Abstract:
Conditions are investigated for local convergence of a decompositional algorithm used in solving optimal control problems and suggested earlier [1, 2]. An algorithm modification is described which expands its application area and makes use of ideas in design of augmented Lagrangean functions.