As autonomous vehicles populate the road, they need to interact with other autonomous and human drivers. Our goal is to create algorithms for autonomous vehicles that exhibit socially-compliant behavior, both in human-like driving and in interactions with human drivers. Tools from social psychology allow us to predict human driver personalities and improve safety.
We use Social Value Orientation, which quantifies the degree to which humans will cooperate with one another, such as “altruistic,” “prosocial,” “egoistic,” or “competitive.” For autonomous driving, these SVO preferences manifest as different behavior. For instance, an altruistic driver may yield and let you merge, while an egoistic driver may not.