Research Statement

My research activity falls into two broad categories:

• Robotics and Artificial Intelligence: The central research question I have explored in robotics is: “How can robots be made easier to program?” To that end, I study varying approaches to machine learning, subsumptive control, and AI planning. I employ inexpensive robots, so studying approaches that are practical on such platforms is an essential aspect of my robotics research. I am also strongly interested in robotics education.
• Computer Science Education: My educational research goals include teaching students how to write reliable software and building software tools to assist student learning:
  • Pluggable Interrupt OS: The Pluggable Interrupt OS enables students to quickly and easily build bare-metal kernels using the Rust programming language. I employ this tool in my Operating Systems course.
  • Grambler: Grambler is a parser generator that uses the Earley algorithm. I find most parser generators require excessive effort to explain to students. For student projects, clarity and ease of use are more important than parsing efficiency. Building a parser generator on the Earley algorithm enables it to accept all context-free grammars, greatly easing the teaching of parsing.
  • Boundalyzer: Boundalyzer is a software modeling and specification tool I developed when teaching the Black Box concept from Cleanroom Software Engineering. It enables students to write specifications of the boundary between a software artifact and its environment. It will automatically check these specifications for completeness and consistency, reporting any problems to the student.