Home // VALID 2023, The Fifteenth International Conference on Advances in System Testing and Validation Lifecycle // View article

In-Service Monitoring and Assessment (ISMA) of Autonomous Driving Vehicles with AI based Algorithms

Authors:
Bangarevva Patil
Thomas Corell
Rudra Narayan Hota

Keywords: verification and validation; in-service monitoring and assessment; silent testing; artificial intelligence functions.

Abstract:
Autonomous Driving technology is anticipated to be a key aspect for achieving a higher level of road safety and efficient mobility. A major challenge for this automation is to verify and validate safety aspects of Highly Autonomous Vehicles appropriately. Due to high system complexity and costs, it requires an exponential increase of test efforts for real world testing. Use of In-service Monitoring and Assessment (ISMA) system can efficiently reduce the verification and validation effort with respect to both cost and time. ISMA is an approach to ensure safety of an Autonomous Driving vehicle during the entire product life cycle by continuously intelligently monitoring and evaluating Autonomous Driving functionality during operation. In this paper, we propose an in-service monitoring and assessment concept with a set of exemplary implementations of Artificial Intelligence based algorithms. This concept aims to identify leading indicators for safety critical scenarios and addresses the assumptions made during the development. The research results demonstrate that the proposed approach can safely and efficiently monitor Autonomous Driving functions in both offline and online mode, and helps to discover critical and unknown unsafe scenarios even before an accident occurs.

Pages: 7 to 12

Copyright: Copyright (c) IARIA, 2023

Publication date: November 13, 2023

Published in: conference

ISSN: 2308-4316

ISBN: 978-1-68558-101-5

Location: Valencia, Spain

Dates: from November 13, 2023 to November 17, 2023