Generalized Dynamic Cognitive Hierarchy Models for Strategic Driving Behavior

While there has been an increasing focus on the use of game theoretic models for autonomous driving, empirical evidence shows that are still open questions around dealing with challenges common knowledge assumptions as well modeling bounded rationality. To address some these practical challenges, we develop a framework generalized dynamic cognitive hierarchy both modelling naturalistic human driving behavior planning vehicles (AV). This is built upon rich model level-0 through automata strategies, interpretable notion rationality safety and maneuver satisficing, robust response planning. Based evaluation two large datasets simulation critical traffic scenarios, show i) strategies suited in level-k framework, ii) proposed to heterogeneous population strategic non-strategic reasoners can be effective approach AV.