Control System Design for the ALINA Lunar Lander

Kurzfassung

This paper presents the control system developed by the German Aerospace Center (DLR) for the ALINA lunar lander. The control system is part of the overall Guidance, Navigation and Control (GNC) subsystem. ALINA, developed by Planetary Transportation Systems GmbH (PTS), is a spacecraft able to semi-autonomously per-form a complete Earth to lunar surface mission and to deliver up to 200 kg of payload. The control system has successfully passed the project’s Preliminary Design Review (PDR). The vehicle, the mission phases and objectives as well as the control requirements are introduced, and the adopted control solutions presented. Of major importance is the spacecraft’s propulsion configuration: it comprises a cluster of throttlable and non-throttlable main engines, as well as attitude control thrusters, fed by several propellant tanks. Therefore, specific challenges arise, such as control allocation and propellant sloshing. The synthesis of the controllers employs optimal control techniques to design the control laws for the different GNC tasks, including detumbling, single-axis pointing, three-axis pointing, maneuver execution, and powered descent, Hazard Detection and Avoidance (HDA) and landing. Attention has been given to the translational and rotational dynamics couplings, as the individual engines do not provide thrust vector control capabilities. The control allocation problem is tackled by minimizing a cost function that takes into account the desired forces and torques, as well as the fuel consumption; the online solution is found by transcribing the problem into a linear programming form and solving it using the SIMPLEX method. Propellant sloshing can disturb thrust vector pointing and can potentially generate critical deviations from the nominal trajectory and instabilities, particularly, as initially more than70 %of the spacecraft mass consists of fuel. Therefore, the plant model includes a representation of propellant sloshing dynamics using mechanical analogies. This has been achieved using the multi-physics object-oriented modeling language Modelica, better suited for large multibody representations; the resulting implementation is consequently embedded within a high-fidelity simulation framework. For the verification process, the control system has been included in the latter. Representative Monte-Carlo campaigns were conducted, with a dedicated focus on the powered descent, HDA and landing. The analysis of the control performance throughout each mission phase shows a good overall performance, well complying with all applicable requirements.