Seminar

Seminar - OSIRIS-REx: To Bennu and Back - Sept. 13

Anonymous — Tue, 03 Sep 2024 13:42:13 +0000

Seminar - OSIRIS-REx: To Bennu and Back - Sept. 13 Anonymous (not verified) Tue, 09/03/2024 - 07:42

Friday, Sept. 13
10:40 a.m. - Seminar in AERO 120
11:30 a.m. - Panel Discussion / Q&A in AERO 111

This seminar will recount the two-year proximity operations and remote sensing campaign at Bennu, including the dramatic sample collection event and the events leading to the landing of the sample capsule in Utah.

A panel discussion will follow, featuring members of the Navigation and Flight Operations Team from NASA Goddard, Lockheed Martin, and KinetX, who will each recount specific challenges faced during the mission and the innovations that were implemented to overcome them.

Featured Speakers:

Dr. Michael C Moreau (AeroEngr MS’97, PhD’01) has worked at NASA’s Goddard Space Flight Center since 2001, and for over 10 years has served in leadership roles on the OSIRIS-REx Mission, as the manager of the Navigation Team during development, launch, and Bennu encounter, then as deputy project manager and leader of the sample return capsule recovery team. Mike’s Ph.D. research at CU focused on applications of the Global Positioning System in high Earth orbits, and contributed to the adoption of GPS on NASA missions such as GOES and Magnetosphere Multiscale. Before attending CU, he earned a BS in Mechanical Engineering at the University of Vermont.

Over three decades, Dr. Peter Antreasian (AeroEngr PhD’92) has made contributions to the navigation of NASA missions, Galileo, NEAR, Mars Odyssey, MER, Cassini-Huygens, GRAIL, and OSIRIS-REx. He began his career at the Jet Propulsion Laboratory in 1992, then joined KinetX 20 years later to lead the OSIRIS-REx navigation team. His expertise in orbit determination and navigation has been crucial in the success of these missions, including the first-ever landing of a spacecraft on an asteroid and the return of an asteroid sample to Earth. Peter earned his BS, MS and PhD in Aerospace Engineering, respectively, from Purdue, University of Texas and University of Colorado.

Dr. Jason Leonard (AeroEngr MS’12, PhD’15) received his Ph.D. in Aerospace Engineering Sciences from the University of Colorado Boulder under the advisement of Dr. George Born. Currently, he is the Orbit Determination Group Supervisor at KinetX Aerospace and Deputy Navigation Team Chief for the NASA OSIRIS-REx and OSIRIS-APEX missions. He has been the Orbit Determination Team Lead for OSIRIS-REx since prior to Launch, during the duration of proximity operations and its successful acquisition of asteroid regolith, and through its return of the sample to Earth. For his contributions to the mission, Jason received the NASA Exceptional Engineering Achievement Medal and the PI Award of Distinction.

Dr. Daniel Wibben is the Maneuver Design Group Supervisor for the Space Navigation and Flight Dynamics practice at KinetX Aerospace, Inc. Since joining the company, he has held the role of Maneuver and Trajectory lead for the OSIRIS-REx asteroid sample return mission. He has also been involved with the planning and operations of the LUCY, LunaH-Map, and DAVINCI missions. He received his B.S. in Aerospace and Mechanical Engineering, and M.S. and Ph.D. in Systems Engineering from the University of Arizona where his research was focused on nonlinear guidance techniques for asteroid proximity operations and planetary landing.

Coralie D. Adam (AeroEngr MS’17) is the Optical Navigation Group Supervisor at KinetX. She holds a B.S. in aerospace engineering and astronomy from the University of Illinois, and an M.S. in aerospace engineering sciences from the University of Colorado at Boulder. During her 12 years at KinetX, Coralie has had lead roles on the navigation teams for NASA’s New Horizons, OSIRIS-REx, Lucy, and OSIRIS-APEX missions. In addition to leading the OSIRIS-REx optical navigation subsystem from development through sample collection, she co-convened the scientific investigation of Bennu’s active particle ejection phenomena. Coralie is currently the deputy Navigation Team Chief on NASA’s Lucy mission, and a navigation lead and science co-investigator on the OSIRIS-APEX extended mission to asteroid Apophis.

