Biography

The overall goal of my research is to introduce software techniques to accelerate the design of electromechanical systems: changing the way we design, what components we use, and what machines we use to fabricate them with.

Modern electronic designs are increasingly complex systems with many specialized processing engines. The complexity of these systems and the severe physical constraints imposed by technology scaling are limiting the hardware efficiency gains possible using traditional manual design. Automated exploration of much larger design spaces is required to uncover further efficiency gains, but will require a much higher level of design description to avoid restricting the set of possible design solutions while maintaining architect productivity. I am working on a set of techniques for accelerating the entire hardware design cycle.

Despite new computer controlled machines that theoretically speed up fabrication of mechanical systems, 3D design remains difficult and fabrication remains manually intensive. In order to address this, I am exploring the ideas of {\it shape compilers} as well as the application of declarative design methods to the design and fabrication of articulated forms to make design and fabrication dramatically more productive. Shape compilers are software applications which transform a high level design into simple and inexpensive parts ready for rapid manufacturing.

Education

  • 1992, Ph.D., Computer Science, University of Massachusetts, Amherst
  • 1989, M.S., Computer Science, University of Massachusetts, Amherst
  • 1985, B.S., Computer Engineering and Cognitive Science, University of California, San Diego

Selected Publications

  • A. Buchan, J. Bachrach, and R. S. Fearing, "Towards a minimal architecture for a printable, modular, and robust sensing skin," in Intelligent Robots and Systems (IROS), 2012 IEEE/RSJ International Conference on, 2012, pp. 33-38.