Mixed Strategy Constraints in Continuous Games (2402.17874v2)

Abstract: When modeling robot interactions as Nash equilibrium problems, it is desirable to place coupled constraints which restrict these interactions to be safe and acceptable (for instance, to avoid collisions). Such games are continuous with potential mixed strategy equilibria, and this combination of characteristics means special care must be given to setting coupled constraints in a way that respects mixed strategies while remaining compatible with continuous game solution methods. Here, we investigate the problem of constraint-setting in this context, primarily focusing on a chance-based method. We first motivate these chance constraints in a discrete setting, placing them on n-player matrix games as a justifiable approach to handling the probabilistic nature of mixing. Then, we describe a numerical solution method for these chance constrained, continuous games with simultaneous pure strategy optimization. Finally, using a modified pursuit-evasion game as a motivating example, we demonstrate the actual behavior of this solution method in terms of its fidelity, parameter sensitivity, and efficiency