CAREER: Modularized Silicon-based Neuromorphic Visual Processing Systems Implemented in Analog VLSI

Rochester Institute Of Tech, Rochester NY

Investigators

Abstract

9984386 Titus The project presented in this proposal combines educational and research activities through exploration and development. The research portion of the work will be formed around the field of analog VLSI systems with an emphasis on neuromorphic processing. A unique approach will be taken to implement an integrated artificial visual system that is formed from distinct components that are designed to work together seamlessly. The entire system will be designed to operate as an entire, functional unit, but will be designed as modular pieces with each piece created individually; the complexity of the visual system requires that it be modeled in manageable pieces. The visual system will consist of the low-level processing functions performed by the retina, mid-level stereopsis processing, and highlevel object recognition. The retina function has been implemented successfully, but modifications will he made to previous designs to accommodate the connections to other levels of processing. An analog VLSI stereopsis processor design that has been created will be redesigned to fully take advantage of top-down connections from an object recognition-processing unit. Finally, a unique object recognition system will be designed based on current psychophysical and biological models of object recognition that is compatible with analog VLSI. The final system will be a silicon-based system that performs multiple visual functions and implemented on an analog VLSI platform. The proposed research will provide new insights into the development of artificial visual systems through the concept of modularization. This type of processing-specific analog circuit is compact and has very low power consumption, making it ideally suited for low cost, high volume, and autonomous exploration devices. Also, this work will generate a set of design standards for analog VLSI, which will aid the mainstreaming of this technology. Students working on the research in this project will gain exposure to diverse fields of science and develop new insights into how to model and synthesize complex systems. The teaching portion of the work will be formed around increasing the use of technology in the classrooms, by addressing the needs of a diverse student body through varied teaching methods, by increasing the understanding of integration and system-level IC design, and by addressing the retention and recruitment problems at the undergraduate and graduate levels, respectively. These objectives are closely linked to one common theme: improving education through commitment. By developing new methods or using the variety of existing methods to transfer knowledge to students, students from all walks of life can find electrical engineering to be exciting and fulfilling. This will be accomplished by continuing to experiment with group learning techniques and alternative forms of evaluation and assessment. The study of analog VLSI systems requires an understanding of all levels of a system, from transistor physics to amplifier design to IC-level layout issues to board-level interconnections. Students must also have an understanding of these levels and their relationship in order to be successful designers of the more complex systems of the future. These concepts will be reinforced by highlighting integration and system constraints in the core electronics class, by introducing a new IC layout elective class and examining the introduction of layout tools at the freshman level. The proposed educational plan is closely tied to the research plan by involving undergraduate students in the research projects and by involving research-oriented graduate students in the educational process through guided supervision. The results from the proposed research will be brought to undergraduate students in the form of group projects, case studies for reinforcing core knowledge, and through examples in class. Student portfolios will be explored as means for aiding students in improving their understanding of material through regular assessment by maintaining an up-to-date website intended for elementary and high school students, they will have access to the ongoing research work. Additionally, student-oriented research-based project demonstrations will be developed that will be brought into classrooms to further engage the students as they are formulating their future interests. ***