Applications of Optimal Stochastic Control Designs to a Ground-Based Tracking System

The use of linear-quadratic stochastic control designs for a ground-based tracking system is investigated. The design approach is motivated by the facts that: (i) the power spectral density of the atmospheric turbulence-induced tilt is known, and can be accurately modeled; and (ii) the angular anisoplanatism effect can be expressed in terms of its temporal correlation. The former generates a stochastic dynamic model that best describes the atmospheric turbulence-induced tilt, and the latter introduces a future state or a delay state depending upon the problem formulation. It is demonstrated that how a design philosophy, taking into account the above available information in its control design, can lead to a closed-loop system with a better tracking performance. For a simple case study, the analysis shows that with the proposed control design, the system could track a much weaker target (has at least 2.5 times less brightness) as compared to the system using the classical control design.