DE102004058359A1 - Vehicle movement control process, involves matching modeled measured values obtained from reference values with modeled control system based on comparison of modeled values with actual measured values of control system - Google Patents

Abstract

The process involves storing reference values in a control system (4), and obtaining actual measured values of the control system for determining updated reference values. The system is modeled, and modeled measured values are obtained from reference values in a modeled control system (9). The modeled values are compared with the actual values, and matched with the control system (9) based on the comparison of values. An independent claim is also included for a movement control system of a vehicle.

Description

The The invention relates to a method and a system for driving motion control of a vehicle in which setpoints are fed into a controlled system and real measured values of the controlled system for the determination of updated Setpoints are taken.

It is known, for example in an electronic stability program (ESP) vehicle dynamics control system based on Sensor signals setpoints to be determined in actuators, for example in the braking system of the vehicle. Based on these driver-independent Interventions in the vehicle by the actuators change the sensor signals and new setpoints are determined again. The interventions in the vehicle should as possible exactly. This is difficult because vehicle parameters, such as the load state, the center of gravity of the Vehicle, the wear of Axles and wheels, the skew stiffness, etc. change and thus influence the sensor signals. continue to occur errors at the measured quantities measured by the sensors by "bias" and scaling error The problem is that the actuators may be inaccurate are.

From the DE 38 19 474 C1 For example, a method and an apparatus for propulsion control of motor vehicles have become known. The method detects an approach of the vehicle to the lateral acceleration limit value independently of the state of the load, the tire, the roadway, etc., by comparing a measured value for the lateral acceleration, the yaw rate or the difference of the front wheel speeds with a calculated value determined by means of a calculation formula , wherein this calculation value is phase-delayed in a filter and corrected in a first correction unit and wherein the used calculation formula is corrected in a second correction unit.

From the EP 0 962 368 A2 a wheel brake control system for motor vehicles with a regulation of the driving state by means of a mathematical vehicle model has become known. It is provided that in the Radbremsregelsystem with sensors, controllable actuators and an electronic control unit for controlling the wheel speed of a wheel, when an actual driving state of a predetermined target driving state, which is calculated as a function of driving parameters by means of a mathematical vehicle model, deviates, during vehicle operation, an automatic correction of the vehicle model is carried out according to a deviation of the currently specified target driving state from the current actual driving state, if present with the current actual driving state is an ideal driving state in which no deviation that leads to a control should occur. At least one condition is defined by which it is detected whether an actual driving condition identical to the ideal driving condition exists.

task The present invention is a method and a travel motion control system to provide with which the control of the driving movement of a vehicle preferably exactly possible becomes.

These The object is achieved by a method of the type mentioned, in the controlled system is modeled and from the setpoints in the modeled Control path modeled measured values are determined, which with the real measured values, with a deviation the modeled controlled system is adapted and the adaptation descriptive Parameters are taken into account when determining nominal values.

By This measure it becomes possible to have one Travel control of a vehicle, in particular a vehicle dynamics control, the driver independent Interference with the brakes, steering, suspension and powertrain undertakes to optimize. In that the descriptive of the adaptation Parameter can be taken into account when determining setpoints, a Online correction of vehicle parameters, measured variables and actuator inaccuracies respectively. It can also be provided that a sensor signal or several sensor signals are used as measured values that are determined by certain Vehicle parameters, measured variables and Actuator inaccuracies are affected. By comparing such measured values with modeled measured values can be a correction the controlled system, the inaccuracies of the vehicle parameters, Measured variables or Actuators considered. Thus, a correction is possible without to know certain vehicle parameters in detail.

The aforementioned actuators, which are controlled by the setpoint values and which perform driver-independent interventions in the vehicle, can in particular be part of an electrohydraulic brake (EHB), an electrohydraulic steering (EHL) or an active body control (ABC). In an electro-hydraulic steering, there is no mechanical penetration of the steering wheel to the wheels. As a result, driver independent steering intervention by actuation of corresponding actuators are possible. The Active Body Control is a system with which the suspension characteristics of a Vehicle can be influenced. Furthermore, the drive train may have actuators. By controlling these actuators by setpoints, a lateral dynamic and longitudinal dynamic intervention in the vehicle can be made. The actuators may include control devices and actuators, wherein the control units can implement the setpoints in manipulated variables for the actuators.

at a preferred variant of the method can be provided that the setpoints are determined based on a vehicle model and the vehicle model is updated based on the parameters. Thereby, that the vehicle model is constantly updated by parameters, the Setpoints are determined particularly accurately when they are based on the updated Vehicle model can be determined. In particular, an accurate Vehicle dynamics control also take place when minor suspension modifications carried out were. In particular, the same method can be used for different series of a vehicle. Also a good adaptation, for example, when switching from summer to winter tires or when driving with a temporary spare wheel. By updating the vehicle model will wear out vehicle parts automatically considered in the scheme.

