文章基本信息
- 标题：Taming the Complexity of Timeline-Based Planning over Dense Temporal Domains
- 本地全文：下载
- 作者：Laura Bozzelli ; Angelo Montanari ; Adriano Peron 等
- 期刊名称：LIPIcs : Leibniz International Proceedings in Informatics
- 电子版ISSN：1868-8969
- 出版年度：2019
- 卷号：150
- 页码：1-14
- DOI：10.4230/LIPIcs.FSTTCS.2019.34
- 出版社：Schloss Dagstuhl -- Leibniz-Zentrum fuer Informatik
- 摘要：The problem of timeline-based planning (TP) over dense temporal domains is known to be undecidable. In this paper, we introduce two semantic variants of TP, called strong minimal and weak minimal semantics, which allow to express meaningful properties. Both semantics are based on the minimality in the time distances of the existentially-quantified time events from the universally-quantified reference event, but the weak minimal variant distinguishes minimality in the past from minimality in the future. Surprisingly, we show that, despite the (apparently) small difference in the two semantics, for the strong minimal one, the TP problem is still undecidable, while for the weak minimal one, the TP problem is just PSPACE-complete. Membership in PSPACE is determined by exploiting a strictly more expressive extension (ECA^+) of the well-known robust class of Event-Clock Automata (ECA) that allows to encode the weak minimal TP problem and to reduce it to non-emptiness of Timed Automata (TA). Finally, an extension of ECA^+ (ECA^{++}) is considered, proving that its non-emptiness problem is undecidable. We believe that the two extensions of ECA (ECA^+ and ECA^{++}), introduced for technical reasons, are actually valuable per sé in the field of TA.
- 关键词：Timeline-based planning; timed automata; event-clock automata
