Convex–concave optimization for a launch vehicle ascent trajectory with chance constraints

Abstract

The objective of this paper is to present a convex–concave optimization approach for solving the problem of a multistage launch vehicle ascent trajectory. The proposed method combines convex–concave decomposition and successive linearization techniques to generate a new sequence of convex subproblems to replace the original non-convex problem. Bernstein approximation is used to transform the chance constraints into convex ones. A hp-adaptive pseudospectral scheme is employed to discretize the optimal control problem into a nonlinear programming problem with less computation cost. The performance of the proposed strategy is compared against other typical techniques in a selection of test case scenarios. Numerical results demonstrate the viability of the method and show pros and cons of the proposed technique.