Description:

Dr. Cathy Wu 

Assistant Professor 

MIT 

Abstract: 

Emerging connected and automated vehicle (CAV) technologies bring hope for addressing long-standing challenges in transportation, including congestion, safety, and environmental sustainability. However, the rich sensing and actuation modalities afforded by CAV technologies significantly increase the complexity of designing future mobility systems, pushing model-based methods to their limits. This talk discusses advances in computational methods for coping with such increasing complexity. The talk presents work that: 1) exposes the brittleness of modern model-free machine learning-based methods, and 2) demonstrates the robustness of hybridized model-based and model-free methods for complex multi-agent coordination problems in transportation. Applications discussed include mixed autonomy traffic, traffic signal control, vehicle routing problems, multi-robot warehousing, and cooperative eco-driving.

Bio: 

Cathy Wu is an Assistant Professor at MIT in LIDS, CEE, and IDSS. She holds a Ph.D. from UC Berkeley, and B.S. and M.Eng. from MIT, all in EECS, and completed a Postdoc at Microsoft Research. Her research interests are at the intersection of machine learning, autonomy, and mobility. She is broadly interested in developing the tools and understanding necessary to confidently integrate automated decisions into societal & industrial systems. Her recent focus consists of two synergistic thrusts: 1) algorithms for contextual control & decision problems, particularly developing robust computational methods for complex multi-agent coordination problems, and 2) traffic benchmarking, particularly large-scale modeling, control, and analysis of performance measures in multi-agent traffic systems. Practically, she applies computational methods to transportation decarbonization, traffic congestion & safety, mixed autonomy traffic, robotic warehousing, and vehicle routing & scheduling. Cathy is received a number of awards, including the NSF CAREER, numerous dissertation awards, and publications with distinctions.