Learning action durations from executions
Learning action durations from executions
复制标题
从执行中学习动作持续时间
DOI：
--
复制
发表时间：
2007
期刊：
影响因子：
--
通讯作者：
D. Borrajo
中科院分区：
文献类型：
--
作者：
D. Borrajo
准确的动作模型对于有效地解决自动规划任务至关重要。一个精确的行动模型允许计划者精确地预见在给定环境中执行行动的后果，从而找到稳健和高质量的计划。但是，当在真实的世界中处理计划任务时，即使手工编码一个简单的BSPS动作模型也是复杂的，因此定义捕捉进一步特征（如执行持续时间或成本）的动作模型变得更加困难。此外，如果这些特征可以在给定时刻被捕获，则它们可以随时间而变化。在本文中，我们自动建模的行动执行的持续时间，从观察计划执行的关系回归树。我们展示了规划者如何将这些模型纳入他们的域定义后，找到更好的计划。
Accurate action models are essential for efficiently solving automated planning tasks. An accurate action model allow the planner to precisely foresee the consequences of executing actions in a given environment and therefore tofind robust and good quality plans. But when addressing planning tasks in the real world, even hand-coding a simple STRIPS action model is complex, thus defining action models capturing further features, like the execution duration or costs, becomes more difficult. Moreover, if these features can be captured at a given instant they may vary over time. In this paper we automatically model the duration of action execution as relational regression trees learned from observing plan executions. And we show how planners find better plans after incorporating these models to their domain definition.