predetermined setpoint. Method according to Claim 6 or 7 , wherein as a mathematical observer (262) a Lünberger observer or a Kalman filter is selected. The method of any one of the preceding claims, wherein variables of the state space model determined in the state feedback module (258) are at least partially returned within the state feedback module (258) and used as inputs to the state feedback module (258). The method of any one of the preceding claims, wherein the at least one actuator (104) is used to provide a tie rod force. A test bench for testing a vehicle steering system (130), comprising at least one actuator (104) for moving the steering system (130) and a controller, the controller configured to generate a mathematical state space model (300) of the test bench (180), and controlling the at least one actuator (104) by means of a controller having at least one control module with a proportional component and an integral component and a state feedback module (258), wherein the controller is further configured to at least one of the control module as at least one input variable by means of State feedback module (258) on the basis of the state space model determined state quantity (232, 236, 238) of the test bench (180) to transmit.

Images