Generation of Time-Averaged Simulated Microgravity and Partial Gravity in a 2-DOF Robotic Device Using Randomized Motion and Gravity Vector Density Analysis

Abstract This study presents a novel path-planning algorithm for a clinostat with two degrees of freedom, designed to generate time-averaged simulated microgravity (taSMG) and partial gravity (taSPG). The existing methods for generating taSMG suffer from an uneven distribution of the gravity vector, long convergence time, or absence of randomness. To overcome these limitations, an algorithm is proposed that directs the gravity direction vector toward under-visited regions on a discretized unit sphere while intermittently introducing random motions to avoid repetitive trajectories. Simulation results indicate that when the endpoints of the gravity direction vectors are mapped onto a unit sphere divided into ten equal intervals in terms of both the latitude and longitude, and 25% of the overall motion follows a random pattern, taSMG converges stably within 10 −3 G in less than 12 h. For taSPG generation, linear and second-order polynomial normalization factors, with tuned hyperparameters, were applied to bias the density of the gravity direction vectors along the latitude, enabling accurate simulation of specific gravitational environments, such as those of Moon (0.17 G) and Mars (0.38 G), with errors below 0.5% and standard deviations under 1%. To the best of our knowledge, this is the first study to propose a clinostat path-planning algorithm that simultaneously achieves rapid convergence, controlled gravity vector distribution, and randomized motion. The findings of this study suggest that the proposed algorithm can provide reliable taSMG or taSPG environments for biological experiments, making it a valuable tool for microgravity and partial gravity research on Earth.