Adaptive KalmanNet: Data-Driven Kalman Filter with Fast Adaptation

TL;DR

Adaptive KalmanNet (AKNet) combines deep learning with Kalman filtering to enable rapid adaptation to changing state space models without retraining, maintaining accurate tracking across diverse noise conditions.

Contribution

The paper introduces AKNet, a novel DNN-aided Kalman filter that uses a hypernetwork for fast, data-driven adaptation to model changes without retraining.

Findings

AKNet maintains accurate state estimation across various noise distributions.

AKNet adapts quickly to model changes without retraining.

AKNet performs well even with limited training data.

Abstract

Combining the classical Kalman filter (KF) with a deep neural network (DNN) enables tracking in partially known state space (SS) models. A major limitation of current DNN-aided designs stems from the need to train them to filter data originating from a specific distribution and underlying SS model. Consequently, changes in the model parameters may require lengthy retraining. While the KF adapts through parameter tuning, the black-box nature of DNNs makes identifying tunable components difficult. Hence, we propose Adaptive KalmanNet (AKNet), a DNN-aided KF that can adapt to changes in the SS model without retraining. Inspired by recent advances in large language model fine-tuning paradigms, AKNet uses a compact hypernetwork to generate context-dependent modulation weights. Numerical evaluation shows that AKNet provides consistent state estimation performance across a continuous range of noise distributions, even when trained using data from limited noise settings.