Since the first extravehicular activities in 1965, the capabilities of EVA astronauts to do useful work outside of their spacecraft have steadily progressed. Likewise, our understanding of EVA astronauts' capabilities and limitations have also progressed through inflight experience, experimentation in neutral buoyancy and parabolic flight, and engineering tests of space suits and EVA tools. Computer models and dynamic simulation are the most recent tools for analyzing EVA capabilities. Computer simulation of EVA has several advantages over physical simulations, including the ability to accurately reproduce forces and displacements in six degrees of freedom and the absence of inherent time and workspace limitations.
One important shortcoming of current EVA models is that they lack an accurate representation of the torques that are required to bend the joints of the space suit. Modern space suits are designed to move with astronauts, using bearings and constant-volume joints to minimize resistance to motion. However, the torques required to perform EVA tasks still have a significant impact on task performance. The torques required to move space suit joints are complicated nonlinear functions of joint position and rate. Uncertainty in the knowledge of these torques leads to large variations in predicted task performance and metabolic costs. The focus of this research project is to quantify the interaction between the human and the space suit, for the Shuttle EMU, developing physics-based, data-driven models of the human-suit interaction. Simultaneous joint position and torque data was obtained using advanced optical motion capture techniques and robotic technology. Results have been applied to developing revolutionary locomotion spacesuits and to providing simulations for astronaut training.