Ryan Olds (AeroEngr BS’04, MS’09) has 19 years of experience in Guidance Navigation and Controls at Lockheed Martin Space supporting NASA Deep Space Exploration Missions. Ryan started his career working on the Pointing Control System for the Spitzer Space Telescope. He developed the reaction wheel control system for the twin-spacecraft GRAIL mission and supported test, integration, launch, and operations at the Moon. Ryan began working on OSIRIS-Rex in 2013 by developing control systems as well as the Natural Feature Tracking system which provided autonomous navigation for OSIRIS-REx during the mission’s sample acquisition phase. Ryan is currently a Guidance, Navigation and Controls manager and continues to support Deep Space Exploration missions such as OSIRIS-REx and DAVINCI.

Off

Traditional

White

Seminar: Impacts of Atmospheric Surface Layer Turbulence on the Wind turbine Drivetrain, and Generalization of Surface Layer Scaling - Apr. 9

Jeff Zehnder — Wed, 02 Apr 2025 15:16:58 +0000

Seminar: Impacts of Atmospheric Surface Layer Turbulence on the Wind turbine Drivetrain, and Generalization of Surface Layer Scaling - Apr. 9 Jeff Zehnder Wed, 04/02/2025 - 09:16

Jim Brasseur
Research Professor, Smead Aerospace
Wednesday, Apr. 9 | 9:35 a.m. | AERO 114

Abstract: I shall discuss two programs of research that have occupied much of my time and the time of two PhD students with whom I am working: Samantha Sheppard and Jarred Kenworthy. Jarred and Samantha’s programs both involve the special turbulence structure created by the impacts of impermeable surfaces on the structure of energy-dominant wall-bounded turbulence eddies. It is in this “surface layer” region that wind turbines operate in the atmospheric boundary layer. I will present key new knowledge extracted from high-fidelity large-eddy simulations by Jarred Kenworthy to understand the nonsteady responses of the wind turbine rotor to the passage of atmospheric surface layer turbulence eddies, with consequent impacts on the wind turbine drivetrain. Jarred’s research shows that strong turbulence-generated nonsteady forcing of the main bearing may underlie the currently excessive levels of main bearing failure.

A key characteristic of surface layer turbulence is the linear growth of integral scale eddy size with distance from the impermeable surface, a key element in what is commonly referred to as “law-of-the-wall” (LOTW) in turbulent boundary layer scaling. I shall describe key elements in a far-reaching fundamental study of surface layer scaling carried out by Samantha Sheppard within a unique experimental program at the University of Colorado Boulder and the Laboratoire de Mécanique des Fluides de Lille in France. We learn that linear scaling of energy-dominant eddy size is generalizable to wider classes of wall-bounded turbulent flows than previously appreciated, extending key LOTW scaling arguments beyond the classical flat-plate turbulent boundary layer for they were developed.

Bio: James Brasseur (Jim) is Research Professor of Aerospace Engineering Sciences at the University of Colorado Boulder. Before, Dr. Brasseur spend 27 years as Professor of Mechanical Engineering, Biomedical Engineering and Mathematics at the Pennsylvania State University, where he retains Emeritus Professor status. Dr. Brasseur is a fluid dynamist with extensive expertise in two general areas of research: (1) turbulence physics and simulation, and modeling within the large-eddy simulation framework, with applications to atmospheric turbulence, wind turbine aerodynamics and turbulent combustion; (2) fluid and solid mechanics integrated within physiology and function of the gastro-intestinal (GI) tract, as well as related issues in drug delivery, including dissolution, transport and absorption in the GI tract and in vitro devices. Dr. Brasseur has served on governing boards of the American Physical Society (APS), the APS Division of Fluid Dynamics, and two medical societies. was president of a medical society and was founding chair of the APS Topical Group on the Physics of Climate. Dr. Brasseur is a member of the Johns Hopkins Society of Scholars and is Fellow of the American Physical Society through the Division of Fluid Dynamics.

