Entry Trajectory Optimization for Cross-Domain Morphing Vehicles by Adaptive Trust-Region Sequential Convex Programming

This paper proposes a virtual-control-based adaptive trust-region sequential convex programming (VATSCP) method for the entry trajectory optimization of cross-domain morphing vehicles (CDMVs). By defining new control and state variables, decoupling of state and control in complex dynamics is achieved, avoiding potential jitters in control profiles. To overcome the artificial infeasibility problem in CDMV entry trajectory optimization, virtual control is introduced to relax the aircraft dynamics, and a penalization method to regulate dynamic relaxation is proposed to expand the feasible set and improve the robustness of the algorithm. On this basis, an adaptive trust region strategy is further proposed to dynamically adjust the trust region to improve the convergence speed of the algorithm. The simulation results show that the proposed method is robust and effective. Compared to the fixed configuration aircraft, CDMV has better flight performance.