Search

A sideslip estimator based on extended Kalman filtering that solves the ESC differentiability problem of a banked corner or a fast growing sideslip angle is presented. That means the proposed approach can handle banked corners and varying friction coefficients only with ESC sensor setup nearly without heuristics. To aid a new approach of integrated vehicle dynamics control the algorithm can estimate axle cornering stiffnesses at the front and the rear axle. The estimator is validated with 262 test drives that include driving on a race track at physical limits, drifting with large sideslip angles, spin outs as well as driving on different low friction surfaces and banked corner angles up to 50%. The parameter adaption is validated with test drives and different tires on country roads and on race track.

Two methodologies are presented for modeling the initial in-cylinder gas state prior to the combustion. The first method is based on measured in-cylinder pressure data and allows the cylinder individual calculation of the gas mass and temperature. A physical equation for the online calculation of the polytropic exponent during the compression stroke is derived in order to receive accurate absolute pressure data from piezo based in-cylinder pressure sensors. In order to model the in-cylinder temperature an additional semi physical model is proposed to account for the heat input during the intake stroke. From the first methodology a second semi-physical model for the average in-cylinder gas mass and temperature is derived which will only need cylinder pressure data for the calibration process of the experimental model.

To avoid collisions in longitudinal direction emergency braking is suitable in many situations. However for higher velocites, an evasive maneuver is better because it can be applied later than the braking maneuver if there is enough free space. This contribution investigates methods for the decision-making of a collision avoidance system with regard to automatic braking and swerving. The resulting system tries to use the latest possible point for intervention if the driver does not react. This decision is based on certain time measures and distances. Further trajectory planning and control methods for swerving are described and some simulations are given.

This contribution presents a new methodology for the online identification, which allows the modeling of the stationary and dynamic behavior of nonlinear combustion engines with many input and output variables in short time with an appropriate model structure. The used models are local polynomial model trees. The necessary enhancements and adaptations of the identification algorithm for local polynomial model trees for the online methodology are presented. To enable an online identification, not only the parameters but also the structure of the models has to be adapted to the ongoing measuring procedure at a test bench. This structure adaptation methods use iteratively recorded measured data for the determination of an optimized model partition and regressor selection regarding the quality and complexity of the model. This maximizes the iterative improvement of the mathematical model, leading to a reduced test bench time. The applicability of the developed methodology is shown for the identification of a model from both, an artificial test function as well as a real diesel engine.

Modeling of turbochargers for calculation of virtual sensors for onboard applications can be obtained by thermodynamic approach. This paper focuses on the identification of a thermodynamic turbocharger model based on isentropic efficiencies by taking the heat transfer from turbine to compressor into account. Partly physical, partly experimental model of a GT1749MV turbocharger is parameterized using engine testbench measurement data under consideration of heat transfer. Summarized eight virtual sensors for the air and exhaust path of turbocharged diesel engine are calculated by presented turbocharger model.

A short view is given on the role of automatic control with regard to technological developments in various process categories like manufacturing, power and process industries, transportation and vehicles, biological systems, and ecological/social systems. Some current developments in the components of automatic control systems are then considered, ranging from smart sensors and actuators, to bus-connected decentralized computer control hardware and control software with model-based control functions. Increasingly, computer-assisted modeling, identification methods and special development tools like hardware-in-the-loop simulation systems support the control function development. A further step is the integration of processes and their control such as the conjoining of mechanics and electronics to form mechatronics. Finally, some examples of the present roles of automation in modern technological and scientific developments are given, and some perspectives for the development in the near future are stated. Throughout the article a shell-oriented graphical representation is used to show the process categories considered, the subsystems and components and related examples of the technological and functional developments. The intention is to explain the role of automatic control, in particular to a general community, not necessarily expert in the field.

Faults in the intake and exhaust path of turbocharged common rail Diesel engines can lead to an increase of emissions and performance losses. Application of turbocharger models can help to detect and diagnose more faults as standard fault detection methods. The modeling of the turbocharger for onboard fault diagnosis can be obtained by different models. The differences between an approach based on the isentropic efficiencies and an approach based on Euler's turbo-machinery equation are investigated in this paper. The two models for a GT1749MV turbocharger are parameterized with data from the engine testbed. The comparison is applied by issues of measured model inputs, number of intern parameters, parameterization effort and model accuracy. Both models are compared regarding the application for onboard diagnosis.

