Welcome to ACXIS Lab

Autonomous Control, eXploration, Intelligence, & Systems

ISO
NEWS
A New Paper Published
Neural-Rendezvous for In-Situ Exploration of Interstellar Objects – Our new paper has finally been published in the AIAA Journal of Guidance, Control, and Dynamics
See here for our 1-min YouTube summary
PlayPlay
Summary
QUICK SUMMARY
ACXIS Laboratory
We aim to establish theoretical foundations for the END-TO-END design and operation of trustworthy autonomous systems, with applications to aerospace/robotic autonomy
(Facility: ARCL, Caltech)
PlayPlay
Control
MOONSHOT 1
Autonomous Control
General theory for every aspect of autonomous decision-making problems in highly uncertain, nonlinear, and large-scale dynamics and environments with severe external disturbances
(Facility: ARCL, Caltech)
PlayPlay
Exploration
MOONSHOT 2
Autonomous Exploration
Fully autonomous explorers for safety-critical aerospace and robotic missions operating in various extreme environments, with significantly challenging exploration objectives
(3D resources: NASA 3D Resources)
PlayPlay
Intelligence
MOONSHOT 3
Autonomous Intelligence
Formal ways to enjoy mathematical guarantees of nonlinear control theory, such as robustness, stability, safety, and optimality in any data-driven, learning-based approaches
(Facility: ARCL, Caltech)
PlayPlay
Systems
MOONSHOT 4
Autonomous Systems
Integrated design, operation, and decision-making frameworks for robust, safe, adaptive, and optimal autonomous systems, which could accomplish any task with formal guarantees
(Image Credit: NASA JPL)
PlayPlay
previous arrow
next arrow

NEWS

Our tutorial session on “Planning and Control with Machine Learning for Autonomous and Robotic Systems,” by Soon-Jo Chung (Caltech), Hiroyasu Tsukamoto (UIUC), and Guanya Shi (CMU) won the Best Tutorial Award at IEEE SMC.

ACXIS Lab has multiple open Ph.D. and M.S. positions (https://hirotsukamoto.com/positions/) for highly motivated students.

We are going to host a tutorial session “Planning and Control with Machine Learning for Autonomous and Robotic Systems” at IEEE SMC. This session gives a tutorial overview of machine learning control systems with safety and stability guarantees.

Hiroyasu Tsukamoto’s doctoral dissertation won the William F. Ballhaus Prize for the best dissertation in Space Engineering, GALCIT, Caltech.

We hosted a tutorial session “Contraction Theory for Machine Learning” at IEEE CDC. This website provides an overview of contraction theory for nonlinear stability analysis and control synthesis, with an emphasis on deriving formal guarantees for learning-based control problems.

Introduction
FACULTY
Hiroyasu Tsukamoto
Incoming Assistant Professor of Aerospace, UIUC; Research Affiliate, NASA JPL.
Interested? Let's talk.

Assistant Professor of Aerospace, UIUC, 2024 –; Adjunct Assistant Professor of Aerospace, UIUC, 2023 –; Postdoctoral Research Affiliate, NASA JPL, 2023 –; Ph.D., Caltech (Best Dissertation, Space Engineering), 2023; M.S., Caltech, 2018; B.S., Kyoto University, 2017.