Nathan Sturtevant is no stranger to Edmonton; from 2003 to 2010, he served as a post-doctoral researcher and assistant adjunct professor in the University of Alberta’s Department of Computing Science. His research on artificial intelligence and pathfinding—code that determines how to navigate from one location to another—led him to work as an AI contractor with Edmonton-based computer game developer Bioware, where he developed and implemented the pathfinding system for the popular fantasy game Dragon Age: Origins.
After almost eight years as an assistant professor continuing his research at the University of Denver, Sturtevant is returning to UAlberta as a full professor, eager to continue his research at the world-class AI hub and to teach students about the science and processes demanded by computer game development.
What brought you to the University of Alberta?
I worked at the University of Alberta as a postdoctoral fellow and then a researcher after completing my PhD. I was thus familiar with the quality of UAlberta, the student body, and life in Edmonton.
In particular, UAlberta has one of the best groups of artificial intelligence researchers in the world, and has a significant subgroup of people working on AI and games. From a work perspective, given my research, it was easy to choose to return; my family enjoyed being here before so it wasn't a hard choice from that perspective, either.
Tell us about your research program.
My research looks at the application of AI techniques to movement, path planning, problem solving, games, and game design. My research group is looking for efficient algorithms for solving these types of problems and their variants. We’re asking: how we can solve problems faster, with higher quality, or with overall better system performance?
Why is this an important field of study?
Path planning is used in a broad range of applications from video games to robotics, so the work is on fundamental technologies that are used in many applications. We often associate games with entertainment, but underlying many good games is the notion of teaching a player to understand how a virtual world works and leveraging that understanding to solve problems. This is very similar to what we want to achieve in education—teaching abstract concepts and helping our students use those concepts to solve problems in the world. Thus, this work has interesting implications regarding how we engage and teach students and the world.
Tell us about your teaching.
This term I'm teaching CMPUT 250 Computers and Games, an interdisciplinary class on computer games that brings students together from across campus to learn about the game development process and to build a small game. This is the introductory course to the game development certificate. The course includes guest lectures by professors from across campus, as well as by people from the game development industry here in Edmonton. I worked on the development of this course about ten years ago, so it is fun to come back and see how things have changed and grown over the years.
In the winter I'll be teaching a graduate class on single-agent search, which encompasses a sub-area of my research. I've been working on building new web-based interactive teaching materials for the course, so I'm excited to continue to improve those and make them available to others that want to learn about the material.
I enjoy teaching both from the perspective of trying to design and implement the best ways for students to learn, and from the joy in watching students learn, grow, and succeed.