Due to more and more stringent emission regulations the optimisation of transient engine operation becomes necessary for future Diesel engine developments. Exemplary for the transient operation a step in acceleration pedal is investigated and optimised with respect to the engine emissions. This gives insights about the emission formation at transient driving and a benchmark for the design of open and closed-loop controls. A model structure to simulate the engine emissions and the engine torque is presented. It consists of a dynamical mean value air path model and a stationary combustion model. Thus the relevant outputs NO x , soot and torque are simulated with respect to the engine actuators. For optimisation the calibrated emission values for steady state are regarded. The actuators of the air path and the crank angle of the main injection are optimised such that the emissions follow a desired trajectory. The results are compared to a non-optimised step in acceleration and to a nonlinear closed loop control of the air path.

Dynamic emission models for Diesel engines are important for future engine development. Emission formation takes place in the cylinder, while the emissions are usually measured downstream the exhaust pipe. Besides operation point dependent gas transportation delays the emission sensors themselves possess a significant dead time and sensor time constant varying for different measured quantities. These sensor individual dynamic measurement characteristics complicate the chronological assignment of cause and effect in emission modelling. Therefore the dynamic characteristics of a NO x , a micro soot sensor and an opacimeter are investigated. Models for the measurement dynamics are derived to separate the measurement dynamics from the engine dynamics. Hence excitation signals which strongly excite the emission formation and only weakly excite the intake and exhaust system are presented. Model results for the measurement dynamics are shown and validated with testbed data.

The number of model-based approaches in modern engine control units (ECUs) increases permanently with the increase of engine complexity. Therefore the efficient storage and the fast computation of these models is a challenging task. Due to their fast computation properties most ECUs utilize 2-dimensional grid map structures. Higher dimensional relations need then to be mapped with nested structures of several 2-dimensional grid maps. The determination of suited structures is a time-consuming task and the number of utilized 2-dimensional grid maps may be large for higher dimensional relations. Therefore in the following a grid map structure, a LOLIMOT structure, a Kernel method and an adaptive polynomial approach are compared with respect to model accuracy, computation time and required memory. Conclusions about the implementation on state of the art ECUs are given. As model output the NO x emissions of a DTH Z19 Opel CR-Diesel engine with VGT-turbocharger and exhaust gas recirculation are regarded. Model structures are presented for a global model approach comprising the engine operation point as input, and for a global-local approach predicting the output by weighting local models.

This paper presents a method to determine the momentary lateral and longitudinal force coefficients by calculating lateral, longitudinal, and vertical forces at the wheels. The longitudinal forces of the wheels are calculated for known accelerationand braking torques. A rigid body model is used to determine the vertical wheel forces. The determination of lateral forces is based on a single-track model. A kinematic Kalman filter, which is independent from any unknown vehicle parameters, is used to estimate the sideslip angle. Using a 3-antenna GPS-receiver the estimation problem can be decoupled into two Kalman filters. The integration of vehicle dynamic sensors and a 3-antenna GPS-receiver enables a high sampling rate and improved accuracy. Based on the state estimation the unknown physical parameters of the wheel force models (such as position of center of gravity, moment of inertia, …) are determined using approximations and least squares parameter estimation.

A semi-physical turbocharger power and heat transfer model built in a CR-Diesel engine is presented. The model is based on physical parameters described by Euler's equation. For the dependency of the turbine power from the VGT-actuator a polynomial approach is chosen. A method for measurement and model parametrisation separating aerodynamic power from heat transfer is given. Further an analytical analysis for choosing ramp times of quasi stationary measurements is included. The model quality and the proportion of heat transfer in the measured temperatures is demonstrated with measured data from the testbed. Finally the parameters of the modelled GT1749MV turbocharger are included in this contribution.

