THE APPLICATION OF DYNAMIC PROGRAMMING TO A STOCHASTIC CONTROL PROBLEM.
AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OHIO SCHOOL OF ENGINEERING
Pagination or Media Count:
Dynamic programming is used for optimization of a first-order control system with stochastic inputs. The dynamic programming solutions were compared with known analytical solutions and were found to be accurate, but there were many computational difficulties. The major difficulties encountered were truncation of the returns which limited the number of stages which could be computed, the large computation time and the critical selection of the incremental parameters. A discussion is made of the optimum control system and the characteristics of the disturbance. IBM 7094 Digital Computer Programs are included in the appendicies. Author