Skip to main content

Research Repository

Advanced Search

A Generative Learning Approach to Sensor Fusion and Change Detection

Gepperth, Alexander R. T.; Hecht, Thomas; Gogate, Mandar


Alexander R. T. Gepperth

Thomas Hecht


We present a system for performing multi-sensor fusion that learns from experience, i.e., from training data and propose that learning methods are the most appropriate approaches to real-world fusion problems, since they are largely model-free and therefore suited for a variety of tasks, even where the underlying processes are not known with sufficient precision, or are too complex to treat analytically. In order to back our claim, we apply the system to simulated fusion tasks which are representative of real-world problems and which exhibit a variety of underlying probabilistic models and noise distributions. To perform a fair comparison, we study two additional ways of performing optimal fusion for these problems: empirical estimation of joint probability distributions and direct analytical calculation using Bayesian inference. We demonstrate that near-optimal fusion can indeed be learned and that learning is by far the most generic and resource-efficient alternative. In addition, we show that the generative learning approach we use is capable of improving its performance far beyond the Bayesian optimum by detecting and rejecting outliers and that it is capable to detect systematic changes in the input statistics.

Journal Article Type Article
Acceptance Date Feb 19, 2016
Online Publication Date Mar 14, 2016
Publication Date 2016-10
Deposit Date Apr 26, 2022
Journal Cognitive Computation
Print ISSN 1866-9956
Electronic ISSN 1866-9964
Publisher Springer
Peer Reviewed Peer Reviewed
Volume 8
Issue 5
Pages 806-817
Keywords Cognitive Neuroscience; Computer Science Applications; Computer Vision and Pattern Recognition
Public URL

You might also like

Downloadable Citations