Christopher W. Geib. The Intentional Planning System: ItPlanS. PhD thesis, University of Pennsylvania, 1995.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....actions and allocating high level goals to the character. This makes possible not only to generate intelligent behaviour, but also to explore the diversity of courses of action. In recent years, several researchers have described the use of planning systems to control characters behaviours. Geib [5] has proposed the use of refinement planning following a detailed study of animation requirements [5] 8] Funge has used situation calculus to generate intelligent behaviours for virtual actors [4] and Cavazza et al. 2] 3] have approached this problem with Hierarchical Task Networks (HTNs) for ....

....behaviour, but also to explore the diversity of courses of action. In recent years, several researchers have described the use of planning systems to control characters behaviours. Geib [5] has proposed the use of refinement planning following a detailed study of animation requirements [5][8] Funge has used situation calculus to generate intelligent behaviours for virtual actors [4] and Cavazza et al. 2] 3] have approached this problem with Hierarchical Task Networks (HTNs) for storytelling, considering the knowledge intensive nature of this kind of applications. The planning ....

Geib, C. The intentional planning system: Itplans. Proceedings of the 2nd Artificial Intelligence Planning Systems Conference, pp. 55-64, 1994

....have been extended to incorporate user intervention. 4. PLAN BASED BEHAVIOURS IN INTERACTIVE STORYTELLING There is a broad agreement on the use of planning techniques for describing high level behaviour of autonomous agents embodied in virtual environments, both for task based simulation [19] [20] and for character based storytelling [12] Our description of characters roles as HTNs naturally led to use these as a starting point for the implementation of a planning system. HTN based planning, also known as task decomposition planning, is among the oldest approaches for providing ....

Geib, C., 1994. The intentional planning system: Itplans. Proceedings of the 2nd Artificial Intelligence Planning Systems Conference, pp. 55-64.

....(i.e. Unreal2 TM) will help representing facial expressions, or detailed non verbal behaviour (e.g. body postures) to improve the dramatisation of events through physical characterisation. Above the planning and interleaving of actions, explicit situated mechanisms for reactive behaviours [11] are needed in order to deal with specific situations (e.g. Ross suddenly meets Rachel on his way) This implies high level action recognition of interactions between characters behaviours. If a narratively meaningful (considering the story genre) situation arises, the mechanism would act on ....

C. Geib, "The Intentional Planning System: ItPlans", Proceedings of the 2nd Artificial Intelligence Planning Conference, AIPS-94, 1994, pp. 55-64.

....sufficient to cope with the executability conditions of actions in a dynamic environment. For instance, Ross might not want to steal Rachel s diary if he can be spotted by Monica. To solve this kind of problem, Geib and Webber have proposed to complement top down planning with situated reasoning [21] [22] This can be successfully applied in conjunction with real time planning: the main plan can be interrupted to cope with situated actions and control later returned to the plan after updating the action pre conditions. For instance, if at an early stage of his plan, Ross bumps into Rachel in ....

....diary but she is using it, he can wait for her to finish (unnoticed by her) rather than find another To a large extent, sub goal independence appears to be a property of narrative representations, though this point deserves further investigation. A similar approach, has been described by Geib [21] as part of his incremental planner ItPlans . However, our system proceeds depth first towards the first executable action. source of information. This form of situated reasoning is based on the duration of actions and the nature of their resources. Fig. 3. Re planning following user ....

Geib, C.: The Intentional Planning System: ItPlans, Proceedings of the 2nd Artificial Intelligence Planning Conference (AIPS-94), pp. 55-64 (1994).

....This point is addressed in the next sub section. 3. 4 Connecting Behaviors to High Level Planners There is a vast literature involving high level planners to guide an animation, and much of it focuses on connecting language to the animation, by interpreting and expanding given text instructions [1, 23, 10, 11]. In particular, Webber et al. [23] identify the limited perception of agents as a main limitation to correctly interpret such text instructions, resulting in a poor knowledge construction. The proposed Smart Object framework can minimize this difficulty by providing in each Smart Object ....

C. W. Geib, "The Intentional Planning System: ItPlanS", Proceedings of AIPS, 1994.

....organizational structure was required. Following the lead of researchers such as Mataric [104] we chose the popular state machine paradigm which successfully introduced the needed intelligence into our system. 2 The Hide and Seek project combines Geib s hierarchical planner, ItPlanS [64], and Moore s search planner [111] with the locomotion control system (LCS) presented in Chapter 7, to produce a fully automatic simulation of the game of Hide and Seek. The system (Zaroff) employs a layered architecture to control player locomotion. LCS controls locomotion directly. The planner ....

Christopher W. Geib. The Intentional Planning System: ItPlanS. PhD thesis, University of Pennsylvania, 1995.

....of the multiple cooperative agents system. As discussed in the previous section, the overall goal of the system is to transport a large object from one place to another while avoiding obstacles and passing through a doorway. Ultimately, this goal will be broken up into subgoals by the task planner [14, 15], though for the current work, we assign the following subtasks manually: ffl It is the responsibility of observation agent A to check the prescribed path for obstacles, and to ensure that the pathway is wide enough at all points to accommodate the two manipulatory agents carrying the object. ....

....items such as pipe, and the navigation necessary to transport items from one location to another. Since humans are better equipped to divide tasks into subtasks, we currently assign this undertaking to the human supervisor, however, we are incorporating the incremental task planner ItPlanS [14, 15] to assist the human supervisor. Application The task level implementations are employed to determine the global plan for the task which is interpreted by the task description translation algorithm (as described in Section 3.2) 2.3.2 Regulation Level There exist minimal interactions which are ....

Christopher W. Geib. The intentional planning system: Itplans. In Proceedings of the Artificial Intelligence Planning Systems Conference, pages 55 -- 64, 1994.

No context found.

Christopher W. Geib. The Intentional Planning System: ItPlanS. PhD thesis, University of Pennsylvania, 1995.

....and visual attention) The actions must be cleanly preemptable, allowing the simulation programmer to pass in a condition for premature termination, which allows an agent to effectively change its mind when certain events occur. Certain complex actions invoke a high level planner called ITPLANS [11, 12]. The planner s function is to expand them into a contextually appropriate structure of more basic actions that can then be carried out, imbuing the agents with a certain level of deliberative intelligence. By altering the logic behind the scheduling of these high level actions, one can ....

C. W. Geib. The Intentional Planning System: ItPlanS. PhD thesis, University of Pennsylvania, 1995.

....about where the other players may be hiding. Adaptivity is also supported through least commitment planning, as the planner only looks ahead one action at each level of its abstraction hierarchy. The implementation follows the two level agent architecture: the Intentional Planning System (ItPlanS)[Geib, 1995] interacts with a Search Planner[Moore, 1993] to perform the Figure 7: The start of a game of hide and seek high level reasoning for the system, and these two components in turn interact with a set of low level SCA nodes based on Stepper (Figs. 7 and 8) 5.1 System Architecture Our division ....

Geib, C.W. (1995) The Intentional Planning System: ItPlanS. PhD thesis, University of Pennsylvania.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC