Moving horizon control in dynamic games

A new solution concept based on moving horizon control is introduced in the area of nonzero-sum infinite-horizon differential games. In this concept the players have a feedback information pattern. Aspects of finite-horizon open-loop Nash equilibria are also incorporated in the moving horizon solution concept. The feedback information pattern makes the concept of practical significance and the open-loop elements result in good analytic tractability. Special attention is paid to the class of linear-quadratic games. The analytic computability of the moving horizon solution is illustrated by an analysis of the scalar case and an economic example, (C) 2002 Elsevier Science B.V. All rights reserved.