After a discussion of passive and active safety systems for automobiles and accident statistics it follows that a further progress in the reduction of accidents can be especially expected by next-generation driver-assistance systems with a sequence of warnings and active interventions. PRORETA is an Industry–University research project with the goal to develop steps towards accident-free driving. The first project considers two vehicles moving in the same direction. The other vehicle is detected by a fusion of LIDAR and camera data providing the system with relative speeds and distance. If the driver does not react to an obstacle on the own lane, the system automatically triggers an emergency braking and/or swerving to avoid a collision. This includes e.g. a fast and precise evasive trajectory control by automatic steering. The second project is dedicated to vehicles moving in opposite directions performing an overtaking maneuver on rural roads. The own vehicle detects the velocities and distances to the preceding and oncoming vehicle by RADAR and lane markings, etc. with a camera. The measured data of the two sensors undergo a sensor fusion with Kalman filters. The overtaking maneuver is predicted by using the measured data of all three vehicles. If an accident-free overtaking is in danger, warnings are given to the driver and if the driver does not react in time a full braking of the own vehicle is triggered such that the driver can turn back behind the overtaken vehicle. The contribution gives a survey on the developed strategies and some basic calculated features and control systems. Measured data are shown and give an impression of the driving experiments on the runway of an air field.

The further improvement of Diesel engine control and diagnosis requires information on the dynamical soot and NO x formation. Investigations of the air and exhaust path show that process and sensor dynamics play an important role for emission modelling. The simplified combustion process and the emission formation is modelled as a stationary batch process. The air path states and the location of mass fraction burned 50% introduce external dynamics to the stationary model structure. It is shown that the main dynamic influences in emission formation come from the air path states at the engine intake. Additional dynamics result from the gas propagation in the exhaust system and the emission sensors. The presented emission models overcome the measurement dynamics leading to relatively fast dynamic models.

A model structure for stationary emission models is presented. The emissions NO x , soot and opacity are separately modeled. Model inputs are the air mass flow rate, charge air pressure, intake temperature, the location of mass fraction burned 50% and the engine operation point. Local polynomials are trained for each operation point, defined by engine speed and injection quantity. To increase accuracy a selection algorithm of the significant regressors is presented. The local models are composed by means of weighting functions, depending on the engine operation points, to global emission models. For a rasterized operation range, the weighting can be interpreted as a linear interpolation of local models. This enables an easy implementation to common electronic control units (ECUs). Measurements from a CR-Diesel engine show the quality of the models.

The problem of dynamic sensor compensation is considered. A local linear model and a physical sensor model are proposed for estimation of nonlinear sensor time constants, both taking into account the nonlinear dependency of the sensor time constant on gas velocity or mass flow. The parameter estimation is performed using total least squares and orthogonal distance regression to cope with the noise present on the regressors and the output. Furthermore, offset errors on the sensors are automatically compensated. An extended Kalman filter is proposed to reconstruct the sensor input where the noise amplification due to the inverse filtering is reduced by a variable filter parameter. Two sensors having different time constants are necessary for the time constant estimation, whereas just one sensor is required for the dynamic input reconstruction.

The identification of the nonlinear dynamic behavior of combustion engines on test benches is described. A method, that is based on the combination of statistical stationary engine measurement (Design of Experiments - DoE) and dynamic identification techniques for nonlinear multi-input systems is shown. The dynamics are excited by several amplitude-modulated pseudorandom-binary-signals (APRBS), which act on the engine simultaneously, in order to allow a short measurement time period. To show the applicability of the described identification and modelling methods several models for a common-rail diesel engine with exhaust gas recirculation and variable geometry turbocharger (VGT) are shown, with five input variables and the outputs nitrogen-oxide emissions, opacity, air-fuel ratio (Lambda) and exhaust gas temperature.

Using the velocity dependent static gains of the one track model it is shown how special combinations of parameters like cornering stiffness and center of gravity can be estimated. Driving with various speed and a constant cornering radius it is possible to estimate three different gains by measurement of the lateral acceleration, the yaw rate and the vehicle slip angle without knowing the moment of inertia of the vehicle. Results are shown for different measurements of a passenger car.

The contribution describes the results of an Industry-University research project on the fault diagnosis of a modern spark-ignition engine with homogeneous and stratified operating mode by using series production sensors. It is shown how faults in the air system, the rail injection system, the ignition system and exhaust gas recirculation system can be detected with nonlinear parity equations and frequency spectrum analysis methods. Based on several generated symptoms the faults are diagnosed by if-then fuzzy rules. A special emphasis was put on showing in which speed-load operating areas of the engine the individual faults can be detected. All results have been obtained by experiments on an engine test stand.

