Advanced control strategies for natural Human-Exoskeleton interaction

The aim of the PhD thesis has been the design, development and test of advanced control strategies for natural and safe interaction between human and exoskeleton interfaces. All the three main applications that involve an exoskeleton interface have been investigated: robot-aided neurorehabilitation, physical interaction with virtual environment and bilateral teleoperation. In robot-aided neuro-rehabilitation, some of the main research questions are: How should we control the robot to make it moving as much natural as possible? How should we control the robot to make it exploitable by a large variety of patients? How can we design more simple and effective rehabilitation protocols? Electromyographic-based controls might be the practical solution to some of the open questions in robot-based rehabilitation, e.g. post-stroke neurorehabilitation. The PhD work presents some techniques based on neuromuscoloskeletal and muscles synergies used to detect the human intention of movement for exoskeleton control. Disaster scenarios, like the Fukushima nuclear accident, clearly show that the capabilities of today’s disaster-response robots are not sufficient for providing the desperately needed support to rescue workers, especially in the first critical hours. Shortly after the accident, the questions were raised: Why do humans need to expose themselves to dangers? Where are the robots? Haptic interfaces, e.g. exoskeletons, and bilateral teleoperation might be the real solution to such set of problems. In a real application, where the human operator need to be telepresent with its whole body in a robot acting remotely, the physical sense of telepresence is defined as the possibility to feel exactly what the remote robot is touching. During the PhD thesis some control techniques have been developed and tested in order to enhance the human-exoskeleton interaction stability for a high quality bilateral teleoperation.