A key characteristic of surface layer turbulence is the linear growth of integral scale eddy size with distance from the impermeable surface, a key element in what is commonly referred to as...

Off

Traditional

White

Seminar - AI in Education: Advanced Applications and Emerging Trends with GenAI - Apr. 4

Jeff Zehnder — Fri, 28 Mar 2025 14:08:08 +0000

Seminar - AI in Education: Advanced Applications and Emerging Trends with GenAI - Apr. 4 Jeff Zehnder Fri, 03/28/2025 - 08:08

Bobby Hodgkinson
Associate Teaching Professor, Smead Aerospace
Friday, Apr. 4 | 12:45 - 2:15 P.M. | AERO N240

Abstract: As AI continues to evolve, so does its role in education. This talk will build on foundational AI literacy and dive deeper into how AI is reshaping education at a national and global level. We will explore cutting-edge implementations in K-12 and higher education, providing insights into how today’s students—our future aerospace engineers—are being exposed to AI tools before they arrive at ��Ҵ�ý�ƽ��.

The session will highlight AI-driven pedagogical advancements at peer institutions, examining initiatives that integrate AI into coursework, assessment, and educational research. We will discuss developments such as AI tutors, automated code reviews, and large-scale AI integration efforts across disciplines. Additionally, we will explore the ethical and pedagogical challenges of these tools, including concerns about bias, academic integrity, and the evolving role of educators in an AI-augmented classroom.

This talk will provide attendees with a forward-looking perspective on AI in education, equipping them with a deeper understanding of the trends shaping the field and the potential implications for engineering education. Attendees will leave with a clearer sense of what to expect from incoming students, how to align with peer institutions, and how to prepare for the next wave of AI-driven educational transformation.

Biography: Bobby is an Associate Teaching Professor in Aerospace Engineering at ��Ҵ�ý�ƽ��, exploring AI-driven education. He has developed AI-assisted grading tools, concept assessments, and code interviews to enhance student learning. He is actively involved in campus-wide AI initiatives and collaborates with colleagues across disciplines to explore AI’s role in education.

Beyond CU, he engages with the broader AI and education community to advance AI literacy. His talks explore practical applications of generative AI in education, from foundational tools to emerging trends shaping the future of learning.

As AI continues to evolve, so does its role in education. This talk will build on foundational AI literacy and dive deeper into...

Off

Traditional

White

Seminar: The Future of GPS - Apr. 4

Jeff Zehnder — Thu, 27 Mar 2025 14:08:08 +0000

Seminar: The Future of GPS - Apr. 4 Jeff Zehnder Thu, 03/27/2025 - 08:08

Mark Crews
PNT Senior Fellow, Lockheed Martin
Friday, Apr. 4 | 10:40 a.m. | AERO 114

Abstract: The Global Positioning System (GPS) has greatly increased efficiency and effectiveness of multiple industries that leverage GPS signals for Positioning, Navigation, and Timing (PNT).

The next generation of GPS III and GPS IIIF satellites are poised to increase performance for GPS users worldwide. GPS consists of three segments: space, ground, and user. The Space Segment includes a constellation of 31 satellites with seven GPS III satellites in the constellation and three more available for launch. When the Follow-on GPS IIIF satellites begin launch in 2026, GPS IIIF satellites will broadcast 60X more anti-jam power for warfighters.

