Generation of Stable Trajectories for Lower Limb Exoskeletons using Neural Oscillators

• Grant number: 09/17593-9
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: April 01, 2010
• End date: March 31, 2011
• Field of knowledge: Engineering - Mechanical Engineering
• Principal Investigator: Adriano Almeida Gonçalves Siqueira
• Grantee: Rafael Gomes Gobbo
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

In this project, a trajectory generator based on neural oscillators for a lower limb exoskeletons will be developed in such an integrated and biologically inspired way that, the purpose of walking of the user during the rehabilitation process will be considered. This work is immersed in a project of building and controlling a lower limb exoskeleton from a commercial orthosis. Such equipment, inicially designed for rehabilitation processes of the impaired, is being built with an activating system based on elastic actuators in series. A trajectory generator making use of cubic splines interpolation has been developed as a part of this project, considering a ZMP stability criteria (Zero Moment Point). Further more, an algorithm using the inverse dynamics of the model and neural networks with MLP architecture (MultiLayer Perceptron) was also developed, being able to adapt the progress pattern during the walk by receiving active forces from the user.