at an advantageous variant of the method is provided that the measured values determined in the modeled controlled system in one Estimation method can be estimated. Particularly good estimates can be you get when as an estimation method a Kalman or Least Square filter is provided. In the estimation procedure can be a size, like For example, a characteristic speed, the one Function u.a. of the wheelbase, the vehicle mass and the skew stiffness is, appreciated become. This value can be measured or measured Values determined characteristic speed compared become. Due to the comparison, the model of the controlled system Getting corrected. While a Kalman filter frequently for estimation of states can be used according to the invention, the Kalman filter for the estimation of Measured variables or be used by vehicle parameters. Using a Kalman filter can be a multidimensional size, for example different readings are estimated. Based on measurement equations becomes the Kalman filter while his business to the real conditions. The balance is done by comparing real measurements with values, which are determined using different models. The Kalman filter is therefore supported by a comparison with reality. The consideration the outputs, i. of the modeled measurements when performing the calibration of the Kalman filter leads to, that it is a recursive estimation method.

Especially It is advantageous, if in each case a model for actuators, the setpoints to implement in a vehicle influencing action, for the movement of the vehicle and for the sensors is formed. Through a model of actuators can be inaccuracies in the determination of manipulated variables or of actuators set by manipulated variables become. Through a model of the vehicle effects can be changed Vehicle parameters, such as stiffness or loading of the vehicle become. Through a model of the sensors inaccuracies of the sensors, for example, due to offsets and scaling errors become.

The Task will as well solved by a travel control system of a vehicle, having a Regulator to carry of driving functions setpoints to a controlled system, in particular to one or more actuators of the controlled system, of the vehicle, and receives sensor signals from one or more sensors of the controlled system. The Setpoints and sensor signals are also fed to a parameter determination device, in which determines based on a model of the controlled system model sensor signals and compared with the measured sensor signals, wherein a deviation the model is adjusted and parameters that change describe the model, are transmitted to the controller. With a such travel control system is an online identification and consideration of Inaccuracies in vehicle parameters, measured variables and actuators possible. Farther If a diagnosis, e.g. of chassis tolerances. Such a movement control system can be used almost independently of the vehicle, because of the adaptation of the model of the controlled system by the Parameter adaptation to almost any vehicle is possible. When using the travel motion control system within the same vehicle, it is possible to change to respond to vehicle parameters. In particular, an accurate Vehicle dynamics control take place when the loading condition of the Vehicle changes or when the road surface changes. Also, the wear of components can in the movement control, especially in the vehicle dynamics control, considered become. The vehicle parameters may be dynamic Sizes or physical, stationary Act sizes.

In a preferred embodiment, it can be provided that the parameter determination device is designed as an estimating device with a comparator and a correction element. Preferably, the treasure is to a model-based estimator. In particular, a Kalman filter can be used as estimator. A Kalman filter can be better matched to the real state than other comparable estimation methods.

A particularly simple and space-saving arrangement results when the controller and the parameter determination device are arranged in a control device are. By the local Closeness becomes one quick update of the vehicle model stored in the controller allows. In particular, it is possible the controller or the vehicle model stored in it in real time To update.

Especially it is advantageous if the parameter determination device on Model of the actuators, a model of the vehicle and a model of the sensors having. As a result, the controlled system is mapped. The in the vehicle included actuators are over Controlled the controller, causing an influence on the movement of the vehicle that leads in turn be detected by means of sensors arranged in the vehicle can. If exactly this controlled system in the parameter determination device can be reproduced and updated, can deviations in vehicle parameters and measurands as well Actuator inaccuracies particularly quickly and easily detected and considered in the further regulation become. This is done by specifying parameters or a parameter set which describe an error in the model and so the model-based controller bring up to date. In the model-based controller can in particular information about the stiffness, the dimensions the vehicle etc. stored.

Further Features and advantages of the invention will become apparent from the following Description of an embodiment the invention, with reference to the figure of the drawing, the essential to the invention details shows, and from the claims. The individual features can each individually for one or more in any combination in a variant be realized the invention.

One embodiment The invention will be explained in more detail with reference to a drawing. The only figure shows a highly schematic representation of a motion control system.

In the 1 is a motion control system 1 shown in which a control unit 2 a regulator 3 includes, based on a vehicle model setpoints for a controlled system 4 determined. The setpoints become actuators 5 fed, which leads to an influence on the movement of the vehicle. This is through the block 6 indicated. The movement of the vehicle is arranged with sensors arranged in the vehicle 7 detected. The sensor signals of the sensors 7 turn turn the controller 3 supplied, in which due to the sensor signals again set values are determined. The sensor signals are influenced by vehicle parameters, such as the tire slip stiffness, the vehicle's center of gravity, the load state of the vehicle, the wear of axles and journal bearings, and the like. Furthermore, the sensor signals are influenced by inaccuracies of measured variables, for example the lateral acceleration and the longitudinal acceleration of the vehicle. The measured variables may be inaccurate because of a "bias" or a faulty scaling 5 or the actuators of the actuators 5 affected. In particular, by the sensors 7 determined physical, stationary variables and driving dynamics variables. The setpoints of the controller 3 are also a parameter determining device 8th , which is formed in the embodiment as a Kalman filter supplied. The parameter determination device 8th includes a model 9 the controlled system 4 , In particular, there is a model of actuators 10 , a model of the vehicle 11 and a model of the sensors 12 available. The aforementioned models can be mathematical models. Based on the model 9 the controlled system 4 , the setpoint values of the controller 3 are supplied, sensor signals are modeled as measured values, in particular estimated. The modeled sensor signals are combined with the real sensor signals or the measured values of the sensors 7 in a comparator 13 compared. Due to the comparison in the comparator 13 is in a correction facility 14 a deviation is determined and becomes the model 9 the controlled system 4 according to the detected deviation in the comparator 13 adapted and corrected.