Lockheed Martin also supports US Space Force operations of the GPS Control Segment, which includes a master and alternate master control stations with 16 worldwide monitor stations that constantly GPS satellite signals. The GPS User Segment includes billions of civilian GPS receivers and low millions of military GPS user equipment. The GPS enterprise has worked together seamlessly for decades to deliver robust PNT services to civil and military users. One of the most important examples of user services is the augmented use of GPS for Space-Based Augmentation System (SBAS) services utilized by civil aviation for precision approaches. SBAS provides integrity assurance for aviation operations without Instrument Landing System (ILS) infrastructure. This presentation provides an overview of GPS satellite modernization and how SBAS leverages GPS for Safety-of-Life operations.

Biography: Dr. Mark Crews is the LM Space Senior Fellow for Positioning, Navigation, and Timing (PNT) at Lockheed Martin. Dr. Crews directs the development of the next generation GPS architecture including laser crosslinks and advanced navigation payloads. He works closely with Government customers to incorporate new PNT requirements into innovative, compliant, and affordable designs.

His previous roles have included GPS Chief Engineer at the GPS Directorate; Chief of the Starfire Optical Range Beam Control Division; Assistant Professor of Electrical Engineering at the USAF Academy; Mission Systems Architect at Ball Aerospace; and Chief Technology Officer at ITT Space Systems Division. Dr. Crews is a retired Air Force Colonel. Dr. Crews has a Ph.D. in Electrical Engineering from the University of Oxford.

The Global Positioning System (GPS) has greatly increased efficiency and effectiveness of multiple industries that leverage GPS signals for Positioning, Navigation, and Timing (PNT). The next generation of..

Off

Traditional

White

Seminar: Emergency Location Service at Google - Mar. 21

Jeff Zehnder — Wed, 12 Mar 2025 14:08:08 +0000

Seminar: Emergency Location Service at Google - Mar. 21 Jeff Zehnder Wed, 03/12/2025 - 08:08

Mohammed Khider
Technical Lead and Manager, Google’s Android Location & Context Team
Friday, Mar. 21 | 10:40 a.m. | AERO 114

Abstract: When emergency services get a 911 call, they need to know the caller’s location to send help and save lives. Current emergency solutions rely on cell tower location and, in some countries, on A-GNSS. A-GNSS can fail with weak signal reception in urban canyons, and especially indoors. On the other hand, cell tower accuracy can be in the range of kilometers. In this presentation we will cover how Emergency Location Service in Android is delivering a more accurate location to emergency services when an emergency call is detected. A few live saving stories will be presented.

Bio: Dr. Mohammed Khider is a Technical Lead and Manager at Google’s Android Location & Context team. Within Android, he is leading a team that works on improving positioning and navigation accuracy of mobile devices in highways and in challenging urban canyon environments. He received his PhD in Communication Engineering with focus on "Multisensor based Positioning and Navigation" from the University of Ulm, Germany. Prior to his role at Google, he was a Research Associate at the Institute of Communication and Navigation at the German Aerospace Center (DLR) where he worked on various research projects related to positioning, navigation and context aware computing. Mohammed has been actively involved in the field of location and context for over 18 years and holds more than 25 patents. His other research interests include multi-sensor fusion, mobility models, signal processing and context-aware services. Mohammed has been an active ION member since 2008: published papers in various conferences & journals, acted as tracks and sessions chair, and as well served in various ION executive committees.

When emergency services get a 911 call, they need to know the caller’s location to send help and...

Off

Traditional

White

Seminar - GNSS Multipath: Characterization, Modeling & Mitigation - Mar. 14

Jeff Zehnder — Sat, 08 Mar 2025 15:08:08 +0000

Seminar - GNSS Multipath: Characterization, Modeling & Mitigation - Mar. 14 Jeff Zehnder Sat, 03/08/2025 - 08:08

Gary McGraw
Systems Engineering Consultant
Friday, Mar. 14 | 10:40 a.m. | AERO 114

