Computer Science Seminar: Brian Plancher (Harvard University)
Speaker: Brian Plancher (Harvard University)
Title: Hardware Acceleration for Realtime Robotics
The seminar will be available for Zoom participation. People in the Barnard/Columbia community who have authorized access to Barnard indoor spaces can participate in person if desired. If you would like to receive the Zoom link, please register using the “Register” button above.
In order for robots to achieve their full potential to assist humans, they need to be able to reliably operate in dynamic and unstructured real-world environments. Opportunities exist for roboticists to leverage hardware acceleration to overcome many of the computational bottlenecks that confine robots to the lab. In this talk I will show how Graphics Processing Units (GPUs) can be used to improve the performance of robotics algorithms through a case study of rigid body dynamics, and preview the potential for further acceleration through the use of custom Application Specific Integrated Circuits (ASICs).
Brian Plancher is a Ph.D. Candidate studying Robotics at Harvard University working with Vijay Janapa Reddi and Scott Kuindersma. His research is focused on developing and implementing open-source algorithms for dynamic motion planning and control of robots by exploiting both the mathematical structure of algorithms and the design of computational platforms (e.g., CPUs, GPUs, FPGAs). As such, his research sits at the intersection of Robotics and both Computer Architecture and Numerical Optimization. Brian also wants to improve the accessibility of STEM education and co-chairs TinyMLedu. He enjoys teaching and designing new interdisciplinary, project-based, open-access courses that lower the barrier to entry of cutting edge topics like robotics and TinyML. Brian enjoys spending his free time with his wife, daughter, and puppy, and ski racing in the winters.