A fixed-point framework for launch vehicle ascent guidance

Zhang, Lijun
Major Professor
Ping Lu
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Aerospace Engineering
Organizational Unit
Journal Issue
Aerospace Engineering

Recent interests in responsive launch have highlighted the need for rapid and fully automated ascent guidance planning and guidance parameter generation for launch vehicles. This dissertation aims at developing methodology and algorithms for on-demand generation of optimal launch vehicle ascent trajectories from lift-off to achieving targeting conditions outside the atmosphere. The entire ascent trajectory from lift-off to final target point is divided into two parts: atmospheric ascent portion and vacuum ascent portion. The two portions are integrated via a fixed-point iteration based on the continuity condition at the switch point between atmospheric ascent portion and vacuum ascent portion;The previous research works on closed-loop endo-atmospheric ascent guidance shows that the classical finite difference method is well suited for fast solution of the constrained optimal three-dimensional ascent problem. The exploitation of certain unique features in the integration procedure between the atmospheric portion and vacuum portion and the finite difference method, allows us to cast the atmospheric ascent problem into a nested fixed-point iteration problem. Therefore a novel Fixed-Point Iteration algorithm is presented for solving the endo-atmospheric ascent guidance problem. Several approaches are also provided for facilitating the convergence of the fixed-point iteration. The exo-atmospheric ascent portion allows an optimal coast in between the two vacuum powered stages. The optimal coast enables more efficient usage of the propellant. The Analytical Multiple-Shooting algorithm is developed to find the optimal trajectory for this portion;A generic launch vehicle model is adopted in the numerical simulation. A series of open-loop and closed-loop simulations are performed. The results verify the effectiveness, robustness and reliability of the Fixed-Point Iteration (FPI) algorithm and Analytical Multiple-Shooting (AMS) algorithm developed in this research. In comparison to Finite Difference (FD) algorithm, the Fixed-Point Iteration algorithm is more adaptive to the "cold start" case for endoatmospheric ascent guiDance; The simulations also validate the feasibility of the methodology presented in this research in rapid panning and guidance for ascent through atmosphere.