This technology allows one to test small-body surface mobility and sampling systems in the laboratory. It is capable of simulating a microgravity environment with relevant terrain. The magnitude of the gravity, the terrain properties, and the surface system being tested are all easily modified to allow for a broad range of experimental setups.
The small body testbed consists of two major components: a 6-DOF (degrees of freedom) microgravity gantry and terrain simulant. The 6-DOF microgravity gantry consists of three active linear DOFs and a passive 3-DOF gimbal. The active DOFs consist of three brushless motors with 500 count encoders and planetary gear heads. Three absolute encoders on the gimbal provide an angular measurement resolution of 1.53 mrad. It should be noted that these resolutions are not the absolute accuracy of the gantry pose estimate, which will also be affected by backlash and deflections of the mechanisms and structures of the gantry. An F/T (force/torque) sensor located at the end of the third stage (but before the gimbal) measures the interaction forces between the robot and the environment. Measurement of the three orthogonal forces can then be used to determine the linear accelerations of the gantry axes in such a way as to simulate a robot of arbitrary mass with a gravity vector of arbitrary magnitude and direction.
The passive gimbal was designed as a compact passive spherical joint that uses ballast masses above the robot in order to make the center of mass (CM) of the robot/ballast system coincide with the intersection of the three gimbal axes. This results in a balanced system that floats freely and responds to external forces with appropriate rotational accelerations at all angles. Some sacrifices are made with this passive gimbal system for the three rotational DOFs as compared with an active system, including the inability to simulate arbitrary rotational inertia, the inability to accommodate changes in the CM due to limb motion, and the relocation of the CM of the robot, which affects the interaction dynamics with the environment. These sacrifices were made to greatly simplify the system.
The terrain simulant resides in a sandbox below the gantry that can be raised and lowered to accommodate a wide range of terrain types and topographies.
The advantage that this small-body testbed has over other existing technologies is the combination of the relatively large workspace, the sensitivity of the interaction force measurement, the accuracy of the ground truth measurement, the high rate of loop closure resulting in smooth and precise motion, and the full 6-DOF simulation.