This paper introduces a self-tuning principle for parametrically excited oscillators that enables large-amplitude oscillations without relying on detailed models. The approach combines slow parameter adaptation with fast feedback modulation to achieve near-optimal amplitude maximization. Experimental results show that the oscillator can adapt itself close to dynamic bifurcation, reaching extreme sensitivity to small perturbations and producing robust large-amplitude oscillations even under model uncertainties and external disturbances.
Why it matters: Parametric oscillators have wide applications in signal amplification, communication, and computation, but their use has been limited by the need for precise tuning and strong modulation. This self-tuning method overcomes those limitations, offering a simple, robust, and practical way to exploit nonlinear dynamics for real-world technologies.