Vetronics Institute 2^nd Annual Summer Workshop Series             

Links to Abstracts

Abstract

In this tutorial, we address fundamental issues in land, air, and space vehicular power systems. Furthermore, a brief description of the conventional vehicle electrical systems and the role of power electronics will be given.   The future vehicle electrical power systems will most likely be a single voltage bus (42V DC) with provision for hybrid and multi-voltage level distribution. Some of the loads considered are power steering, anti-lock braking, air-conditioning, throttle actuation, ride-height adjustment, rear-wheel steering, electrically heated catalyst, and active suspension systems.

Advanced aircraft and spacecraft power systems are also multi-converter power electronics based systems. In these systems, different converters such as AC/DC rectifiers, DC/DC choppers, and DC/AC inverters are used to provide power at different voltage levels in both DC and AC forms. Some of the loads considered are electro-mechanical and electro-hydraulic flight control actuators, 270V DC switched reluctance starter/generator, electric anti-icing, electro-mechanical valve control, air-conditioning system, utility actuators, weapon systems, and different electric motor drives for pumps and other applications.

This is an in-depth course for power electronics, motor drives, automotive, and aerospace electronics engineers and researchers at all levels.

Advanced Vehicle Systems Research Program

Department of Electrical Engineering

Texas A&M University

College Station, Texas      77843

Phone: 979-845-7582

Email: ehsani@ee.tamu.edu

Return to Titles 

Roadmap to Models for Emulating, Predicting, and Timing Intelligent Behavior

Abstract

A quarter century of research in Artificial Intelligence, Cognitive Science, Human Computer Interaction, and Computer Simulation has resulted in powerful tools for cognitive and performance modeling. An engineer faced with the need to model or automate human tasks is faced with competing and complementary theories of cognition, cognitive architectures, and engineering modeling languages and tools including ACT-R, GOMS, Soar, 3-CAPS, EPIC, GLEAN, IMPRINT. In this workshop, we present the main theories, architectures, and models, and explain how they relate to each other. We conclude with a roadmap to what to use when.

Presenter Info

 Dr. Fatma Mili

Associate Professor

School of Engineering and Computer Science

Oakland University, Rochester, MI48309-4401

mili@oakland.edu, (248)370-2246, www.secs.oakland.edu/~mili

Return to Titles 

Integration of a Processor System Model with Modelica

Abstract

Modelica is a unified object-oriented language for modeling of large, complex, and heterogeneous physical systems. The language can be used for mechatronic models in robotics, automotive, military and aerospace applications involving mechanical, electrical, electronic, hydraulic, and control subsystems. Modelica models are mathematically described by differential, algebraic and discrete equations. Models of standard components are typically available in model libraries. Modelica allows high-level modeling using instantiation of existing components.

Although, Modelica is suitable for the simulation of an embedded control system, a method for simulating a CPU is not currently available for this environment. Integration of a processor system model with a Modelica model is important. It allows simulation of a software based intelligent control system of a mechatronic system on one end. It also gives the capability to study processor behaviors under a complex real-time physical environment on the other end.

Intelligent vehicles and cockpits of intelligent vehicles use extensive embedded processing of data in real-time. Those processors often need to make decisions based on complex real-time events. Uses of these systems are increasing in emergency service vehicles, as well as military and space exploration vehicles.  The automobile industry is also interested in the drive-by-wire and other advanced automobile design concepts and in more intelligent vehicles. Yet, the capability to study the CPU behavior under these workloads is limited. The combination of a processor system with a real-time operating system and workload, and a physical system modeled in Modelica is a powerful test-bench for these types of studies. In the workshop, we explore ways to model a processor system in the Modelica simulation environment. We then explore several applications of the new capabilities.

Presenter Info

Afzal Hossain, Ph.D., Assistant Professor,

University of Michigan - Dearborn

Email: afzal@umich.edu, Phone: (313) 583-6318.

Return to Titles