Study and development of sensorimotor interfaces for robotic human augmentation

This thesis presents my research contribution to robotics and haptics in the context of human augmentation. In particular, in this document, we are interested in bodily or sensorimotor augmentation, thus the augmentation of humans by supernumerary robotic limbs (SRL). The field of sensorimotor augmentation is new in robotics and thanks to the combination with neuroscience, great leaps forward have already been made in the past 10 years. All of the research work I produced during my Ph.D. focused on the development and study of fundamental technology for human augmentation by robotics: the sensorimotor interface. This new concept is born to indicate a wearable device which has two main purposes, the first is to extract the input generated by the movement of the user's body, and the second to provide the somatosensory system of the user with an haptic feedback. This thesis starts with an exploratory study of integration between robotic and haptic devices, intending to combine state-of-the-art devices. This allowed us to realize that we still need to understand how to improve the interface that will allow us to feel the agency when using an augmentative robot. At this point, the path of this thesis forks into two alternative ways that have been adopted to improve the interaction between the human and the robot. In this regard, the first path we presented tackles two aspects conerning the haptic feedback of sensorimotor interfaces, which are the choice of the positioning and the effectiveness of the discrete haptic feedback. In the second way we attempted to lighten a supernumerary finger, focusing on the agility of use and the lightness of the device. One of the main findings of this thesis is that haptic feedback is considered to be helpful by stroke patients, but this does not mitigate the fact that the cumbersomeness of the devices is a deterrent to their use. Preliminary results here presented show that both the path we chose to improve sensorimotor augmentation worked: the presence of the haptic feedback improves the performance of sensorimotor interfaces, the co-positioning of haptic feedback and the input taken from the human body can improve the effectiveness of these interfaces, and creating a lightweight version of a SRL is a viable solution for recovering the grasping function.