Automated patch point placement for autonomous spacecraft trajectory targeting

Galen K Harden, Purdue University

Abstract

The lofty goal of autonomous, onboard trajectory design motivates this investigation of automated patch point placement techniques. The motion of a particle affected by the gravitational fields of two massive particles moving on two concentric circular orbits, as defined in the Circular Restricted Three Body Problem (CR3BP), provides the dynamical environment for this study. Differential correctors are employed in converging multiple-segment trajectories subject to continuity and other constraints. A new approach for the Two-Level Corrector is derived and demonstrated. The relationship between Local Lyapunov Exponents (LLE) and the convergence behavior of parallel shooting targeters is explored, then formalized as an error prediction model. Lastly, an automated patch point placement algorithm is developed using that error prediction model. The placement algorithm is not restricted to the current error model, nor is the analysis restricted to the CR3BP. The error prediction model and patch point placement algorithm are applied to initial guesses obtained by randomly perturbing a viable solution, simulating the conditions used in deriving the error model. The patch point placement algorithm is then used in a sample design problem, demonstrating its ability to achieve convergence where arbitrary patch point placement causes the targeter to diverge.

Degree

M.S.A.A.

Advisors

Howell, Purdue University.

Subject Area

Aerospace engineering

COinS