Optimum-Path-to-Go Guidance of Command Adjusted Trajectory Projectiles.
Abstract
Execution of optimal guidance and control algorithms in real time with a microprocessor is feasible using polynomial networks to store, in compact and almost instantly retrievable form, information about a large number of pre-computed optimum two-point boundary-value (TPBV) trajectory solutions. The mathematical basis for computing these optimum solutions has been derived in this SBIR Phase I work using the calculus of variations to obtain adjoint differential equations that are initialized for the TPBV solution. Historically, these initializations have usually necessitated iterative computations because the adjoint equations are not analytically integrable. It is shown that polynomial networks fitted, off-line, to a data base of optimum trajectories can provide the appropriate initializations in real time. These polynomial networks offer optimum-path-to-go (OPTG) and target motion prediction capabilities that can significantly improve performance of maneuverable projectiles, particularly in cases of thruster impulse limitations and substantial uncertainties about target behavior. This report presents the derivation of the governing variational equations, a simulation for three-degree-of-freedom trajectory guidance analyses of command adjusted trajectory (CAT) projectiles engaging maneuvering targets, and results for a 120mm CAT projectile. The results show that the blended guidance law has a smaller miss distance and greater divert capability than the baseline guidance, and that pure OPTG guidance requires fewer divert squib firings than the baseline guidance.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 22, 1988
- Accession Number
- ADA193377
Entities
People
- John F. Elder Iv
- Roger L. Barron