InfoSSM: Interpretable Unsupervised Learning of Nonparametric State-Space Model for Multi-modal Dynamics
This work addresses the problem of accurately describing nonparametric, multi-modal dynamics like aircraft motions for researchers in system identification and control, representing an incremental advance over traditional Gaussian process state-space models.
The paper tackled the challenge of modeling complex, multi-modal dynamics in system identification by proposing InfoSSM, an interpretable unsupervised learning framework using multiple Gaussian process transition models with mutual information regularization, achieving improved performance and interpretability in experiments including a high-fidelity flight simulator.
The goal of system identification is to learn about underlying physics dynamics behind the time-series data. To model the probabilistic and nonparametric dynamics model, Gaussian process (GP) have been widely used; GP can estimate the uncertainty of prediction and avoid over-fitting. Traditional GPSSMs, however, are based on Gaussian transition model, thus often have difficulty in describing a more complex transition model, e.g. aircraft motions. To resolve the challenge, this paper proposes a framework using multiple GP transition models which is capable of describing multi-modal dynamics. Furthermore, we extend the model to the information-theoretic framework, the so-called InfoSSM, by introducing a mutual information regularizer helping the model to learn interpretable and distinguishable multiple dynamics models. Two illustrative numerical experiments in simple Dubins vehicle and high-fidelity flight simulator are presented to demonstrate the performance and interpretability of the proposed model. Finally, this paper introduces a framework using InfoSSM with Bayesian filtering for air traffic control tracking.