A Learning Framework for Versatile STL Controller Synthesis
| Autor: | V. Dimarogonas, Péter Várnai, Dimos |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
0209 industrial biotechnology
Task specifications Artificial neural network Controllers Computer science Atomic propositions Control (management) Control policy Controller synthesis Sample (statistics) Input energy Temporal logic 02 engineering and technology Control Engineering Temporal specification Task (project management) Specifications Learning frameworks 020901 industrial engineering & automation Signal temporal logic Computer engineering Target cost Control theory Reglerteknik Energy (signal processing) |
| Zdroj: | 2019 IEEE 58th Conference on Decision and Control (CDC) CDC |
| ISSN: | 2020-1125 |
| DOI: | 10.1109/cdc40024.2019.9029727 |
| Popis: | In this paper, we aim towards providing a practical framework for learning to satisfy signal temporal logic (STL) task specifications for systems with partially unknown dynamics. We consider STL tasks whose satisfaction can be guaranteed by enforcing a priori known temporal specifications imposed on the atomic propositions that compose them. First, a neural network is trained offline as a control policy to satisfy such temporal specifications while also minimizing a target cost, such as the input energy of the system. The obtained controller then serves as a guide that aids exploration while learning to satisfy any specific STL task optimally using policy improvement, greatly increasing the sample efficiency of the procedure. The promise of the approach towards a versatile STL learning framework is demonstrated through simulations. QC 20201125 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|