Application of quantized control to human self-rotation maneuvers in microgravity

We consider motion planning of human self- rotations; that is, human-body rotations without external torques, which are common in microgravity, diving, and gymnastics. In these cases it may be difficult to formulate an objective function that leads to maneuvers that are appropriate for humans to perform in high-stress situations. For example, cognitive complexity of a given motion can be difficult to quantify and incorporate in an objective function. In this paper, we pose the self-rotation planning problem using a quantization of motion, by defining a set of specific finite-time trajectories called motion primitives. By incorporating complexity considerations in the definition of the motion primitives, the quantized control method permits us to develop complete maneuvers that can be easily learned and are well-suited for high-stress situations. An application to astronaut motion is analyzed in order to lower the workload when a reorientation needs to be performed. Using the proposed motion-quantization methodology, we show that difficult-to-perform astronaut maneuvers can be replaced by concatenations of simpler motions, formed from rotations about two axes.