Eugene Cliff

Dynamics and Control

1990-present, Reynolds Metals Professor, 1980-1990, Professor, 1975-1980, Associate Professor, 1971-1975, Assistant Professor 1969-1971, Aerospace and Ocean Engineering, Virginia Polytechnic Institute and State University; Visiting Assistant Professor, Aerospace and Mechanical Engineering, University of Arizona.

AOE Honorifics Committee, College Committee on Post Tenure Review, College Computing Committee, Dean's Planning Committee, Associate Director of Interdisciplinary Center for Applied Mathematics, Minority Workshop Facilitator; Virginia Polytechnic Institute and State University; Associate Editor of J. Guidance Control and Dynamics; Panel Member of NSF Graduate Fellowship Program; Reviewer for Journal of Guidance, Control and Dynamics, Journal of Optimization, Theory and Applications, and others; Associate Fellow of AIAA; Senior Member of IEEE; SIAM.

Trajectory Optimization

A variety of mathematical models have been used to study trajectory optimization for atmospheric flight vehicles. In many cases, however, the optimal-trajectory characteristics revealed by simple models have serious shortcomings as approximations to the characteristics for more complex models. In recent years studies of supermaneuverable aircraft has led to interest in models which include rigid-body dynamics. Research is directed at understanding and interpreting these phenomena. Broader interests include numerical optimization procedures for a variety of atmospheric and exoatmospheric flight problems.

Optimal Design

While there has been considerable progress in aerodynamic prediction codes there is continuing need for design tools. Some approaches to optimization-based design employ an approximate-then-optimize approach. There is research interest in methods that retain the continuum model and introduce numerical approximation later. In sensitivity equation methods we derive continuum models for the ``partial-derivative'' of the flow with respect to a design parameter. Numerical solutions for these sensitivities can often be accomplished by modifying the underlying analysis code. All-at-once methods derive on optimality system for the continuum problem and then seek numerical solutions. Adjoint methods use optimality condition to describe the sensitivity of cost and/or constraint functionals.