A Minimax Filter for Systems with Large Plant Uncertainties Using Measurements Corrupted by Colored Noise

A minimax filter is derived to estimate the state of a system, using observations corrupted by colored noise, when large uncertainties in the plant dynamics and process noise are present.     

Analysis of generalized guidance laws for homing missiles

It is shown that the optimal guidance problem can be handled as a Bolza-type problem which can be solved by straightforward calculation. The solution proves to be different from that given by Yang, et al. (1989) and is verified to be a minimum performance index solution. Also presented is an alternate unified approach to the solution of the nonlinear system of differential equation of the generalized proportional navigation law discussed earlier by Yang, et al. This unified equation has been solved for all the different cases of proportional navigation for which closed-form solutions have been published by earlier authors     

Minimax Approach to Trajectory Optimization of Multistage Launch Vehicles

An optimum pitch steering program is developed, using a minimax technique, for a multistage launch vehicle in the presence of large parameter uncertainties. The uncertainties are characterized by deterministic bounds on the respective parameters. The pitch steering program is obtained by maximizing two independent scalar performance ance indexes. 1) the coast apogee velocity for a specified altitude; 2) the perigee of the satellite orbit. The product of dynamic pressure and angle of attack is constrained so as to minimize the structural loads during the atmospheric flight. The values of the uncertain parameters are determined by minimizing the same performance indexes in order to achieve a worst case design. The existence of saddle point solution to this class of problems is shown using the techniques of differential game theory. The conjugate gradient algorithm has been used for computer er aided design of the minimax technique.     

Sensitivity Considerations in Trajectory Optimization

A technique by which the trajectory optimization problem can be formulated to include the trajectory sensitivity functions in the performance index is presented. It is shown that an explicit steering law, which can be derived for the upper atmospheric flight of a vehicle, is a function of the sensitivity state, adjoint vectors, and the parameters of the chosen trajectory dynamics. The new steering law is compared with the one without sensitivity considerations. A computational method is presented to implement the new steering law.