AAE Research Seminar Series - Two Seminars by Dr David Rosen and Dr Jianyong Wang

Seminar 1: Certifiably Correct Machine Perception

Abstract

Many fundamental machine perception tasks require the solution of a high-dimensional nonconvex estimation problem; this class includes (for example) the foundational problems of simultaneous localization and mapping (in robotics), 3D reconstruction (in computer vision), and sensor network localization (in distributed sensing).  Such problems are known to be computationally hard in general, with many local minima that can entrap the smooth local optimization methods commonly applied to solve them.  The result is that standard machine perception algorithms (based upon local optimization) can be surprisingly brittle, often returning egregiously wrong answers even when the problem to which they are applied is well-posed.

In this talk, we present a novel class of certifiably correct algorithms that are capable of efficiently recovering globally optimal solutions of challenging machine perception problems in many practical settings.  In brief, our approach directly tackles the problem of nonconvexity by employing a convex relaxation (which can be efficiently and globally solved) whose minimizer we prove provides an exact (globally optimal) solution to the original estimation problem under moderate measurement noise.  We illustrate the design of this class of methods using the fundamental problem of pose-graph optimization as a running example, culminating in the presentation of SE-Sync, the first practical method provably capable of recovering exact (globally optimal) PGO solutions.

Speaker

Dr. David M. Rosen is an Assistant Professor in the Department of Electrical and Computer Engineering at Northeastern University (NEU).  His research addresses the design of practical provably robust methods for machine perception and control, using a combination of tools from optimization, geometry, algebra, and probabilistic inference.  He holds the degrees of BSc in Mathematics from the California Institute of Technology (2008), MA in Mathematics from the University of Texas at Austin (2010), and ScD in Computer Science from the Massachusetts Institute of Technology.  Before joining NEU, he was a Postdoctoral Scholar at MIT’s Laboratory for Information and Decision Systems (LIDS) and a Research Scientist at Oculus Research (now Meta Reality Labs) in Seattle.

His work has been recognized with a Best Paper Award at the 2016 International Workshop on the Algorithmic Foundations of Robotics, an RSS Pioneer Award at Robotics: Science and Systems 2019, and a Best Student Paper Award at Robotics: Science and Systems 2020.

Seminar 2: Computational & Experimental Investigations on the Turbulent Heat Transfer of Supercritical Fluids

Abstract

Driven by the great potential as ideal heat transfer mediums being utilized in a variety of crucial technologies, such as regenerative cooling for hypersonic vehicles and rocket walls, cooling design for nuclear reactors and Concentrating Solar Thermal (CST) power plants, supercritical fluids (for instance supercritical water, CO2 and hydrocarbon fuels) have been attracting considerable attentions worldwide. Lots of studies have been performed to gain a better understanding on the supercritical fluid flow and heat transfer characteristics. This presentation will introduce the unique properties and relevant applications of supercritical fluids, and demonstrate Dr Wang’s past, present and future works through the combined methods of simulations & experiments.

Speaker

Dr Jianyong Wang received his Bachelor of Power Engineering of Aircraft and Master of Aerospace Engineering from Nanjing University of Aeronautics & Astronautics (NUAA) in Nanjing, China. He then went to the University of Queensland in Brisbane, Australia to pursue his Doctorate. As part of the Australia Solar Thermal Research Initiative (ASTRI) supported by the Australian Government, his PhD project used the Computational Fluid Dynamics (CFD) to study the supercritical carbon dioxide flow & heat transfer features from fundamental aspects. After completion of his PhD in 2019, Dr Wang started his new research career as an assistant professor in Sun Yat-sen University (SYSU) and continue the supercritical research with the help of CFD and experiments.