AURORA: Autonomous Updating of ROM and Controller via Recursive Adaptation
Jiachen Li, Shihao Li, Dongmei Chen

TL;DR
AURORA is a supervisory framework that autonomously updates reduced order models and controllers in real-time for high-dimensional nonlinear systems, improving performance and diagnostics without expert tuning.
Contribution
It introduces a recursive, agent-based adaptation framework with performance classification and guarantees for linear systems, validated empirically on complex benchmarks.
Findings
Achieves 6-12% better tracking than expert baselines.
Diagnoses operational issues with 91% accuracy.
Validated on systems with up to 5177 dimensions.
Abstract
Real time model based control of high dimensional nonlinear systems presents severe computational challenges. Conventional reduced order model control relies heavily on expert tuning or parameter adaptation and seldom offers mechanisms for online supervised reconstruction. We introduce AURORA, Autonomous Updating of ROM and Controller via Recursive Adaptation, a supervisory framework that automates ROM based controller design and augments it with diagnostic triggered structural adaptation. Five specialized agents collaborate through iterative generate judge revise cycles, while an Evaluation Agent classifies performance degradation into three operationally distinct categories, subspace inadequacy, parametric drift, and control inadequacy, and routes corrective action to the responsible agent. For linear ROMs, we analytically prove that this classification is correct under mild…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsModel Reduction and Neural Networks · Advanced Control Systems Optimization · Control Systems and Identification
