| Summary: | Episodic memory is an important component of "minds" of many longliving virtual agents, because equipping such an agent with his personal history increases his e ciency and believability. So far, research on episodic memory modeling in the context of these agents has focused mostly on producing the memory content on-line, that is, when the agent is being simulated. In this work, we address a complementary issue, automatic generation of the memory content off-line. We see a possible need of a tool for generating memories that anticipate the start of the simulation. Hence we created a complex design method enabling a designer to specify high-level requirements on an agent's history and use planning to automatically generate this history according to these requirements. We detail the structure of the high-level language used for the description of the requirements and the part of this method that concerns itself with the planning. In a set of experiments, we tested the performance of several planners on our task and we present here the results we gained.
|
|---|
