SENSITIVITY ANALYSIS OF OPTIMAL BALLISTIC MISSILE WAR STRATEGIES.
Abstract
A ballistic missile war is modeled by a set of first-order, nonlinear differential equations with time-varying coefficients. The model includes the effects of initial missile stocks, missile firing rate, counterforce and countervalue missile effectiveness. Termination conditions related to the tolerable number of casualties and reduced missile strength are used. The problem of determining optimal wartime strategies, where the enemy strategy is known, is formulated as an optimal control problem and extremal strategies are determined from an application of Pontryagin's Maximum Principle. Strategies for the optimizing country are composed of a missile firing rate and targeting variable and are viewed as control functions. The model and adjoint system dynamics are programmed on an analog computer and extremal control functions are determined by a visual searching technique that is developed from a computer study of an example system. Several assumptions that reduce the complexity of the extremal terminal adjoint variable conditions are used to reduce the number of unknown initial adjoint conditions from four to two. The assumptions on the terminal adjoint conditions and some characteristics of the adjoint dynamics are used to determine a 'playable' set of initial adjoint values from an analog computer study of the example system. The example system is used as a reference model for a sensitivity analysis. The sensitivity of the optimal solution of the reference model to variations in system parameters and initial conditions is determined and a simplified estimate of the cost of these variations is given. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 1970
- Accession Number
- AD0708478
Entities
People
- Michael J. Gunn
Organizations
- University of California, Los Angeles