Optimality conditions in a control problem for a linear system with bilinear functional (Q1794384)

In this paper, a nonconvex optimal control problem with a bilinear functional and a linear phase system is studied. Sufficient optimality conditions based on the maximum principle are proved. In this context, the notion of a strongly extremal control that maximizes the Pontryagin function on the set of phase or conjugate trajectories is used. Under the considered aspects, optimality conditions with the help of explicit expressions for extremal values of auxiliary problems are transformed. Consequently, optimality conditions in the desired form and, finally, inequalities with respect to scalar functions on a certain time interval are obtained. This approach reflects the dual character of obtained conditions, i.e. they represent pairs of independent symmetric correlations. Some examples illustrate the advantage of this approach.