Demo abstract: real-time out-of-distribution detection on a mobile robot
In a cyber-physical system such as an autonomous vehicle (AV), machine learning (ML) models can be used to navigate and identify objects that may interfere with the vehicle’s operation. However, ML models are unlikely to make accurate decisions when presented with data outside their training distrib...
Saved in:
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Conference or Workshop Item |
Language: | English |
Published: |
2023
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/162460 http://2022.rtss.org/call-for-demo/ |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-162460 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1624602023-02-04T23:32:30Z Demo abstract: real-time out-of-distribution detection on a mobile robot Yuhas, Michael Easwaran, Arvind Interdisciplinary Graduate School (IGS) School of Computer Science and Engineering IEEE Real-Time Systems Symposium - RTSS@Work 2022 Energy Research Institute @ NTU (ERI@N) Engineering::Computer science and engineering Duckietown Out-of-Distribution Detection In a cyber-physical system such as an autonomous vehicle (AV), machine learning (ML) models can be used to navigate and identify objects that may interfere with the vehicle’s operation. However, ML models are unlikely to make accurate decisions when presented with data outside their training distribution. Out-of-distribution (OOD) detection can act as a safety monitor for ML models by identifying such samples at run time. However, in safety critical systems like AVs, OOD detection needs to satisfy real-time constraints in addition to functional requirements. In this demonstration, we use a mobile robot as a surrogate for an AV and use an OOD detector to identify potentially hazardous samples. The robot navigates a miniature town using image data and a YOLO object detection network. We show that our OOD detector is capable of identifying OOD images in real-time on an embedded platform concurrently performing object detection and lane following. We also show that it can be used to successfully stop the vehicle in the presence of unknown, novel samples. Ministry of Education (MOE) Submitted/Accepted version This research was funded in part by MOE, Singapore, Tier-2 grant number MOE2019-T2-2-040. This research is part of the programme DesCartes and is supported by the National Research Foundation, Prime Minister’s Office, Singapore under its Campus for Research Excellence and Technological Enterprise (CREATE) programme. 2023-01-30T04:29:29Z 2023-01-30T04:29:29Z 2022 Conference Paper Yuhas, M. & Easwaran, A. (2022). Demo abstract: real-time out-of-distribution detection on a mobile robot. IEEE Real-Time Systems Symposium - RTSS@Work 2022, 26-28. https://hdl.handle.net/10356/162460 http://2022.rtss.org/call-for-demo/ 26 28 en MOE2019-T2-2-040 10.21979/N9/INOCLV © 2022 The Author(s). Published by RTSS. All rights reserved.This paper was published in Proceedings of IEEE Real-Time Systems Symposium - RTSS@Work 2022 and is made available with permission of The Author(s). application/pdf |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Computer science and engineering Duckietown Out-of-Distribution Detection |
spellingShingle |
Engineering::Computer science and engineering Duckietown Out-of-Distribution Detection Yuhas, Michael Easwaran, Arvind Demo abstract: real-time out-of-distribution detection on a mobile robot |
description |
In a cyber-physical system such as an autonomous vehicle (AV), machine learning (ML) models can be used to navigate and identify objects that may interfere with the vehicle’s operation. However, ML models are unlikely to make accurate decisions when presented with data outside their training distribution. Out-of-distribution (OOD) detection can act as a safety monitor for ML models by identifying such samples at run time. However, in safety critical systems like AVs, OOD detection needs to satisfy real-time constraints in addition to functional requirements. In this demonstration, we use a mobile robot as a surrogate for an AV and use an OOD detector to identify potentially hazardous samples. The robot navigates a miniature town using image data and a YOLO object detection network. We show that our OOD detector is capable of identifying OOD images in real-time on an embedded platform concurrently performing object detection and lane following. We also show that it can be used to successfully stop the vehicle in the presence of unknown, novel samples. |
author2 |
Interdisciplinary Graduate School (IGS) |
author_facet |
Interdisciplinary Graduate School (IGS) Yuhas, Michael Easwaran, Arvind |
format |
Conference or Workshop Item |
author |
Yuhas, Michael Easwaran, Arvind |
author_sort |
Yuhas, Michael |
title |
Demo abstract: real-time out-of-distribution detection on a mobile robot |
title_short |
Demo abstract: real-time out-of-distribution detection on a mobile robot |
title_full |
Demo abstract: real-time out-of-distribution detection on a mobile robot |
title_fullStr |
Demo abstract: real-time out-of-distribution detection on a mobile robot |
title_full_unstemmed |
Demo abstract: real-time out-of-distribution detection on a mobile robot |
title_sort |
demo abstract: real-time out-of-distribution detection on a mobile robot |
publishDate |
2023 |
url |
https://hdl.handle.net/10356/162460 http://2022.rtss.org/call-for-demo/ |
_version_ |
1757048203653414912 |