A Real-Time Novelty Detector for a Mobile Robot

TL;DR

This paper introduces a real-time novelty detection algorithm for mobile robots that learns environmental normality from sonar scans, highlighting new stimuli and adapting over time to changing features.

Contribution

It presents a novel online learning algorithm for environmental novelty detection using sonar data, with the ability to forget and recover features over time.

Findings

Effective detection of novel environmental features in real-time

The model adapts to environmental changes by forgetting and relearning features

Successful robot experiments demonstrating the algorithm's operation

Abstract

Recognising new or unusual features of an environment is an ability which is potentially very useful to a robot. This paper demonstrates an algorithm which achieves this task by learning an internal representation of `normality' from sonar scans taken as a robot explores the environment. This model of the environment is used to evaluate the novelty of each sonar scan presented to it with relation to the model. Stimuli which have not been seen before, and therefore have more novelty, are highlighted by the filter. The filter has the ability to forget about features which have been learned, so that stimuli which are seen only rarely recover their response over time. A number of robot experiments are presented which demonstrate the operation of the filter.