Optimisation in Aerospace Engineering

WC-04: Optimisation in Aerospace Engineering
Stream: Optimisation in Aerospace Engineering
Room: Lagrange
Chair(s): Edmondo Minisci

Minimum-fuel Low-thrust Earth-Moon Transfers in the Sun-Earth-Moon System
Richard EPENOY
In this work, a minimum-fuel optimal control problem will be stated subject to the dynamics of the Sun-Earth-Moon Bicircular Restricted Four-Body Problem. To overcome the huge sensitivity of the problem, a massive derivative-free exploration will be used to find the zeros of the associated shooting function. Different families of medium-duration transfers will be identified so as long-duration trajectories exploiting the counterparts of invariant manifolds defined in both Sun-Earth and Earth-Moon Circular Restricted Three-Body Problems leading to so-called low-energy transfers.

Analysis of Nature Inspired Methods for Interplanetary Trajectory Problems
Edmondo Minisci, Gabriela Ochoa
Many nature inspired optimisers have been used in the past decades to solve interplanetary trajectory problems. These problems are thought to have funnel or multi funnel objective function landscapes, but only few relevant works have tried to properly characterise the landscape and, more importantly, link the shape of the objective functions with the performance of the metaheuristics and heuristics. To contribute to this research area, a Local Optima Networks will be used to analyse, visualise and compare the behaviour of different types of solvers.

Interoperating direct and indirect solvers in optimal control
Olivier Cots, Jean-Baptiste Caillau, Pierre Martinon
Methods for solving optimal control problems fall into two main categories. Indirect methods based on Pontrjagin Principle are fast and accurate but require more work, in terms of mathematical analysis and a priori knowledge of structure of the solution, than direct transcription approaches that offer a good tradeoff between robustness and accuracy. For challenging problems, one may start with a direct method to find a rough solution, then refine it through an indirect method. We discuss this combined approach on applications and the interfacing between two solvers: Bocop and HamPath.

Theme: Overlay by Kaira