On May 30, 2024, Brian Plancher, assistant professor of computer science, presented new research, titled “Parallel Optimization for Robotics: An Undergraduate Introduction to GPU Parallel Programming and Numerical Optimization Research,” at the IEEE International Symposium on Parallel and Distributed Processing Workshops and Phd Forum (IPDPSW). The paper focuses on a new robotics course Plancher designed at Barnard, COMS BC3159 Parallel Optimization for Robotics. Parallel programming, particularly on graphics processing units (GPUs), and numerical optimization, are increasingly crucial branches of computer science that can be used to tackle complex computational challenges across disciplines. As Plancher explains, however, sometimes daunting barriers to entry restrict the accessibility and diversity of these areas of study. Plancher developed Parallel Optimization for Robotics, an undergraduate course with minimal prerequisites, in an effort to narrow this gap. His curriculum focuses on providing an application-driven introduction to the topics through the lens of robotics paired with a large final project, allowing students to connect theory to practice.
The course was first taught at Barnard in Spring 2023, to 27 students without prior experience in parallel programming or numerical optimization. During this pilot run, Plancher tracked his students’ academic performance as well as how satisfied they were with their progress in learning the subject matter. Students showed improved performance in all key topics, and an overwhelming majority felt that they successfully learned both technical and soft skills that left them feeling prepared for future study in these fields.
Plancher presented his findings at IPDPSW as part of EduPar-24, the 14th NSF/TCPP Workshop on Parallel and Distributed Computing Education. He taught Parallel Optimization for Robotics again in Spring 2024, this time to a larger class of nearly 40. The course will be offered to Barnard and Columbia students a third time in the upcoming Fall 2024 semester.