Abstract: Global Navigation Satellite Systems (GNSS) like the US Global Positioning System (GPS) are critical to the operation of many aspects of our modern society and are used daily by billions of people. A major error source for GNSS is multipath propagation and mitigating these errors is a major concern in high-accuracy positioning applications like survey, aircraft landing, and precision construction and agriculture. This talk presents an introduction to what multipath is, how it affects GNSS measurements, and ways to mitigate its effects. The use of carrier smoothing of code pseudorange measurements is discussed as an effective multipath and noise mitigation technique that does not require access to the inner workings of the receiver like other mitigation techniques require. Recent developments in approaches to statistically model multipath errors as stochastic processes and how the multipath error statistics are affected by carrier-code smoothing will be introduced. These techniques are useful in safety of life applications, like aviation navigation, where it is necessary to have statistical bounds for error sources.

Biography: Dr. Gary McGraw is a systems engineering consultant, specializing in Positioning, Navigation and Timing (PNT). He retired as a Technical Fellow at Collins Aerospace where he led the development of several high accuracy and high integrity navigation systems for civil aviation and military applications. His current research areas are focused on the use of communication data links for positioning and timing and GNSS-based aircraft landing system developments. He received the B.S. degree in Electrical Engineering and Mathematics from Iowa State University, and the M.S. and Ph.D. degrees in Electrical Engineering from the University of California, Los Angeles. Dr. McGraw is a Fellow of the Institute of Navigation (ION) and is a Senior Member of the IEEE. He is an Associate Editor of the ION NAVIGATION journal, was the recipient of the 2011 Johannes Kepler Award from the ION, and currently serves as the ION President.

Global Navigation Satellite Systems (GNSS) like the US Global Positioning System (GPS) are critical to the operation of many aspects of our modern society and are used daily by...

Off

Traditional

White

Seminar - AI Literacy: Foundations and Practical Applications of GenAI in Education - Mar. 14

Jeff Zehnder — Fri, 07 Mar 2025 22:53:54 +0000

Seminar - AI Literacy: Foundations and Practical Applications of GenAI in Education - Mar. 14 Jeff Zehnder Fri, 03/07/2025 - 15:53

Bobby Hodgkinson
Associate Teaching Professor, Smead Aerospace
Friday, Mar. 14 | 12:45 - 2:15 P.M. | AERO N240

Abstract: Artificial intelligence is no longer a distant concept—it’s here, and can help shape the way we teach, learn, and assess. This talk will provide a foundational understanding of generative AI (GenAI) in education, focusing on what faculty, staff, and students at ��Ҵ�ý�ƽ�� can do right now with available tools and resources. We will explore how chatbots, custom GPTs, and applications like NotebookLM and deep research are transforming studying, research, and classroom engagement.

Beyond these entry points, the talk will examine AI-augmented applications, including automated grading assistants, AI-enhanced concept exams, and AI-driven code interviews. Attendees will also gain insight into how generative AI is being used across campus, including the Leeds School of Business, the Office of Information Technology, and the Center for Teaching and Learning. We will then extend the discussion to the broader community, highlighting AI initiatives within the Boulder Valley School District (BVSD) and the Rocky Mountain AI Interest Group (RMAIIG).

This session is designed to provide a clear, practical roadmap for getting started with generative AI today—without requiring technical expertise or financial investment. By the end, participants will leave with an understanding of the immediate applications of AI in their professional and educational environments, as well as awareness of the growing AI ecosystem in Boulder and beyond.

Artificial intelligence is no longer a distant concept—it’s here, and can help shape the way we teach, learn, and...

