Associate Professor Matt Campbell in the lab. On screen are screenshots of the software his group has written to produce generative grammars.
Associate Professor Matt Campbell left for Munich, Germany in late May to begin a fellowship at the Technische Universität München (Technical University of Munich) Institute for Advanced Study (TUM-IAS) . His initial stay will last from June 1 through December 12, 2010 with follow-up visits during the next three years, where he will be working with faculty and graduate students in the Advanced Computation area. Campbell will help TUM-IAS outfit research labs and work with Professor Kristina Shea in a new research area for the institute — advanced computation, which he helped define.
Dr. Campbell received the Hans Fischer Fellowship, which is awarded to up to four "outstanding scientists a year who intend to explore innovative, high-risk topics." The Fellowship is named after Professor Hans Fischer (1881-1945) who was awarded the Nobel Prize in Chemistry in 1930. Fellows participate in university events and are expected to contribute to the intellectual life of the university.
Matt Campbell's Biographical Sketch
Dr. Campbell received his Bachelor of Science and his Master of Science in Mechanical Engineering in 1995 and 1997 from Carnegie Mellon University, Pittsburgh. In 2000 he was awarded his Ph.D. degree also at Carnegie Mellon University (dissertation: The A-Design Invention Machine: A Means of Automating and Investigating Conceptual Design). Campbell joined the Manufacturing and Design program of the Mechanical Engineering Department of The University of Texas at Austin as Assistant Professor in that same year and founded the Automated Design Lab. Since 2006 he has been working as an Associate Professor and William J. Murray Jr. Fellow at the Mechanical Engineering Department. He teaches two undergraduate courses, Machine Elements (ME338) and Design Methodology (ME366J), as well as two graduate courses, Design Optimization and Automation (ME392C) and Design as Cultural Production (ME397). Additionally, he serves as the department’s Associate Chair for Information Technology.
Dr. Campbell’s Research in Advanced Computation
Dr. Campbell’s research incorporates math, engineering, language theory (generative grammars), computer science and psychology. He first described his work as "artificial intelligence (AI)." Recently, his area of research has focused on computational design synthesis – a field which he has helped pioneer for the last eight years. In short, the research is focused on computational methods that are solely capable of designing a product. The goal of the research is not necessarily to allow computers to invent new products without human intervention, but rather to augment and bolster human creativity and productivity. As part of this fellowship, he will work with the Institute of Product Development at TUM towards a "fundamental computational theory generative grammar synthesis."
Matt Campbell uses a diagram to explain the linking of sections of code or design elements, somewhat like an object in object-oriented programming, that is used in his generative engineering design grammars.
Understanding Generative Grammars
The study of “generative grammars†originated in the study of the system of rules found in all languages, and was first articulated by M.I.T. Professor Emeritus Noam Chomsky in 1957, in his book Syntactic Structures.
"Chomsky's assumption was that a grammar is finite, but that the sentences which people produce are theoretically infinite in length and number. Thus, a grammar must generate, from finite means, all and only the infinite set of grammatical sentences in a language."
The theory has been used in other areas besides language, including music, art and architecture. Dr. Campbell cited a paper written in 1981 by graduate architecture students H. Koning and J. Eizenberg, called “The language of the prairie: Frank Lloyd Wright’s prairie houses.†In the paper, they meticulously analyzed the core designs of prairie house style in 11 homes Frank Lloyd Wright designed, and were able to define 238 possible variations on this core. They designed three new houses based on the design language found in the 11 homes studied.
Dr. Campbell’s Research Goals
Campbell’s goal is to encode the rules-of-thumb that define innovative design in a series of computational grammar rules, in which the grammar is mathematical, not linguistic . Unlike the equation-based analytics of applied physics, engineering design is difficult to understand and put to use. This is due in part, to the cognitive aspect of design and the concept of creativity, which is difficult to quantify. With generative grammars, we allow the computer the ability to construct complex shapes and configurations. Since the grammar rules are extracted from actual engineered products, the computer is now able to build a wide variety of feasible solutions to a single problem. Dr. Campbell’s current efforts are focused on powertrain systems and satellite design.
Professors Campbell and Shea plan to write a book on their research that will someday make it possible to design engineering systems that are now too complex to design without the use of artificial intelligence. AI frees programmers from having to define an infinite number of scenarios, thus greatly simplifying the design process.
