Return to search

Dynamical formulations and control of an automatic retargeting system

The Poincare equations, also known as Lagrange's equations in quasi coordinates,
are revisited with special attention focused on a diagonal form. The diagonal
form stems from a special choice of quasi velocities that were first introduced by Georg
Hamel nearly a century ago. The form has been largely ignored because the quasi
velocities create so-called Hamel coefficients that appear in the governing equations
and are based on the partial derivative of the mass matrix factorization. Consequently,
closed-form expressions for the Hamel coefficients can be difficult to obtain
and relying on finite-dimensional, numerical methods are unattractive. In this thesis
we use a newly developed operator overloading technique to automatically generate
the Hamel coefficients through exact partial differentiation together with numerical
evaluation. The equations can then be numerically integrated for system simulation.
These special Poincare equations are called the Hamel Form and their usefulness in
dynamic modeling and control is investigated.
Coordinated control algorithms for an automatic retargeting system are developed
in an attempt to protect an area against direct assaults. The scenario is for
a few weapon systems to suddenly be faced with many hostile targets appearing together.
The weapon systems must decide which weapon system will attack which
target and in whatever order deemed sufficient to defend the protected area. This
must be performed in a real-time environment, where every second is crucial. Four different control methods in this thesis are developed. They are tested against each
other in computer simulations to determine the survivability and thought process of
the control algorithms. An auction based control algorithm finding targets of opportunity
achieved the best results.

Identiferoai:union.ndltd.org:tamu.edu/oai:repository.tamu.edu:1969.1/4836
Date25 April 2007
CreatorsSovinsky, Michael Charles
ContributorsHurtado, John E.
PublisherTexas A&M University
Source SetsTexas A and M University
Languageen_US
Detected LanguageEnglish
TypeBook, Thesis, Electronic Thesis, text
Format367172 bytes, electronic, application/pdf, born digital

Page generated in 0.0021 seconds