Faculty with Computational Neuroscience Research

George Konidaris

My research aims to build intelligent, autonomous, general-purpose robots capable of completing a wide variety of tasks in environments that are unstructured and partially specified. I focus on understanding how to build hierarchical robot controllers that can construct and execute long-horizon plans in unknown or semi-specified environments. I develop and apply techniques from machine learning, reinforcement learning, optimal control and planning to construct well-grounded hierarchies that result in fast planning for common cases, and are robust to uncertainty at every level of control. Link to website

< back to People