Parameters that change the model 9 the controlled system 4 describe, in turn, the controller 3 according to arrow 15 supplied, so that there stored model of the vehicle can be corrected based on the parameters. The vehicle model of the controller corrected on the basis of the parameters 3 describes the vehicle in which the driving motion control system 1 is used, more precisely, so that also the setpoints for the controlled system 4 can be determined more precisely. The parameters that the controller 3 can be supplied, can be a parameter set. They take into account, for example, when wear occurs on the vehicle or on the tires or if there is a fault in the model. Through the correction element 14 becomes the model 9 to the actual controlled system 4 customized. This adjustment is via the parameters to the controller 3 forwarded. A particularly good measure of the condition of the vehicle results in the slip angle. It can therefore be provided that the swimming angle descriptive sensor signal from the controlled system 4 the comparator 13 is fed and in the model 9 the controlled system 4 A slip angle is estimated, which compares with the real slip angle 13 is compared. In particular, the values for the slip angle can describe a plurality of physical or driving dynamics variables, ie, if values for the slip angle are known, the values of other physical or dynamic parameters can be deduced or calculated or determined, so that the vehicle model in the controller is based on the float angle 3 can be corrected.

At this point, again the essence or the core of the method according to the invention or the device according to the invention is to be represented: In the method according to the invention or in the device according to the invention, for a real controlled system (FIG. 4 ) coming from the vehicle ( 6 ), the actuators arranged in the vehicle ( 5 ) and the sensors also contained in the vehicle ( 7 ), a modeled controlled system ( 9 ). Starting from the setpoints, which are controlled by the controller ( 3 ) are calculated in the modeled controlled system ( 9 ) modeled measured values. These modeled measurements are taken in a comparator ( 13 ) compared with the real measured values. If there is a deviation between the modeled and the real measured values, the modeled measuring sections ( 9 ), which is done by altering, ie adapting, parameters contained or stored or used in the modeled controlled system. The adjusted parameters are in turn assigned to the controller ( 3 ), ie the adjusted parameters are taken into account when determining the setpoints. The controller and thus also the determination of the setpoints is adapted to the current conditions, whereby the quality of the control is increased. By the method according to the invention or the device according to the invention, influences not taken into account in the case of a comparable control can be detected so far which find their way into the control.

Claims (9)

Method for controlling the movement motion of a vehicle, in which setpoint values in a controlled system ( 4 ) and real measured values of the controlled system ( 4 ) for determining updated setpoint values, characterized in that the controlled system ( 4 ) and from the setpoint values in the modeled controlled system ( 9 ) modeled measured values are compared with the real measured values, whereby in the case of a deviation the modeled controlled system ( 9 ) and the parameters describing the adaptation are taken into account when determining nominal values. Method according to claim 1, characterized in that the setpoint values are determined on the basis of a vehicle model and the vehicle model is updated based on the parameters. A method according to claim 1 or 2, characterized in that in the modeled controlled system ( 9 ) are estimated in an estimation procedure. Method according to claim 3, characterized that the estimation method realized by means of a Kalman or Least Square filter. Method according to one of the preceding claims, characterized in that in the modeled controlled system in each case a model ( 10 ) for actuators that implement the setpoints in an action affecting the vehicle, a model ( 11 ) for the movement of the vehicle and a model ( 12 ) is provided for the sensors. Traveling motion control system of a vehicle, in particular for carrying out the method according to one of the preceding claims, with a controller ( 3 ), which for carrying out driving functions setpoints to a controlled system ( 4 ), in particular to one or more actuators ( 5 ) of the controlled system ( 4 ), and sensor signals from one or more sensors ( 7 ) receives the controlled system, characterized in that the setpoint values and sensor signals also a parameter determination device ( 8th ), in which a model ( 9 ) of the controlled system ( 4 ) Model sensor signals are detected and compared with the measured sensor signals, whereby in case of a deviation the model ( 9 ) and parameters that change the model ( 9 ) to the controller ( 3 ). Traveling control system according to claim 6, characterized in that the parameter determining device ( 8th ) as an estimator with a comparator ( 13 ) and a correction term ( 14 ) is trained. Traveling control system according to claim 6 or 7, characterized in that the controller ( 3 ) and the parameter determination device ( 8th ) in a control unit ( 2 ) are arranged. Traveling control system according to one of claims 6 to 8, characterized in that the parameter determining device ( 8th ) a model ( 10 ) of the actuators, a model ( 11 ) of the vehicle and a model ( 12 ) of the sensors.