An adaptive data sampling strategy for stabilizing dynamical systems via controller inference

2025-06-02

AI Summary

In unstable dynamical systems, data collection often leads to trajectory divergence, hindering acquisition of informative and safe training data. Method: This paper proposes an adaptive “sample-while-stabilize” data generation strategy that jointly optimizes online sampling and real-time feedback control. It integrates excitation signal design under stability constraints with a model-agnostic controller inference framework to simultaneously ensure data informativeness and system safety. Contribution/Results: We theoretically establish that the resulting dataset is both sufficient—enabling stable controller learning—and minimal—eliminating redundancy. Numerical experiments demonstrate that the method achieves successful stable controller learning using only 10% of the data required by conventional unguided sampling, significantly improving controllability and data efficiency under edge and extreme operating conditions.

Technology Category

Search and Optimization: Sampling/Simulation-based SearchIntelligent Robots: Behavior Learning & ControlMultiagent Systems: Adversarial Agents

Application Category

Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAISecurity and Privacy: Data transparency and provenanceSemantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Learning stabilizing controllers from data is an important task in engineering applications; however, collecting informative data is challenging because unstable systems often lead to rapidly growing or erratic trajectories. In this work, we propose an adaptive sampling scheme that generates data while simultaneously stabilizing the system to avoid instabilities during the data collection. Under mild assumptions, the approach provably generates data sets that are informative for stabilization and have minimal size. The numerical experiments demonstrate that controller inference with the novel adaptive sampling approach learns controllers with up to one order of magnitude fewer data samples than unguided data generation. The results show that the proposed approach opens the door to stabilizing systems in edge cases and limit states where instabilities often occur and data collection is inherently difficult.

Problem

• Learning stabilizing controllers from unstable system data
• Adaptive sampling prevents instability during data collection
• Minimizes data samples needed for effective controller inference

Innovation

• Adaptive sampling stabilizes during data collection
• Minimal informative datasets for stabilization
• Fewer samples than unguided data generation