WiP Abstract : Robust Out-of-distribution Motion Detection and Localization in Autonomous CPS

TL;DR

This paper introduces a real-time out-of-distribution motion detection and localization framework for autonomous cyber-physical systems, combining classical optic flow with variational autoencoders to improve robustness.

Contribution

It presents a novel OOD detection method that integrates optic flow and VAE for enhanced robustness and includes a technique for localizing OOD factors in images.

Findings

Outperforms related methods in robustness on driving simulation data

Effectively detects unusual movements in real-time

Provides localization of OOD factors in images

Abstract

Highly complex deep learning models are increasingly integrated into modern cyber-physical systems (CPS), many of which have strict safety requirements. One problem arising from this is that deep learning lacks interpretability, operating as a black box. The reliability of deep learning is heavily impacted by how well the model training data represents runtime test data, especially when the input space dimension is high as natural images. In response, we propose a robust out-of-distribution (OOD) detection framework. Our approach detects unusual movements from driving video in real-time by combining classical optic flow operation with representation learning via variational autoencoder (VAE). We also design a method to locate OOD factors in images. Evaluation on a driving simulation data set shows that our approach is statistically more robust than related works.