Hybrid Direct-Indirect Strategy for Optimal Landing Guidance of Reusable Rockets

Abstract

The present work focuses on developing a fast and accurate algorithm to find feasible trajectories for reusable rockets’ landing while minimizing their fuel consumption. Trajectory exploits aerodynamic forces, thus an Optimal Aerodynamic Powered Landing Problem is faced. A hybrid strategy is adopted, combining convex direct optimization with a novel indirect collocation scheme. A Covector Mapping Theorem is exploited to bridge the two methods. Development of the algorithm is organized in two steps: firstly, the structure of the optimal solution is derived solving the problem with a single shooting indirect method combined with a double homotopic continuation scheme; in second instance, an algorithm tailored on the optimal solution structure is presented and discussed. The suggested strategy is finally compared with the homotopic continuation scheme considering accuracy and computational times. Outcome is a net superiority of the designed algorithm over the homotopic technique; the power of a hybrid approach is therefore demonstrated over traditional solution methods.