The presented method for nonlinear system identification is based on the LOLIMOT algorithm introduced by Nelles and Isermann [1996]. The LOLIMOT algorithm divides the input space by a tree-construction algorithm and interpolates the local linear models by local membership functions. Instead of assuming local linear models, the presented algorithm utilizes general local nonlinear functions, which make the algorithm more flexible. These are approximated by a multidimensional Taylor series. Since the amount of regressors grows fast with the number of inputs and the expansion order, a subset selection procedure is introduced. It reveals significant regressors and gives information about the local functional behavior. The local subset selection is implemented as a stepwise regression with replacement of regressors. Mallows’ C p -statistic is used for the subset selection algorithm and is also implemented for final model selection. The benefit of the extended algorithm lies in the higher flexibility in the local models, which results in less partitions of the input space by a similar approximation quality.

As fault detection and fault diagnosis are more and more finding their way into modern industrial mechatronic products, it is now time to take the next step. Based on the research efforts for fault detection and diagnosis, a platform has been prepared for research on fault management, i.e. automatic reactions of the system to continue operation after the detection of faults. These reactions may employ hardware redundancy (i.e. switching from a faulty actuator to another, intact one) or analytical redundancy (i.e. switching from a faulty sensor to a “model sensor” or “soft sensor”). A total fault tolerance concept must encompass all components of a system, i.e. the actuators and drives, the sensors as well as the controller and communication. In many cases, a degradation of functions has to be accepted after a fault has appeared. This contribution will center on actuators and drives showing real-world examples of prototypical realizations and examples from industrial products. Concentrating on the most widespread actuation principles, the paper will focus on electric drives and hydraulic actuators. First, a review is given on fault tolerance principles and general structural considerations e.g. hot-standby and cold-standby, focusing on the scheme of an overall fault-tolerant control system. Then, fault statistics for real actuators and drives will be presented. These fault statistics give hints on the parts of the actuators that are most susceptible to faults. Different designs of fault tolerant actuators and drives, which have been realized, shall be presented and evaluated with respect to their capabilities of sustaining faults. Finally, an outlook for fault-tolerant mechatronic systems will be given.

The nonlinear system identification of block-oriented systems like Hammerstein and Wiener systems is considered. The static nonlinearity is modeled by local affine models whose input space decomposition can be fixed a-priori or can be learnt during optimization. The optimization is done in output error configuration and is solved via the Levenberg-Marquardt algorithm using finite differences. The local affine model used is LOLIMOT which is flexible and robust and allows to model all kinds of smooth and unique nonlinear functions. Furthermore, a regularization operator especially for local affine models is applied which preserves the local interpretability of the submodels during the global optimization which optimizes the parameters of all submodels and the linear dynamic subsystem simultaneously.

Monitoring of the tire pressure is usually performed by direct pressure measurement and wireless transmission to the vehicle ECU's. In order to reduce cost and avoid battery tire sensors, so called indirect tire pressure monitoring is an alternative. First the wheel speeds can be compared to each other. However, this is not sufficient. Therefore, additional measurements are used in this contribution: the vertical wheel speed oscillation and torsional wheel speed oscillation which both depend on the tire pressure. Therefore three methods for pressure loss detection in tires due to punctures (faults) are developed and compared to each other. Finally, an appropriate combination (fusion) of these methods allows to detect small pressure changes in tires. This is shown by experimental test drives with a research car.

Due to the rising complexity of many technical processes modern diagnosis systems have to supervise a multitude of hydraulic, mechanical, electromechanical and mechatronic components. Therefore model-based methods of fault-detection and diagnosis have been developed. These methods use mathematical process models to relate data of several measurable variables. Thus the diagnosis quality depends on the available sensor data. In order to obtain additional information with the given sensor configuration special input excitation signals can be used. This paper will describe a method to locate faults in multivariable systems using such input excitation and its application to the intake air system of a modern common rail Diesel engine. The presented method uses the knowledge of fault effects on the measured output, when the inputs are successively excited quasi-stationary, to determine the location of the fault. It has been applied successfully to differentiate air mass sensor faults from other process faults.

