Trajectory Optimization for Hypersonic Vehicles Under Aerodynamic Uncertainty via Risk-Neutral Sequential Convex Programming

Traditional trajectory optimization methods for hypersonic under aerodynamic uncertainty often lead to overly conservative solutions. To address this issue, this paper proposes a novel trajectory planning method using sequential convex programming (SCP) with chance-constraint in response to aerodynamic uncertainty. The core of our approach is mapping aerodynamic uncertainty to the control input, formulating a tractable probabilistic optimization problem. A risk-neutral surrogate function, named Scaled and Translated Adaptive Proxy (STA-Proxy), is designed to approximate the non-smooth chance constraint. By incorporating an error compensation method, the STA-Proxy function avoids overly conservative solution while maintaining numerical accuracy. The resulting STA-Proxy-SCP algorithm developed can solve the trajectory optimization problem efficiently. Numerical simulation demonstrates that the proposed method outperforms both robust optimization and the Split-Bernstein approach in terms of performance index, thus highlighting its superior balance between computational efficiency and reliability.