Future space robotic systems that will enable exploration of the lunar surface represent complex platforms, characterized by numerous operational constraints and by the need to operate in a wide range of environmental conditions. A key element of the development of these systems is a strong integration with the active control of the vehicle’s attitude and distance from the ground, which is necessary to ensure payload interaction with the surface, the stabilization of the chassis, and the facilitation of the traversal of environments characterized by steep slopes and irregularities. In this context, predictive control techniques based on a linear model of the system struggle to ensure optimal performance over the entire operating envelope and to correctly describe the kinematic and dynamic nonlinear effects to which the system is subjected.