Research

• Model-based mechatronic system design in order to increase the dynamics, the accuracy, and the overall equipment efficiency
• Development of control strategies which systematically account for the nonlinear characteristics of sensor and actuator systems
• Modeling and control of resonance structures
• Design of sensorless actuator systems using suitable observer strategies and optimizing the construction of the actuator system
• Application of smart materials in mechatronic actuator systems

Applications

• Electrohydraulic actuator systems

Electrohydraulic actuator systems are capable of meeting both the increasing demands on high forces or torques at large displacement velocities while comprising a very compact installation space at the same time. On the other hand, electrohydraulic actuators exhibit a significantly nonlinear behavior and in general possess a low energetic efficiency. Therefore, the scope of this research area is the development of modern nonlinear control strategies for electrohydraulic actuator systems. Furthermore, constructive changes for the improvement of the overall energetic efficiency are developed.

Example: Control of a self-supplied, variable axial piston pump:



In this project a nonlinear control strategy for the volume flow and the load pressure was developed in order to realize a desired dynamical behavior of the closed-loop system independently of the actual (fast changing) load, which was assumed to be unknown. For this purpose, the control strategy was extended by a nonlinear load observer. One specific field of application of this control strategy are injection molding machines, where the mold is closed by a hydraulic actuator. If the mold is closed completely, the load changes rapidly which consitutes a major challenge for the control strategy.

Example: Impedance control of electrohydraulic systems:
Force or position control are the classical control concepts for electrohydraulic actuators. However, in many applications it is desired to control the dynamical relationship between force and position in such a way that the closed-loop response is equal to a desired (mechanical) impedance system. This control task, which is well known from robotics, was extended to electrohydraulic systems. Thereby, a main challenge is the inherent nonlinear characteristics of electrohydraulic systems. Furthermore, constructional changes were considered in order to increase the energetic efficiency of electrohydraulic impedance systems. Possible fields of application are active suspension systems, construction machines (excavators) or electrohydraulically actuated robots.

Example: Closed-Center electrohydraulic power steering [Kemmetmüller et al., IEEE/ASME Mechatronics, 12(1), p. 85-97, 2007]:



In recent years, increasing demands on driving comfort and stability require the development of new power steering systems. In this project, a closed-center electrohydraulic power steering system was developed in cooperation with industrial partners, which combines the advantages of a classical hydraulic power steering system with respect to the energy supply and the steering feel with the advantages of an electromechanical power steering system, i.e. in particular the energetic efficiency. Thereby, it is shown that the desired steering feel can be achieved by means of an impedance control strategy. The control strategy was tested on a steering test stand and implemented in a test vehicle.

• Modeling and control of micro-electromechanical sensors

MEMS gyroscopes are widely used in automotive industry as yaw sensors for vehicle dynamics control systems. Other fields of application are found in consumer products such as digital cameras (picture stabilization), navigation systems or interactive video game consoles. The demands on gyroscopes, in particular for automotive applications, are extraordinary high, e.g., in terms of cost efficiency, accuracy and robustness. Hence, a number of control engineering tasks result from these general demands.
The systematic derivation of analytical mathematical models for specific classes of MEMS gyroscopes (capacitive or piezoelectric) is a main focus of our research activities. The sensors under consideration typically are harmonically excited, weakly damped resonance structures. In view of these characteristic properties of MEMS gyroscopes, special emphasis is laid on the derivation of comprehensive models that solely capture the essential dynamics of these resonance structures. The models in turn serve as a basis for the development of linear and non-linear control strategies, e.g., for the stabilization of the primary oscillation. Moreover, models comprising an individually adapted level of detail can be used in the synthesis of the overall sensor design for efficient time-transient as well as steady-state simulations.

Funding: Eurimus-Project EM 103 "RESTLES - Reliable systems level integration of stacked chips on MEMS"

• Electrorheological actuators

The topic of this field of research is the modeling, control and application of electrorheological fluids in actuators. Basically, an electrorheological fluid significantly and reversibly changes its apparent viscosity upon application of an electric field. This effect is based on the formation of chains of the polyurethane particles which are suspended in silicone oil.

Example: Semi-active suspension for a heavy-duty offroad vehicle:



In this project, a new semi-active suspension system based on electrorheological dampers was developed and tested for a heavy-duty offroad vehicle. Thereby, the special characteristics of electrorheological fluids necessitated the development of new control strategies for the dampers. In order to reach a compromise between driving comfort and driving stability a hybrid ground- and sky-hook strategy was developed. The electrorheological dampers and the control strategy were implemented and tested in a test vehicle. The project was accomplished in cooperation with FLUDICON and WTD41 of the German Armed Forces (press coverage).

• Sensorless control of electromechanical actuators

In some applications the usage of sensors is undesirable due to additionally arising costs and the potential risk of failures. In worst case, the failure of a sensor can lead to the failure of the entire system. In recent years, large effort has been made in the development of sensorless control strategies, wherein the sensors are replaced by virtual sensors based on observer strategies utilizing cross-sensitivities of the system. Within these research activities new sensorless control and observer strategies for electromechanical actuators are developed. Furthermore, questions concerning the adaption of the construction in order to improve the observability are examined.

Example: Sensorless control of a traction drive:



In this project a new velocity observer for an industrial traction drive based on an induction motor was developed. Thereby, the existing control strategy was adapted in order to allow for a sensorless operation. Since the injection of additional signals for the observation of the velocity usually influences the quality of the controller, in this project an observer based strategy was employed. For this, an estimation of the rotor flux is necessary which in general causes problems due to the time integration of the stator voltage equations. In this new adaptive observer strategy this time integration can be omitted which makes the observer feasible. The overall control strategy was tested on different traction drives of tramways and municipal railways.