Verifying Plans and Scripts for Robotics Tasks Using Performance Level Profiles

Alexander Kovalchuk, Shashank Shekhar, Ronen I. Brafman

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Performance-Level Profiles (PLPs) were introduced as a type of action representation language suitable for capturing the behavior of functional code for robotics. This paper considers two issues that PLPs raise: (1) Their formal semantics. (2) How to verify a script or plans that schedule the use of components that have been documented by PLPs. We discuss formal semantics for PLPs that maps them to probabilistic timed automata (PTAs). We also show how, given a script that refers to components specified using PLPs, we derive a PTA specification of the entire system. Using a model checker, we can now verify various properties of the system and answers queries about its behavior. Finally, we empirically evaluate an implemented system based on these ideas that use the UPPAAL-SMC model checker and demonstrate its scalability. The result is a pragmatic approach for verifying various properties of component-based robotic systems.

Original languageEnglish
Title of host publication31st International Conference on Automated Planning and Scheduling, ICAPS 2021
EditorsSusanne Biundo, Minh Do, Robert Goldman, Michael Katz, Qiang Yang, Hankz Hankui Zhuo
PublisherAssociation for the Advancement of Artificial Intelligence
Pages673-681
Number of pages9
ISBN (Electronic)9781713832317
StatePublished - 2021
Event31st International Conference on Automated Planning and Scheduling, ICAPS 2021 - Guangzhou, Virtual, China
Duration: 2 Aug 202113 Aug 2021

Publication series

NameProceedings International Conference on Automated Planning and Scheduling, ICAPS
Volume2021-August
ISSN (Print)2334-0835
ISSN (Electronic)2334-0843

Conference

Conference31st International Conference on Automated Planning and Scheduling, ICAPS 2021
Country/TerritoryChina
CityGuangzhou, Virtual
Period2/08/2113/08/21

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Information Systems and Management

Fingerprint

Dive into the research topics of 'Verifying Plans and Scripts for Robotics Tasks Using Performance Level Profiles'. Together they form a unique fingerprint.

Cite this