REAL TIME IDENTIFICATION OF AUTOMOTIVE DRY CLUTCH FRICTIONAL CHARACTERISTICS USING TRUST REGION METHODS

Real-time knowledge of the variable characteristics in dry-clutch subsystems is crucial for understanding the dynamic behaviour of an automotive dry-clutch system. The automotive dry-clutch system plays a key role in passenger car gearboxes based on automated manual transmissions as well as in broader class of vehicle driveline systems. Thus, in order to have a good performance of vehicle longitudinal dynamics, whose response is strongly related to the effectiveness of the control strategy, real-time estimation of the dry-clutch torque transmission characteristics is required.This paper presents an original development of an on-line torque transmission model characteristics estimation methodology along with ability of vehicle powertrain dynamics consideration. The method focused to provide the direct capability to observe the kiss-point value by using measurable signals in real-time. Real-time simulations in various conditions demonstrate the effectiveness of the algorithm. Very promising results are expected in different scenarios.