Dynamic controllability of simple temporal networks with uncertainty: Simple rules and fast real-time execution

作者: Massimo Cairo , Romeo Rizzi

DOI: 10.1016/J.TCS.2018.11.005

关键词: Simple (abstract algebra)Existential quantificationInterval (mathematics)AlgorithmLocal consistencyEquivalence (measure theory)ControllabilityCompleteness (statistics)Set (abstract data type)Computer science

摘要: Abstract Simple Temporal Networks (STNs) are a well-studied model for representing temporal constraints. They comprise set of time-points (real-valued variables execution times) and binary difference constraints among them. with Uncertainty (STNUs) extend STNs in that some (called contingent) treated as exogenous variables, whose times, bound to fall within given interval from the corresponding activation (as specified by “contingent link”), get revealed only during real-time execution. An STNU is dynamically controllable (DC) if there exists strategy execute its satisfying all constraints, regardless times contingent In this work we present new system constraint propagation rules STNUs, which sound-and-complete DC checking. Our comprises just three which, differently ones proposed previous works, generate unconditioned particular, after applying any our rules, network remains an respects. Moreover, completeness proof short non-algorithmic, based on explicit construction valid strategy. This substantial simplification theory underlies efficient algorithms DC-checking. analysis also shows: (1) existence late strategies (2) equivalence notion several variants semantics (3) fast algorithm runs O ( K N ) total time ≥ 1 links points, considerably improving 3 -time bound.

参考文章(17)
Paul Morris, Dynamic Controllability and Dispatchability Relationships integration of ai and or techniques in constraint programming. pp. 464- 479 ,(2014) , 10.1007/978-3-319-07046-9_33
Jonas Kvarnström, Patrick Doherty, Mikael Nilsson, EfficientIDC: a faster incremental dynamic controllability algorithm international conference on automated planning and scheduling. pp. 199- 207 ,(2014)
Carlo Combi, Luke Hunsberger, Roberto Posenato, An Algorithm for Checking the Dynamic Controllability of a Conditional Simple Temporal Network with Uncertainty - Revisited international conference on agents and artificial intelligence. ,vol. 449, pp. 314- 331 ,(2013) , 10.1007/978-3-662-44440-5_19
Paul Morris, A structural characterization of temporal dynamic controllability principles and practice of constraint programming. pp. 375- 389 ,(2006) , 10.1007/11889205_28
John Stedl, Brian Williams, Julie Shah, Paul Robertson, A fast incremental algorithm for maintaining dispatchability of partially controllable plans international conference on automated planning and scheduling. pp. 296- 303 ,(2007)
Paul Morris, Robert Morris, Francesca Rossi, Lina Khatib, Temporal constraint reasoning with preferences international joint conference on artificial intelligence. pp. 322- 327 ,(2001)
Nicola Muscettola, Paul Morris, Thierry Vidal, Dynamic control of plans with temporal uncertainty international joint conference on artificial intelligence. pp. 494- 499 ,(2001)
Brian Williams, Robert Effinger, Gerard Kelly, Michael Sheeny, Dynamic controllability of temporally-flexible reactive programs international conference on automated planning and scheduling. pp. 122- 129 ,(2009)
Nicola Muscettola, Paul Morris, Temporal dynamic controllability revisited national conference on artificial intelligence. pp. 1193- 1198 ,(2005)
Jonas Kvarnström, Patrick Doherty, Mikael Nilsson, Incremental dynamic controllability revisited international conference on automated planning and scheduling. pp. 337- 341 ,(2013)