• Adaptive oscillators
• Assisitive robotics
• Phase of gait

Longer life expectation and lower birth rate are challenging themes of industrialized society which has to deal with ageing of population who is still required to remain healthy and productive. Therefore, many research teams have focused their studies in designing wearable robotic devices for lower limb rehabilitation and assistance. Comfortable and ergonomic pHRI, non invasive and reliable cHRI, bioinspired mechatronic design of control and actuation system are still opening questions of this research field. In this work an active pelvis orthosis for gait assistance will be presented and described in detail in all its mechatronic components.
Our study will be focused on developing and validating a phase prediction algorithm based on adaptive frequency oscillators; an adaptive oscillator is a mathematical tool capable to learn the features of a non-sinusoidal signal such as its frequency (and consequently the phase) and envelope modeling the inherent limit cycle of its dynamical system. The dynamical system has been extended to synchronize oscillator phase with a specific event of the gait cycle. The phase predictor is exploited to instantaneously access the phase of gait of the wearer simply monitoring biomechanical signals through non invasive and reliable sensors. The phase of gait is then used to generate a phase-based torque aiming at assisting specific moments of the gait cycle. An additional experimental session was carried on to prove the feasibility of the developed assistive strategy.