Home | Center for Autonomy

Welcome to the Center for Autonomy

The Center for Autonomy brings together several research groups that address fundamental challenges in developing autonomous systems through contributions in controls, machine learning, game theory, information theory, and formal methods. Its primary objective is to create a unified front in attracting the best researchers to UT Austin and empowering them to solve the pressing problems toward developing autonomous systems that can make a net positive impact.

Upcoming Events

The Role and Value of Information in Multi-Agent Decisions

Event Status

Scheduled

Monday May 13, 2024, 11:00 am - 12:30 pm

POB 6.304

Uncertainty is ubiquitous in virtually all real-world interactions between agents. Without knowing the state of the world or the intentions of others, rational agents must use the limited information at their disposal for decision-making, maximizing their expected payoff and controlling risk. In this setting, information itself becomes valuable: statistical decision theory (SDT) defines the value of information (VoI) as the expected gains from informed decision-making. When a decision-maker has the opportunity to first gather additional information at some cost, VoI provides a principled basis by which to make that decision. We introduce VoI in the multi-agent setting, within games of incomplete information. We use VoI to define a general information design problem, in which an agent has the opportunity to decide how to preemptively gather information, prior to a non-cooperative scenario. We introduce two approaches for solving this problem: a Bayesian optimal, decision-theoretic approach, and an information-theoretic approach based on hypothesis clustering. We demonstrate the decision-theoretic method in a smooth, Blotto-like tower defense game. The clustering approach is demonstrated in sensing allocation for a traffic routing control problem, against exogenous data-poisoning attacks.

News

Takashi Tanaka, Promoted to Associate Professor of Aerospace Engineering

April 29, 2024

Professor Takashi Tanaka, a member of the Center for Autonomy, has been promoted to Associate Professor of Aerospace Engineering with tenure as of September 1st, 2024. Congratulations to Professor Tanaka!

The Community Driven Approaches to Research in Technology & Society visioning workshop organizers

Visioning Workshop Report Released: Community Driven Approaches to Research in Technology & Society

March 14, 2024

The Community Driven Approaches to Research in Technology & Society visioning workshop organizers, Timnit Gebru (Distributed Artificial Intelligence Research Institute), Ufuk Topcu (University of Texas at Austin), and Suresh Venkatasubramanian (Brown University), along with support from Haley Griffin (CCC), Nasim Sonboli (Brown University), and Leah Namisa Rosenbloom (Brown University), have written a workshop report that synthesizes the amazing ideas discussed throughout the visioning workshop. This report is a result of the ideas, experiences, recommendations, and visions of the incredible group of workshop participants.

This staged scene and illustration represents what you might see when NASA’s University Innovation project awards teams of students with funds to conduct real-world aeronautical research that will help the agency transform aviation for the 21st century.

NASA Selects University Teams to Explore Innovative Aeronautical Research

Feb. 23, 2024

NASA collaborates with George Washington University and the University of Texas at Austin to investigate measures that can be taken to safeguard autonomous aircraft flying in high-density urban airspace from cyberattacks that could disrupt safe operations.

Research Projects

AI generated photo of city with skyscrapers to demonstrate new Chishik.AI research

Chishiki.ai: An integrated community of Computational Infrastructure professionals (CIPs) across AI and CEE sectors

The Chishiki project adopts four strategies to build a sustainable, diverse, and integrated community of CI professionals for AI in CEE by (1) fostering collaboration between CIPs and domain experts through initiatives such as research summits, graduate and undergraduate fellowships, joint research initiatives, and industrial partnerships; (2) offering personalized and scalable learning environments powered by AI; (3) developing innovative AI-enabled CI architectures for reproducible and efficient workflows; and (4) creating a diverse, sustainable CIP community through engagement with historically underrepresented institutions through recruitment and research initiatives.

Autonomous Aerial Cargo Operations at Scale

The project addresses the challenge of "safe and efficient growth in global operations, with the expected outcome of an algorithmic foundation that will help realize increasingly autonomous and collaborative air traffic management for all classes of airspace and vehicles and support scalable and efficient operations that rapidly adapt to meet changing demands and to respond to system disruptions.

Learning Mixed Strategies in Trajectory Games

In multi-agent settings, game theory is a natural framework for describing the strategic interactions of agents whose objectives depend upon one another’s behavior. Trajectory games capture these complex effects by design.

A recent paper by members of the DCIST alliance develops a perception-action-communication loop framework using Vision-based Graph Aggregation and Inference (VGAI). This multi-agent decentralized learning-to-control framework maps raw visual observations to agent actions, aided by local communication among neighboring agents.

Learning Decentralized Controllers with Graph Neural Networks