Robust ascent trajectory design and optimization of a typical launch vehicle

Abstract

Robustness and reliability of the designed trajectory are crucial for flight performance of launch vehicles. In this paper, robust trajectory design optimization of a typical LV is proposed. Two formulations of robust trajectory design optimization problem using single-objective and multi-objective optimization concept are presented. Both aleatory and epistemic uncertainties in model parameters and operational environment characteristics are incorporated in the problem, respectively. In order to uncertainty propagation and analysis, the improved Latin hypercube sampling is utilized. A comparison between robustness of the single-objective robust trajectory design optimization solution and deterministic design optimization solution is illustrated using probability density functions. The multi-objective robust trajectory design optimization is executed through NSGA-II and a set of feasible design points with a good spread is obtained in the form of Pareto frontier. The final Pareto frontier presents a trade-off between two conflicting objectives namely maximizing injection robustness and minimizing gross lift-off mass of launch vehicle. The resulted Pareto frontier of the multi-objective robust trajectory design optimization shows that with 1% increase in gross mass, the robustness of the design point to the considered uncertainties can be increased about 80%. Also, numerical simulation results show that the multi-objective formulation is a necessary approach to achieve a good trade-off between optimality and robustness.

A Convex Approach for Trajectory Optimization of Super-Synchronous-Transfer-Orbit Large Launch Systems with Time-Position Constraints

“…Among the trajectory optimization algorithms developed in recent years, including indirect methods and direct methods [9][10][11][12][13][14], convex optimization algorithms have great advantages in onboard aerospace applications because the convex optimization problem can be solved in polynomial time with no need for initial guesses supplied by the user [15]. However, most of the original trajectory optimization problems are infinite and nonconvex, which cannot be solved by convex optimization methods directly [16].…”

“…IPM is a typical and widely used algorithm for SOCP. For any given initial guess and accuracy, a globally optimal solution can be found by IPM within the predetermined upper bound of iteration times on condition that the feasible solution exists [10]. To ensure the problem can be solved by IPM, the flip-Radau pseudospectral discretization method is adopted to convert the continuous and infinite problem into a finite one, and the thrust magnitude constraint is relaxed based on lossless convexification.…”

Direct Solution of Multi-Objective Optimal Control Problems Applied to Spaceplane Mission Design

“…[18] performed a multiobjective optimisation of a two stage launcher minimising cost and maximising the payload and Roshanian et al [19] performed robust design optimisation of a two stage launch vehicle by means of multiobjective optimisation, minimising both the mean and the variance of the gross take-off mass when several design and operative parameters where subject to uncertainty. In these last three works the control laws had a simple parametric shape.…”

“…The method proposed in this paper differentiates from Ober-Blöbaum et al [14] and Kaya and Maurer [15] in that it combines a global exploration and local convergence with a smooth transition between Chebyshev and Pascoletti-Serafini scalarisation [16] and incorporates an automatic and unsupervised procedure to generate feasible first guesses. It also differentiates from [17][18][19][20][21] in that it does not use a generic MOGA but proposes a more efficient memetic approach and implements a more general direct transcription method.…”

Online Trajectory Replan for Gliding Vehicle Considering Terminal Velocity Constraint

“…Wang et al [12] proposed the spline-line based energy management guidance for ascending solid rocket vehicles. Ascent trajectory design concept was developed to increase flight performance of launch vehicle [13]. Terminal speed control has also been studied for re-entry vehicles, however, the methods developed are not suitable for ballistic missiles.…”