Modeling, Estimation, and Control of Automotive Systems

---

Project number: 0120-053

Period: June 2002 -

Summary: The application of various control systems in vehicles results in significant improvements in drivability, safety, comfort, fuel economy, and emissions. The development of automotive control systems is based on the knowledge of mathematical models of automotive elements and systems. Since a control system often requires knowledge of powertrain quantities that cannot be conveniently measured, it is necessary to develop on-line estimators of these quantities.

In known models of automotive elements and systems, some important dynamic effects are often neglected (e.g., dynamics of friction in clutches, servo-system gears, and tires), or the models are too complex from the standpoint of control system development (e.g., finite-element-based models). This project proposes the development of simple, physical, control-oriented, and experimentally validated dynamic models of automotive powertrains. Alternatively, the application of black-box models based on neural networks is also considered. The emphasis is on modeling the internal combustion engine, electronic throttle body, wet clutch of an automatic transmission, and tires. A detailed analysis of the dynamic behavior of these elements and the overall powertrain with automatic transmission is conducted to facilitate model parameterization and control system design.

Estimation of engine torque, vehicle traction force, and other quantities of an automotive powertrain is usually conducted in an open loop. This project proposes the development and experimental verification of closed-loop nonlinear estimators to increase the static and dynamic accuracy of estimation in the presence of significant process parameter variations. Special attention is given to the choice of estimator structure to ensure estimator robustness across a wide operating range.

Based on the developed models and estimators, various high-performance automotive control systems are designed and experimentally tested. The emphasis is on idle speed control of the gasoline engine, active compensation of torque pulses in hybrid vehicles, servo control of the electronic throttle, and vehicle dynamics control systems, including ABS and traction control.