Learning IMM Filter Parameters from Measurements using Gradient Descent

TL;DR

This paper introduces a method to automatically optimize IMM filter parameters directly from measurement data using gradient descent, eliminating the need for ground-truth information and simplifying complex sensor system tuning.

Contribution

It presents a novel approach to tune IMM filter parameters solely from measurements via gradient descent, bypassing the need for ground-truth data and expert tuning.

Findings

The method achieves performance comparable to ground-truth parameterized filters.

It simplifies the tuning process for complex sensor systems.

The approach is validated through simulation experiments.

Abstract

The performance of data fusion and tracking algorithms often depends on parameters that not only describe the sensor system, but can also be task-specific. While for the sensor system tuning these variables is time-consuming and mostly requires expert knowledge, intrinsic parameters of targets under track can even be completely unobservable until the system is deployed. With state-of-the-art sensor systems growing more and more complex, the number of parameters naturally increases, necessitating the automatic optimization of the model variables. In this paper, the parameters of an interacting multiple model (IMM) filter are optimized solely using measurements, thus without necessity for any ground-truth data. The resulting method is evaluated through an ablation study on simulated data, where the trained model manages to match the performance of a filter parametrized with ground-truth values.