RESEARCH ON OPTIMIZATION THEORY AND THE CONTROL OF STOCHASTIC SYSTEMS.
Abstract
Work is reported on a general theory of operator equations which include, as special cases, ordinary differential equations, Volterra integral equations, and functional differential equations with retardations. The results obtained made it possible to look at optimal control problems from a novel and conceptually very advantageous point of view, and to obtain a Pontryagin type maximum principle for control problems with a wide class of 'dynamics'. A computationally simple approximation to the minimum energy controller for a linear system was achieved by reducing the complexity of, the mathematical model of the system. After specifying the criterion for identifying an optimal model of lower order, the required model boundary conditions were deduced. Controller performance was expressed as a scalar valued functional of the model coefficient matrices, and a computational algorithm was suggested for evaluating the best model. The suboptimal control was then expressed in terms of the derived model parameters. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1969
- Accession Number
- AD0700148
Entities
People
- Lucien W. Neustadt
Organizations
- University of Southern California