Nombre: CAMILO ARTURO RODRIGUEZ DIAZ
Fecha de publicación: 14/03/2014
Supervisor:
Nombre |
Papel |
|---|---|
| ANSELMO FRIZERA NETO | Advisor * |
| TEODIANO FREIRE BASTOS FILHO | Co-advisor * |
Junta de examinadores:
| Nombre |
Papel |
|---|---|
| ANSELMO FRIZERA NETO | Advisor * |
| EDUARDO ROCON DE LIMA | External Examiner * |
| EVANDRO OTTONI TEATINI SALLES | Internal Examiner * |
| TEODIANO FREIRE BASTOS FILHO | Co advisor * |
Sumario: This work presents the development of a robotic walker to aid people with motion disabilities. The device may be used for functional compensation or rehabilitation purposes, with a focus on a gait strategy based on interaction forces. For this purpose, a initial study of some gait supporting devices is performed, among which the most current and cited Smart Walkers in the literature are highlighted. A special attention is given to the walkers with force sensors used for physical interaction strategies. In this context, it is presented the UFES Smart Walker Project, developed at the Federal University of Esp´ırito Santo (UFES) in the Post-Graduate Program in Electrical Enginnering, giving continuity to the Spanish Project SIMBIOSIS, developed at the Spanish National Research Council. To interact with the user, the UFES Smart Walker is equipped with triaxial force sensors under the forearm supporting platforms, a laser range sensor to detect the users legs, and assisted locomotion. In addition, embeded electronic modules are used for real time signal processing. In order to develop a more natural force interface between the user and the walker, a rigorous signal processing based on adaptive filters is also presented.
The goal of this Smart Walker is to provide a more natural, safe and adapted
rehabilitation tool to the user. For this, the validation of the device comprises several experiments in which all signals are analyzed, with the objetive of determining the behavior of the human-machine interaction. The validation process is divided into three stages: interaction forces between the user and the walker, extraction of parameters related with the gait and response of high and low level controllers.