Off

Traditional

White

Seminar - Differentiable Physics: A physics-constrained and data-driven paradigm for scientific discovery - Mar. 12

Jeff Zehnder — Thu, 06 Mar 2025 16:37:57 +0000

Seminar - Differentiable Physics: A physics-constrained and data-driven paradigm for scientific discovery - Mar. 12 Jeff Zehnder Thu, 03/06/2025 - 09:37

Romit Maulik
Assistant Professor, Data Science, Pennsylvania State University
Wednesday, Mar. 12 | 10 a.m. | AERO 114

Abstract: Machine learning stands poised to revolutionize the process of scientific discovery across various disciplines. In this talk, we will introduce a state-of-the-art scientific machine learning paradigm - differentiable physics (DiffPhys). DiffPhys can be considered a system identification paradigm that can be applied to determine neural network approximations of governing laws given data. It can also be used to improve first-principles-based simulations of physical phenomena by learning corrections to governing laws (for instance for closure modeling in multiscale applications). Notably, optimizing these neural networks necessitates a differentiable programming paradigm where gradients of a loss function can be propagated through a numerical solver. In this talk, we will introduce DiffPhys algorithms that (1) can learn models for dynamical systems from sparse data, (2) efficiently compute sensitivities for systems exhibiting deterministic chaos, (3) leverage graph neural networks for geometry-invariant learning, and (4) provide physically meaningful interpretations for neural network behavior thereby engendering scientific discovery. We will demonstrate the capabilities of DiffPhys on canonical and realistic scientific computing problems and close with a discussion of the future possibilities of this approach.

Bio: Romit Maulik is an Assistant Professor of Data Science in the College of Information Sciences and Technology at Pennsylvania State University and a Joint Appointment Faculty at the Mathematics and Computer Sciences Division at Argonne National Laboratory (Argonne). He obtained his PhD in Mechanical and Aerospace Engineering at Oklahoma State University in 2019 and was the Margaret Butler Fellow and then a Staff Scientist at Argonne National Laboratory before joining Penn State in 2023. His research centers around machine learning for scientific computing with an emphasis on scalable, physically consistent, and robust algorithm construction for simulation-based scientific discovery of multiscale physics from multifidelity data. He has led research projects sponsored by multiple agencies and is an Early Career Awardee of the Army Research Office.

Machine learning stands poised to revolutionize the process of scientific discovery across various disciplines. In this talk, we will introduce a state-of-the-art scientific machine...

Off

Traditional

White

Seminar: Surrogate Modeling for the Multi-Physics Design Analysis and Optimization of Hypersonic Systems - Mar. 7

Jeff Zehnder — Mon, 03 Mar 2025 16:56:05 +0000

Seminar: Surrogate Modeling for the Multi-Physics Design Analysis and Optimization of Hypersonic Systems - Mar. 7 Jeff Zehnder Mon, 03/03/2025 - 09:56

Charbel Farhat
Professor, Aeronautics and Astronautics, Stanford University
Friday, Mar. 7 | 1 p.m. | AERO 111 | Hybrid - Zoom

Abstract: The AFOSR MURI effort, titled “A Robust Multi-Physics Design Analysis and Optimization Framework for Hypersonic Systems Grounded in Rigorous Model Reduction,” unites a multi-disciplinary team of engineering scientists from Stanford University, the University of Minnesota, Rice University, and the University of Notre Dame. This ambitious and collaborative initiative is dedicated to tackling the computational challenges inherent in the development of hypersonic systems. The team’s primary goal is to design a computationally efficient framework for multi-physics design analysis and optimization under uncertainty, harnessing the power of surrogate modeling techniques to achieve remarkable efficiency gains. The lecture will begin with a comprehensive yet succinct overview of the global research effort, offering insights into the project’s objectives, broader impact, and the computational obstacles it seeks to overcome. Among these challenges are the surrogate modeling of quantities of interest versus spatio-temporal fields, the complexities of low-dimensional modeling in the presence of shocks, shock-shock and shock-boundary layer interactions, turbulence, and the intricacies of training surrogate models in high-dimensional parameter spaces for coupled vehicle-trajectory problems and optimization. The discussion will then shift to recent contributions by the speaker, including advanced nonlinear projection-based reduced-order models that address the closure error in the latent space using a deep learning approach. These models have been validated for hypersonic benchmark flow problems and have led to the development of an efficient global optimization method. Finally, the lecture will explore the application of these methods to simulate hypersonic flight dynamics and optimize trajectory planning, demonstrating their potential for advancing the design and analysis of hypersonic systems.

