Encoding Defensive Driving as a Dynamic Nash Game
This work addresses safety in multi-agent robotic systems, offering a less conservative and more strategic alternative to existing methods, though it appears incremental in its application to specific scenarios.
The paper tackles the problem of ensuring safe robot navigation among multiple agents by proposing a novel robustness formulation based on general-sum dynamic game theory, inspired by defensive driving, and demonstrates its effectiveness in various traffic scenarios.
Robots deployed in real-world environments should operate safely in a robust manner. In scenarios where an "ego" agent navigates in an environment with multiple other "non-ego" agents, two modes of safety are commonly proposed -- adversarial robustness and probabilistic constraint satisfaction. However, while the former is generally computationally intractable and leads to overconservative solutions, the latter typically relies on strong distributional assumptions and ignores strategic coupling between agents. To avoid these drawbacks, we present a novel formulation of robustness within the framework of general-sum dynamic game theory, modeled on defensive driving. More precisely, we prepend an adversarial phase to the ego agent's cost function. That is, we prepend a time interval during which other agents are assumed to be temporarily distracted, in order to render the ego agent's equilibrium trajectory robust against other agents' potentially dangerous behavior during this time. We demonstrate the effectiveness of our new formulation in encoding safety via multiple traffic scenarios.