Granular Matter in Extraterrestrial Environments - Modeling and Simulation of Regolith in Planetary Exploration

Kurzfassung

In everydays life, moving on soft, sandy soils is a well known task for everyone us. Regardless, the interaction physics of and with granular matter is still not fully understood. Whilst in terrestrial applications a direct recovery of locomotion gear is possible by external means, planetary exploration are fend for themselves.The talk will introduce enhanced but yet efficient simulation methods that are used in development of locomotion gear. Validated application models allow for in-depth analysis of the interaction, even beyond the possibilities given by measurements: As shown in figure 1 flows of granular matter may be tracked in any part of the medium, forces and locomotion behaviour can be predicted and mutual influences of locomotion gear and sand can by analyzed. Given the possibilities of the models, particle simulation poses a great chance to further improve future locomotion gear. Thus the focus will be on improvements of the particle-based Discrete Element Method (DEM). One of the tasks for modeling the correct shear strength of granular matter is efficient coverage of grain rolling and interlocking. To allow for efficient simulations the complex shapes of real grains need to be simplified. Covering certain features like angularity or rounded edges is crucial. In the proposed approach two dimensional rolling geometries are attached to the spherical grains and superimposed to 3D equivalent bodies [1]. Resulting in an improved non-linear torque behaviour, it becomes possible to cover behaviour like tilting or rotational interlocking. Details on the method will be presented alongside torque-laws for different shapes, including edge rounding. A second challenge in particle simulations is parameter identification. Real-world sand grains are too small to directly identify inter-grain contact parameters. The state of the art treats this problem with multi-parameter calibration to measured data, leading to non-unique solutions in long lasting processes [2]. A method to identify a parameter set without the need of preliminary simulations or calibration has been developed [3] and will be explained throughout the talk. Additionally methods to improve efficiency, e.g. dynamic boundaries, will be shown. The presented methods will be presented alongside their applications in current planetary exoploration mission. These range from “the Mole”, a self-impelling nail for NASA’s InSight Mission to Mars, over MASCOT, DLR’s asteroid hopping currently on its way to asteroid Ryugu, to Wheels of planetary exploration rovers (Figure 1).