Assured runtime monitoring and planning: Toward verification of neural networks for safe autonomous operations

Esen Yel, Taylor J. Carpenter, Carmelo Di Franco, Radoslav Ivanov, Yiannis Kantaros, Insup Lee, James Weimer, Nicola Bezzo

Research output: Contribution to journalArticlepeer-review

14 Scopus citations

Abstract

Autonomous systems operating in uncertain envi ronments under the effects of disturbances and noises can reach unsafe states even while using finetuned controllers and precise sensors and actuators. To provide safety guarantees on such systems during motion planning operations, reachability analysis (RA) has been demonstrated to be a powerful tool. RA, however, suffers from computational complexity, especially when dealing with intricate systems characterized by high-order dynamics, making it hard to deploy for runtime monitoring.

Original languageEnglish
Article number9068251
Pages (from-to)102-116
Number of pages15
JournalIEEE Robotics and Automation Magazine
Volume27
Issue number2
DOIs
StatePublished - Jun 2020

Fingerprint

Dive into the research topics of 'Assured runtime monitoring and planning: Toward verification of neural networks for safe autonomous operations'. Together they form a unique fingerprint.

Cite this