Convex Predictor–Corrector Aerocapture Guidance

Existing aerocapture predictor–corrector guidance algorithms apply significant simplifications to the problem in order to reduce the computational complexity involved in the control plan correction phase. Such simplifications include using a reduced set of equations of motion, exclusion of path constraints, the use of parametric control profiles, and the requirement of separate lateral logic to determine the bank-angle direction. This work presents an advanced predictor–corrector aerocapture guidance algorithm based on sequential convex programming, which enables the inclusion of system-level constraints, namely, peak aerodynamic loading and convective heat load, and eliminates the need for a lateral-logic law while maintaining a feasible level of computational expense. The developed algorithm is dubbed the “convex predictor–corrector aerocapture guidance” (CPAG) algorithm. CPAG’s performance is compared to the state-of-the-art fully numerical predictor–corrector aerocapture guidance (FNPAG) algorithm for aerocapture at Neptune, Mars, and Earth. Monte Carlo analyses indicate that CPAG produces similar in-plane [Formula: see text] requirements and improved out-of-plane [Formula: see text] requirements relative to FNPAG. Additionally, CPAG is able to enforce aerodynamic and aerothermodynamic constraints, even in the presence of uncertainty, while FNPAG is unable to do so. Finally, the use of an energy-based apoapsis targeting expression that can account for perturbations from nonspherical gravitational fields is presented, and its predictions for the required in-plane [Formula: see text] are compared to standard Keplerian expressions. The energy-based expression for targeting an apoapsis radius relaxes the Keplerian motion assumption and improves the accuracy in [Formula: see text] predictions by almost 25%.