ESA GNC Conference Papers Repository

During the atmospheric flight, launcher guidance al gorithms commonly rely on an open-loop process. Thi s guidance scheme is relatively easy to implement. Ne vertheless, the inability to take into account the actual in- flight conditions generally leads to kinematic cond itions at the end of the implicit guidance scheme w hich can be far from the optimal ones. Many sources of distu rbances can affect flight conditions, either coming from the launcher (over- or under- propulsion, non-nominal p ropellant loading or structural masses) or from the environment (atmospheric density, wind); it address es the question of the feasibility and of the inter est of designing an explicit launcher guidance scheme duri ng atmospheric flight. The fundamental idea of the predictive control is b ased on an anticipating effect with regard to a pre -defined reference trajectory, by using the prediction of th e future behavior of the system while minimizing th e gap between this prediction and the reference trajector y. While widely considered in chemical and bio proc ess, robotics, applications of the predictive control to spacecraft flight control have been mainly studied for orbital systems: attitude control, in-orbit rendezvous, clo se-formation flight. An application to the atmosphe ric flight phase of a conventional launcher is proposed in thi s work. The paper presents a description of the flight phas e to guide, the formulation in term of nonlinear mo del predictive control, the problem solving, and the as sociated results. Some axes of improvements close t his paper. The work presented in this paper has been performed in the frame of CNES Launchers’ Research and Technology program.