Forthcoming mobile robots require efficient generalizable algorithms to operate in challenging and unknown environments without human intervention while collaborating with humans. Today, despite the rapid progress in robotics and autonomy, no robot can deliver human-level performance in everyday tasks and missions such as search and rescue, exploration, and environmental monitoring and conservation.
Maani Ghaffari,
University of Michigan
In this talk, I will put forward a vision for enabling efficiency and generalization requirements of real-world robotics via computational symmetry and learning. I will walk you through structures that arise from combining symmetry, geometry, and learning in various foundational problems in robotics and showcase their performance in experiments ranging from perception to control. In the end, I will share my thoughts on promising future directions and opportunities based on lessons learned in the field and on campus.
Maani Ghaffari received the Ph.D. in Engineering (Robotics) from the Centre for Autonomous Systems (CAS), University of Technology Sydney, NSW, Australia, in 2017. He is currently an Associate Professor at the Department of Naval Architecture and Marine Engineering and the Department of Robotics, University of Michigan, Ann Arbor, MI, USA, where he directs the Computational Autonomy and Robotics Laboratory (CURLY). His work on sparse, globally optimal kinodynamic motion planning on Lie groups received the best paper award finalist title at the 2023 Robotics: Science and Systems conference. He received the 2025 AFOSR Young Investigator Research Program (YIP) and the 2021 Amazon Research Awards. His research interests lie in the theory and applications of robotics and autonomous systems. He has expertise in developing efficient generalizable algorithms for robot autonomy operating in unstructured, dynamic, and uncertain environments. He has developed geometric state estimation, geometric control methods, geometric deep neural network architectures (Lie Neurons), information-theoretic motion planning, and mapping algorithms for mobile robots.