This paper introduces a framework for enabling policy-aware autonomous agents to reason about potential penalties for non-compliant behavior, and act accordingly. We employ the Authorization and Obligation Policy Language ( \(\mathcal {AOPL}\) ) for policy specification and Answer Set Programming (ASP) for reasoning about policies and penalties. We build upon existing work by Harders and Inclezan on simulating the behavior of policy-aware autonomous agents and test our work on two different domains. We conclude that our framework produces higher quality plans than the previous approach. Depending on the nature of the domain, optimal plans may also be computed more efficiently by our framework.