Abstract:
Many autonomous systems use perception, learning, and control to execute complex tasks ranging from autonomous driving to planetary exploration. The next generation of autonomous systems will require resilient algorithms that possess strong performance guarantees to enable safe operation in diverse environments. Predicting the performance of today’s autonomous systems is challenging as ad-hoc design strategies are often employed for algorithms responsible for critical tasks, such as perceiving the environment or avoiding obstacles. This talk will present a recently developed framework that utilizes nonlinear and adaptive control theory to predict and improve the performance of autonomous systems when different sources of uncertainty are present.

Bio:
Brett Lopez is a Postdoctoral Scholar at the NASA Jet Propulsion Laboratory in the Robotic Aerial Mobility Group where he leads a team of engineers and researchers designing the next generation of autonomous aerial robots for the DARPA Subterranean Challenge. He obtained his PhD (2019) and SM (2016) from MIT working with Prof. Jonathan How. He obtained his BS (2014) from UCLA where he received the Aerospace Engineering Outstanding Bachelor of Science award. His research establishes performance guarantees for complex autonomous systems through nonlinear/adaptive control theory and optimization.