Id: arxiv/2607.02750 Url: https://arxiv.org/abs/2607.02750 Tags: math.OC, eess.SY Score: 0 Title: Online Modeling and Sequential Convex Programming for Lunar Landing Trajectory Optimization Domain: math.OC Source: arxiv Authors: Zhenbo Wang Reviews: 0 Subject: math.OC Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-07-02 Page: https://delphi.science/r/arxiv/2607.02750 This paper presents a guidance framework for lunar powered descent and landing that combines sequential convex programming (SCP) with real-time online model identification. A nonconvex energy-optimal landing problem is developed and then reformulated into a sequence of convex second-order cone programs (SOCPs) through a change of variables, successive linearization, and a lossless second-order cone relaxation of the thrust direction constraint. An online identification layer, built from a recursive least squares (RLS) filter with exponential forgetting and an exponential moving average (EMA)…