Diversity in Action: General-Sum Multi-Agent Continuous Inverse Optimal Control
This work addresses the challenge of realistic reward inference and prediction for autonomous vehicles in interactive traffic, though it appears incremental by building on existing game-theoretic approaches.
The paper tackles the problem of modeling interactive traffic scenarios by developing a game-theoretic method for multi-agent continuous inverse optimal control, which infers individual rewards and predicts actions without perfect communication, leading to better alignment with real-world behavior in experiments.
Traffic scenarios are inherently interactive. Multiple decision-makers predict the actions of others and choose strategies that maximize their rewards. We view these interactions from the perspective of game theory which introduces various challenges. Humans are not entirely rational, their rewards need to be inferred from real-world data, and any prediction algorithm needs to be real-time capable so that we can use it in an autonomous vehicle (AV). In this work, we present a game-theoretic method that addresses all of the points above. Compared to many existing methods used for AVs, our approach does 1) not require perfect communication, and 2) allows for individual rewards per agent. Our experiments demonstrate that these more realistic assumptions lead to qualitatively and quantitatively different reward inference and prediction of future actions that match better with expected real-world behaviour.