A New Trust-Region Constraint Method for Sequential Convex Programming Framework for Entry Guidance

This paper introduces an improved method for imposing trust-region constraints within the sequential convex programming framework, particularly aiming to reduce unnecessary iterations in applications for the entry guidance problem. The proposed method leverages the dual solution to detect oscillations in the solution across iterations and to determine whether the solution is near optimal. Subsequently, a penalized form of the trust-region constraint is applied to enhance convergence and accurately satisfy the nonlinear constraints. Numerical simulations on the entry guidance problem are presented to validate the performance of the proposed method. The results demonstrate a significant reduction in the number of iterations, with only a minimal trade-off in cost functional value and convergence success rate.