Goal-Driven Autonomy with Case-Based Reasoning

Abstract

Abstract. The vast majority of research on AI planning has focused on automated plan recognition, in which a planning agent is provided with a set of inputs that include an initial goal (or set of goals). In this context, the goal is presumed to be static; it never changes, and the agent is not provided with the ability to reason about whether it should change this goal. For some tasks in complex environments, this constraint is problematic; the agent will not be able to respond to opportunities or plan execution failures that would benefit from focusing on a different goal. Goal driven autonomy (GDA) is a reasoning framework that was recently introduced to address this limitation; GDA systems perform anytime reasoning about what goal(s) should be satisfied [4]. Although promising, there are natural roles that case-based reasoning (CBR) can serve in this framework, but no such demonstration exists. In this paper, we describe the GDA framework and describe an algorithm that uses CBR to support it. We also describe an empirical study with a multiagent gaming environment in which this CBR algorithm outperformed a rule-based variant of GDA as well as a non-GDA agent that is limited to dynamic replanning.

Case-Based Reasoning Research and Development

“…Most GDA agents (e.g., CB-gda (Jaidee et al, 2010)) are given knowledge about state expectations, discrepancies, goals to achieve, and the means to achieve the goals (e.g., the plans or the policies). While some prior work has focused on learning some of these, GRL is the first GDA agent to learn all of them simultaneously.…”

Goal‐driven Autonomy for Responding to Unexpected Events in Strategy Simulations

“…This reduces the knowledge engineering tasks for system designers to either annotating expert gameplay traces or simply collecting them. For example, CB-gda uses observed discrepancies as the retrieval cue to select task goals in a team shooter game (Muñoz-Avila et al 2010). EISBots performed competently against the built-in AI of Starcraft by selecting goal states using the current state as the retrieval cue from a library of game play traces (Weber, Mateas, and Jhala 2010).…”

Learning and Reusing Goal-Specific Policies for Goal-Driven Autonomy

“…GDA agents use a four-step strategy to respond competently to unexpected situations in their environment: (1) detect any discrepancy between the observed state and the expected state(s), (2) explain this discrepancy, (3) formulate a goal to resolve it (if needed), and (4) manage this new goal along with its pending goals (Molineaux et al, 2010;Muñoz-Avila et al, 2010). In step 3, these agents use a variety of models to formulate new goals.…”