In this paper, we focus on the "Tidy-up" robotic service which needs extremely complex and enormous kinds of knowledge about objects, environments, places, and user's status. Especially, we have tried to find the way for specifying of appropriate places of daily-life objects for the "Tidy-up" robotic service. As the first phase, we categorized daily-life objects into three categories: Trashable, Reusable, and Washable. In the specification of an appropriate place, there are two methods: object category based and individual object trajectory based. For the first method, object state machine has been provided for each object category. In the second method, the object trajectories in term of places have been collected. Based on these historical trajectories, we could recognize the trajectory pattern and use it to provide and appropriate place for individual object.