Combining decision making and dynamical systems for monitoring and executing manipulation tasks
- PDF / 1,152,425 Bytes
- 7 Pages / 613.276 x 859.89 pts Page_size
- 96 Downloads / 174 Views
Elektrotechnik & Informationstechnik https://doi.org/10.1007/s00502-020-00816-7
Combining decision making and dynamical systems for monitoring and executing manipulation tasks M. Saveriano , J. Piater
In this paper, we propose a unified framework for online task scheduling, monitoring, and execution that integrates reconfigurable behavior trees, a decision-making framework with integrated low-level control functionalities, and reactive motion generation with stable dynamical systems. In this way, we realize a flexible and reactive system capable of coping with unexpected variations in the executive context without penalizing modularity, expressiveness, and readability of humans. The framework is evaluated in a simulated sorting task showing promising results in terms of flexibility regarding task scheduling and robustness to external disturbances. Keywords: decision making; task execution monitoring; learning from demonstration; reactive motion planning
Überwachung und Ausführung von Roboter-Tätigkeiten durch kombinierte Entscheidungsfindung mit dynamischen Systemen. Für die Online-Aufgabenplanung von Robotern sowie deren Ausführung und Überwachung haben wir eine vereinheitlichte Architektur entwickelt, die rekonfigurierbare Verhaltensbäume, Entscheidungsfindung mit integrierter Regelungstechnik und reaktive Bewegungserzeugung mit stabilen dynamischen Systemen integriert. Das Ergebnis ist ein flexibles und reaktives System, das unerwartete Schwankungen im Ausführungskontext bewältigen kann, ohne Modularität, Ausdruckskraft oder menschliche Lesbarkeit zu beeinträchtigen. Das System wird mittels einer simulierten Sortieraufgabe evaluiert, die vielversprechende Ergebnisse in Bezug auf Flexibilität bei der Aufgabenplanung und Robustheit gegenüber externen Störungen liefert. Schlüsselwörter: Entscheidungsfindung; Ausführung und Überwachung von Tätigkeiten; Nachahmungslernen; reaktive Bewegungsplanung
Received May 31, 2020, accepted July 16, 2020 © The Author(s) 2020
1. Introduction Robots operating in everyday environments need increased flexibility to cope with sudden and unexpected variations in their executive context. In dynamic environments, the robot needs to perform sensing and planning operations at low cost, as well as to monitor its own actions in a goal-oriented fashion [1, 2]. High flexibility and adaptation capabilities are required at each level of the senseplan-act loop to deploy robust and effective robotic solutions [3]. At the same time, the enhanced reasoning and acting capabilities of cognitive robots have to be implemented considering facets like modularity, reusability, design effectiveness, and human-readability. In this work, we propose a unified framework that combines highlevel decision making, continuous execution monitoring, and online motion generation. In our framework, task scheduling and execution monitoring are handled by Reconfigurable Behaviour Trees (RBTs) [4], while stable dynamical systems are exploited for online motion generation. RBTs extend the traditional behavior trees [2
Data Loading...
