Contribution of artificial proprioception on coordinated manipulation

Lacking an efficient sensory feedback system has caused upper-limb robotic prostheses to lose their popularity among users as they cannot reflect realistic feeling of performing a task. One of the obstacles in achieving this realism is missing proprioceptive feedback, which is the sense of relative position, velocity and kinesthetic force of neighboring parts of the body. It has been proved that proprioception plays an important role in body coordination during movements. Identifying the exact contribution of artificial proprioception on coordinated manipulation is a keynote in the research field yet to be explored. In order to study artificial proprioception, we have developed a novel virtual-reality-based experimental setup comprising of a two degrees of freedom (DOF) haptic device and an input device working with a virtual environment. Later, we conducted a psychophysical test in which subjects compared real and virtual springs with different stiffness constants. Results showed that the setup was able to render intended stiffness with high success rate. Eventually, a psychophysical test was conducted. An unstable task called "Strength-Dexterity Test", based on buckling of compression springs, has been employed. Relative contributions of vision, position and force and their sensory substitution method on feedback control was quantified with this 2-DOF task. Subjects interacted with a virtual spring with the index finger of their dominant hand through either the haptic interface, the input device, or a force sensor. Therefore, input condition were either isotonic (when the haptic device or the input device is used and the finger moves) or isometric (when the force sensor is used and the finger is stationary). Three feedback conditions were tested: visual only, wrong- modality sensory substitution of position and force (through vibration) and modality-matched sensory substitution. Results show that modality matched sensory susbtitution feedback did not have a notable effect on subjects' performance.