In multi-stand cold rolling mills an accurate synchronization of the mill drives is necessary for a good product quality and for a safe process operation. During the standard commissioning procedure the control performance of each drive is optimized individually. However, in normal operation mode the performance is then not always optimal due to the coupling of the drives by the strip being rolled. Based on a model of the drives and the strip tension force the control performance of the coupled drives is analyzed. A linear observer is proposed to support the standard PI speed controllers. Thus control performance can be improved with a minimum of additional effort. The observer needs only a few parameters and can cover all operating points of a tandem cold rolling mill.

Driver assistance systems need more and more precise information about the vehicle driving situation. While a certain number of sensors is available, some important variables are not measurable. A sensor platform is designed, which consists of a sensor fault detection and diagnosis system, the design of a fault tolerant sensor system and the estimation of non-measurable variables. The features are generated by parity equations, parameter estimation and fault observers and are reduced to a suitable set. After locating the defective sensor the faulty sensor signal is substituted by a calculated one using an Extended Kalman Filter and a two-track model. Additionally, the roll angle is estimated and the road surface is classified to provide these information to the control systems.

In this paper, an Extended Kalman-Bucy-Filter approach for fault detection and diagnosis of an hydraulic servo-axis is presented. The filter improves the results with the EKF as reported in literature since it uses a new model structure, considers nonlinear valve spool displacement-flow relationships and incorporates temperature effects. The performance of the newly proposed architecture is evaluated thoroughly at a testbed in the presence of both process faults and sensor faults. For sensor faults, the smallest reliably detectable sensor fault is determined. The method is finally compared to parity equations and parameter estimation, which have also successfully been applied to the supervision oif hydraulic servo axes.

The optimized control of combustion engines with regard to minimized fuel consumption and emissions requires nonlinear models. Because of an increase of control inputs, like fuel mass flow, injection angle, exhaust gas recirculation flow and several outputs like torque, nitrogen oxides (NOx), hydrocarbons (HC) and particulates the classical grid-based measurement techniques take too long time and do not include dynamics. Therefore different measurement strategies for the stationary and dynamic behavior are described, like Design of Experiments (DoE) and use of suitable neural networks and Pseudo-Random-Binary-Signals (PRBS). As the structure of the models is not precisely known a-priori, nonlinear identification methods in form of special versions of neural networks are good candidates. Therefore, it will be shown how with special amplitude-modulated pseudo random binary signals (APRBS), simultaneous excitation of several input signals, nonlinear multi-input multi-output models can be obtained in relatively short time.

A precise knowledge about the current driving condition is getting increasingly important for future driver assistance systems like global chassis control or collision avoidance systems for avoiding any critical driving situation. Moreover a precise knowledge about the driving situation can be used in testing, in evaluation, and for comparison of new passenger cars. A two degree of freedom model of vehicle lateral dynamics is used to derive a characteristic velocity stability indicator (CVSI). The CVSI is used to distinguish between different driving and stability conditions (i.e. understeering, oversteering, and neutralsteering). This forms the basis for a driving condition detection system with fixed thresholds. It is then extended to a detection system with fuzzy logic thresholds. The CVSI and the fuzzy systems are compared experimentally using (i) a slalom test drive on an icy road and (ii) a stationary circular test drive on a dry asphalt road.

A diagnosis system is designed to detect faults of sensors of lateral vehicle dynamics. Contrary to further publications, an algorithm is introduced which allows a differentiation of the road surface. Therefore, it is possible to adapt the diagnosis system either on dry asphalt or on ice. Thresholds that are used to detect an error can be decreased in order to detect smaller faults.

Applying the continuous-time one track model for lateral vehicle dynamics it is shown how the cornering stiffness of the front and rear tires can be estimated. To consider as much information as available, the standard nonrecursive as well as the recursive least squares estimation methods are based on two equations in parallel governing different dynamics. Its performance is shown by simulations.

