Use of AI algorithms and discrete modeling to improve control of a flexible parachute canopy during maneuvers

Context : The complexity and the unstable and uncertain non-linear nature of the response of large flexible parachute canopies, interacting with a surrounding fluid under the effect of manoeuvring actions, are obstacles for deterministic flight mechanics control models and algorithms. Conventional weakly or strongly coupled finite element simulation methods are costly and unstable, and cannot be used to improve their construction.

Project : To meet this modelling and simulation challenge, the team is developing a research code for structural calculations based on a discrete method [1]. This code is frugal in terms of calculation time and can be used to assess the influence of local effects of tissue deformation on manoeuvring forces. The aim of the thesis is to improve the existing tool to enable dialogue with artificial intelligence (AI) algorithms [2,3] for the analysis of the stability and steerability of the sail. The aim of the thesis is to meet the ambition of getting the sciences of structural mechanics to communicate and interact with those of control and AI mathematical algorithms.