Towards the disruption of plans

Abstract

In order for an agent or a group of agents (such as a team) to achieve a goal, a sequence of actions have to be performed. These actions bring about state transitions that constitute a plan. Multiple ways of achieving the goal may exist. In some situations, one may want to prevent or delay an agent or group of agents from achieving a goal. We argue that plans can be disrupted by preventing particular state transitions from happening. We propose four algorithms to identify which state transitions should be thwarted such that the achievement of the goal is prevented (total disruption) or delayed (partial disruption). In order to evaluate the performance of our algorithms we define disruption (partial and total) and also provide metrics for its measurement. We do acknowledge that the disruptor may not always have an accurate representation of the disruptee’s plans. Thus, we perform an experimental analysis to examine the performance of the algorithms when some of the state transitions available to the disruptee are unknown to the disruptor.