A collision avoiding vehicle driver assistance system is developed using braking, steering or a combination of steering and braking. To guide a vehicle on a desired trajectory for emergency steering, two feedback control approaches are presented. The combined steering and braking intervention requires a further control system. The estimation of the sideslip angle is based on a nonlinear vehicle state estimation approach. Simulation results with a validated nonlinear two-track model and results from test drives are shown. Topic: G (Mechatronic systems and applications)

Modern Diesel engines with direct fuel injection and turbo charging have shown a significant progress in fuel consumption, emissions and driveability. Together with exhaust gas recirculation and variable geometry turbochargers they became complicated and complex processes. Therefore, fault detection and diagnosis is not easily done and need to be improved. This contribution shows a systematic development of fault detection and diagnosis methods for two system components of Diesel engines, the intake system and the injection system together with the combustion process. By applying semiphysical dynamic process models, identification with special neural networks, signal models and parity equations residuals are generated. Detectable deflections of these residuals lead to symptoms which are the basis for the detection of several faults. Experiments with a 2.0 l Diesel engine on a dynamic test bench as well as in the vehicle have demonstrated the detection and diagnosis of several implemented faults in real time with reasonable calculation effort.

Many technical processes and products in the area of mechanical and electrical engineering are showing an increasing integration of mechanics with digital electronics and information processing. This integration is between the components (hardware) and the information-driven functions (software), resulting in integrated systems called mechatronic systems. Their development involves finding an optimal balance between the basic mechanical structure, sensor and actuator implementation, automatic information processing and overall control. Simultaneous design of mechanics and electronics, hardware and software and embedded control functions resulting in an integrated component or system are all of major importance. This technical progress has a very large influence on a multitude of products in the area of mechanical, electrical and electronic engineering and changes the design, for example, of conventional electromechanical components, machines, vehicles and precision mechanical devices with increasing intensity. This contribution summarizes ongoing developments for mechatronic systems, shows design approaches and examples of mechatronic products and considers various embedded control functions and system's integrity. One field of ongoing developments, automotive mechatronics, where especially large influences can be seen, is described in more detail by discussing mechatronic suspensions, mechatronic brakes, active steering and roll stabilization systems.

Different methods for change detection of signal features are compared for a residual generated at a hydraulic servo axis. The residual is based on a physical model of the pressure buildup inside the hydraulic cylinder. Nevertheless, it is augmented with an observer to counteract slight model impurities even over prolonged periods of operation. Different sensor faults are introduced, which affect the mean and/or the variance of the residual. Methods have been implemented, which allow to detect changes in the mean and variance of a signal. These methods are compared in terms of the size of the smallest detectable fault, time-to-detection and computational expense. All results have been verified experimentally.

Model based feedback control strategies for electrohydraulic braking and active steering systems in passenger cars are considered. Simulation results with a complex vehicle model and measurements from test drives with a real car are shown for brake control. For a μ-split road surface additional drive dynamics control is required like ESP or active front steering (AFS). Therefore, an AFS feedback controller is applied to stabilize the vehicle. Furthermore, the braking control is improved by AFS, assisting the driver to brake in critical situations.

In this paper, different models of the pressure buildup inside a hydraulic servo-axis are compared. These models are obtained using RBF networks, local linear models and support vector machines (SVMs), with a particular focus on the latter. For SVMs, a reduction method is derived, which allows to reduce the number of support vectors without losing the generalization abilities of the SVM. Experimental results obtained at a hydraulic servo-axis and a comparison of the different modelling techniques conclude this paper.

After discussing various principles of suspensions with variable dampers and springs as well as active components, mathematical models of these systems are derived. It is shown how the unknown parameters can be obtained experimentally through parameter estimation by using body accelerometers, wheel accelerometers, and suspension deflection sensors in different combinations. Experimental results are described for suspensions on a test rig and in cars driving over road surfaces. Through recursive parameter estimation these parameters can be obtained on-line in real time. Then, feedback principles are derived for controlling the damping ratio of dampers with proportional magnetic valve actuators. The mathematical models are then used for fault detection and diagnosis of the damper by combining parameter estimation and parity equation methods.

The further improvement of drive dynamic control needs model-based concepts. Two model-based applications are described. Firstly, the one-track model is used to indicate critical driving situations like oversteering or unstable behavior. Experiments with a middle-class vehicle show the results of this supervision and fault detection method. Secondly, it is shown by simulation how the understeering and oversteering behavior can be influenced by anti-roll bar actuation. Then a nonlinear yaw-rate controller for active steering is designed and it is shown by simulation how the steering behavior can be improved by anti-roll actuation.

