Dynamic Bayesian Approach for decision-making in Ego-Things

TL;DR

This paper introduces a dynamic Bayesian framework combining multisensory data, feature selection, and neural gas clustering with Markov Jump particle filtering for improved abnormality detection and decision-making in dynamic systems.

Contribution

It proposes a novel integrated approach using neural gas clustering and Markov Jump particle filtering for enhanced abnormality detection and feature selection in dynamic systems.

Findings

Effective abnormality detection in vehicle data

Improved feature selection for prediction accuracy

Validated on real vehicle dataset

Abstract

This paper presents a novel approach to detect abnormalities in dynamic systems based on multisensory data and feature selection. The proposed method produces multiple inference models by considering several features of the observed data. This work facilitates the obtainment of the most precise features for predicting future instances and detecting abnormalities. Growing neural gas (GNG) is employed for clustering multisensory data into a set of nodes that provide a semantic interpretation of data and define local linear models for prediction purposes. Our method uses a Markov Jump particle filter (MJPF) for state estimation and abnormality detection. The proposed method can be used for selecting the optimal set features to be shared in networking operations such that state prediction, decision-making, and abnormality detection processes are favored. This work is evaluated by using a real dataset consisting of a moving vehicle performing some tasks in a controlled environment.