Bio: Charbel Farhat is the Vivian Church Hoff Professor of Aircraft Structures in the School of Engineering at Stanford University, where from 2008 to 2023, he chaired the Department of Aeronautics and Astronautics. He is also Professor in the Institute for Computational and Mathematical Engineering. He was designated by the US Navy recruiters as a Primary Key-Influencer and flew with the Blue Angels during Fleet Week 2014. He is a Member of the National Academy of Engineering (US); a Member of the Royal Academy of Engineering (UK); a member of the Lebanese Academy of Sciences; a Doctor Honoris Causa from Ecole Nationale Supérieure d’Arts et Métiers, Ecole Centrale de Nantes, and Ecole Normale Supérieure Paris-Saclay; a designated ISI Highly Cited Author in Engineering; and a Fellow of AIAA, ASME, IACM, SES, SIAM, USACM, and WIF. He has trained so far about 100 PhD and post-doctoral students. For his research on aeroelasticity, aeroacoustic scattering, CFD, dynamic data-driven systems, fluid-structure interaction, high performance computing, model reduction, and physics-based machine learning, he has received many professional and academic distinctions including: the Ashley Award for Aeroelasticity and the Structures, Structural Dynamics and Materials Award from AIAA; the Spirit of St Louis Medal and a Lifetime Achievement Award from ASME; the Gordon Bell Prize and the Sidney Fernbach Award from IEEE; the Gauss-Newton Medal from IACM; the Grand Prize from the Japan Society for Computational Engineering Science; the John von Neumann Medal from USACM; and the Olof B. Widlund Prize for Excellence in Domain Decomposition Methods from DDM.org.

The AFOSR MURI effort, titled “A Robust Multi-Physics Design Analysis and Optimization Framework for Hypersonic Systems Grounded in Rigorous Model Reduction,” unites a...

Off

Traditional

White

Seminar - Learning for Autonomy: From Policies to Operators to Distributions - Mar. 10

Jeff Zehnder — Thu, 27 Feb 2025 17:15:16 +0000

Seminar - Learning for Autonomy: From Policies to Operators to Distributions - Mar. 10 Jeff Zehnder Thu, 02/27/2025 - 10:15

Adam Thorpe
Postdoctoral Researcher, University of Texas at Austin
Monday, Mar. 10 | 10 a.m. | AERO 114

Abstract: Autonomous systems must learn, adapt, and make decisions in novel, unpredictable environments. However, data-driven approaches often struggle to generalize and can be fragile in such environments. My research addresses this challenge by developing learning-based methods that explicitly incorporate mathematical structure, enabling autonomy to adapt, scale, and generalize.

Central to this approach are structured representations of policies, operators, and distributions that leverage Hilbert space theory, statistical learning, and neural network models. In this talk, I will present results demonstrating how autonomous systems can adapt and transfer to new scenarios within seconds using minimal online data—without the need for additional retraining.
I will highlight advances in neural operator learning, where efficient function-to-function mappings achieve orders-of-magnitude improvements in accuracy over state-of-the-art methods, with significant implications for autonomy. Finally, I will discuss efforts to design autonomous systems that operate safely around humans by tailoring responses to individual needs and preferences.

Bio: Adam Thorpe is a postdoctoral researcher at the Oden Institute for Computational Engineering and Sciences at The University of Texas at Austin. He received his PhD at the University of New Mexico in 2023. Adam's research interests are in the area of data-driven and learning-based control, with applications to humans and autonomy, space systems, and robotics.

Autonomous systems must learn, adapt, and make decisions in novel, unpredictable environments. However, data-driven approaches often struggle to...

Off

Traditional

White