For the improvement of reliability, safety and efficiency advanced methods of supervision, fault-detection and fault diagnosis become increasingly important for many technical processes. This holds especially for safety related processes like aircraft, trains, automobiles, power plants and chemical plants. The classical approaches are limit or trend checking of some measurable output variables. Because they do not give a deeper insight and usually do not allow a fault diagnosis, model-based methods of fault-detection were developed by using input and output signals and applying dynamic process models. These methods are based, e.g., on parameter estimation, parity equations or state observers. Also signal model approaches were developed. The goal is to generate several symptoms indicating the difference between nominal and faulty status. Based on different symptoms fault diagnosis procedures follow, determining the fault by applying classification or inference methods. This contribution gives a short introduction into the field and shows some applications for an actuator, a passenger car and a combustion engine.

This paper proposes a control system for active steering of a passenger car based on Local Linear Models. Furthermore, it is investigated how a active suspension system can improve the performance of the active steering system in critical driving situations. Active antiroll bars at the front and rear axle are introduced as active elements in the suspension system. Based on the characteristic velocity of the vehicle a control concept for the antiroll bar stiffness variation at the front and rear axle is proposed. Simulations with a threedimensional vehicle model are carried out to verify the effectiveness of the proposed control system.

A precise knowledge about the current driving condition is getting increasingly important for future driver assistance systems to support the prevention of any critical driving situation. Moreover a precise knowledge about the driving situation can be used in testing and comparing new passenger cars. After a short introduction of a lateral vehicle model, an analytical approach for an online calculation of different driving conditions (i.e. stability, understeering, oversteering, and neutralsteering) is given. A characteristic velocity stability indicator is defined, which allows on-line computation of the present driving condition Real test-drives show the application.

An automobile fault detection and diagnosis method is described for an active vehicle suspension system. The method is based on mathematical models of the suspension system. It is shown how the unknown parameters of these mathematical models can be obtained experimentally by parameter estimation. Furthermore, the semi-physical approach of the local linear neuronal network LOLIMOT is applied for the generation of parity equations. Both, parameter estimation and parity equations are then used for model based fault detection and identification. Various faults are simulated with test rig data and it is shown, how these faults are detected.

Globalization and growing new markets, as well as increasing emission and fuel consumption requirements force the car manufacturers and their suppliers to develop new engine control strategies in shorter time periods. This can mainly be reached by development tools and an integrated hardware and software environment enabling rapid implementation and testing of advanced engine control algorithms. The structure of a Rapid Control Prototyping (RCP) system is explained, which allows fastmeasurement signal evaluation, and rapid prototyping of advanced engine control algorithms. A hardware-in-the-Ioop simulator for Diesel engine control design is illustrated, simulation results for a 40 tons truck are presented. Providing efficient engine models for the proposed development tools, a dynamic local linear neural network approach is explained and appliedfor modelling the NOx emission characteristics of a 1.9 liter direct injection Diesel engine. Furthermore the application of a RCP system is exemplified by the application of combustion pressure based closed-loop ignition timing control for a SI engine. Experimental results are shown for a 1.0 liter SI engine on a dynamic engine test stand.

Monitoring of tyre a pressure in passenger vehicle is a major aspect of improved active vehicle safety. In this contribution, a new velocity compensated method for monitoring tyre pressure using vertical body acceleration signals is presented. It is found that the wheel hop frequency is strongly correlated with damping coefficient and nature of road excitation. Limitations of using the shift of wheel hop frequency are presented. To avoid the influence of nature of road excitation, a virtual transfer function has been derived between front and rear body acceleration signals based on certain predefined event of the vehicle. The delay in transfer function is strongly related to velocity of the vehicle. The characteristic features are generated from the frequency response of the virtual transfer function. The sensitivity analysis presents the effects of the variation of mass of the vehicle and the spring stiffness has no effect on the frequency response but the damping coefficient has an influence. A simple threshold is found from experiment data for classification of relative tyre deflation. The tyre pressure diagnostic algorithm is implemented online and tested for different relative tyre pressure deflations.

After a short introduction into the topic of active vehicle suspension systems, a mathematical model of the considered active vehicle suspension, which is presented in a test rig, is derived. It is shown how the unknown parameters can be obtained experimentally by parameter estimation. The results of parameter estimation are used for model based fault detection and identification, in order to obtain reliable knowledge of the system's state. All results are shown for measurements from an active suspension on a test rig.