02559naa a2200217 a 450000100080000000500110000800800410001910000130006024500640007326000090013730000160014652018770016265000280203965000160206765300290208365300140211265300180212670000110214470000150215577301710217010055142011-03-17       1997    bl uuuu     u00u1 u    #d1 aPENG, T.  aProbabilistic temporal reasoning in the situation calculus.  c1997  ap. 264-273.  aTheories of action and change usually involve reasoning about time and uncertainty with regard to the truth of propositions and the occurrences of actions. In most situations, however, there is not enough information to make certain predictions. When agents devise plans for execution in the real world, they face two forms of uncertainty: they can never have complete knowledge about the situation of the real world, and they do not have complete control, as the effects of actions are uncertain, and the temporal expression of knowledge: when some propositions hold true and how long they persist. While most classical (planning) approaches avoid explicit uncertainty reasoning and the methods based on situation calculus avoid temporal information as well, we believe that uncertainty should be explicitly represented and reasoned about, and the techniques for temporal information are needed. This paper presents a formalism for temporal and probabilistic reasoning in the framework of the situation calculus. The ontology of the situation calculus is enriched by introducing a new sort for times. Starting from a given situation, an action may perform over some special time to change the world into different resulting situations. The uncertainty of the occurrences of actions is captured by a probability distribution over the possible actions with regard to the corresponding situation. The formalism allows representation of the probability that relevant fluents will hold at a specified time, and provides a mechanism with quantitative answers to questions such as "Starting from a given situation, with what probability will the world change into situation sit after some actions perform sequentially?", etc. Also, this paper presents a discussion on the frame problem form uncertain actions and provides a simple solution (sometimes) in the extended formalism.  aArtificial intelligence  aProbability  aInteligência artificial  aOntologia  aProbabilidade1 aMA, J.1 aKNIGHT, B.  tIn: CONGRESO INTERNACIONAL DE INGENIERIA INFORMÁTICA, 3., 1997, Buenos Aires. Proceedings... Buenos Aires: Universidad de Buenos Aires, Facultad de Inginieria, 1997.