Co-evolutionary computation for constrained min-max problems and its applications for pursuit-evasion games

The co-evolutionary computation method for solving constrained min-max problems is proposed. Many engineering problems can be practically expressed as constrained min-max problems. Min-max problems have two groups of variables. Each group will minimize or maximize payoffs and is subject to equality and inequality constraints. Lagrange multipliers are implemented for handling constraints. Primal constrained min-max problems are transformed into dual unconstrained min-max problems by using the Lagrange multipliers. The co-evolutionary computation is used for solving the dual min-max problems. The proposed method deals with separable and inseparable constraints of two groups. The proposed algorithm is applied for pursuit-evasion games with various constraints. Numerical results are compared with those of conventional methods.