Human-Robot-Environment Interaction Strategies for Walker-Assisted Gait
Name: MARIO FERNANDO JIMENEZ HERNANDEZ
Publication date: 19/12/2018
Advisor:
Name |
Role |
|---|---|
| ANSELMO FRIZERA NETO | Advisor * |
| TEODIANO FREIRE BASTOS FILHO | Co-advisor * |
Examining board:
| Name |
Role |
|---|---|
| ANDRE FERREIRA | External Examiner * |
| ANSELMO FRIZERA NETO | Advisor * |
| ELIETE MARIA DE OLIVEIRA CALDEIRA | External Examiner * |
| TEODIANO FREIRE BASTOS FILHO | Co advisor * |
Summary: Smart Walkers (SWs) are robotic devices that may be used to improve balance and locomotion stability of people with lower-limb weakness or poor balance. Such devices may also offer support for cognitive disabilities and for people that cannot safely use conventional walkers, as well as allow interaction with other individuals and with the environment. In this context, there is a significant need to involve the environment information into the SW's control strategies. In this Ph.D. thesis, the concept of Human-Robot-Environment Interaction (HREI) for human locomotion assistance with a smart walker developed at UFES/Brazil (turned UFES's Smart Walker - USW) is explored. Two control strategies and one social navigation strategy are presented. The first control strategy is an admittance controller that generates haptic signals to induce the tracking of a predetermined path. When deviating from such path, the proposed method varies the damping parameter of the admittance controller by means of a spatial modulation technique, resulting in a haptic feedback, when is perceived by the user as a hard locomotion towards the undesired direction. The second strategy also uses an admittance controller to generate haptic signals, which guide the user along a predetermined path. However, in this case, the angular velocity of the smart walker is implemented as a function of a virtual torque, which is defined using two virtual forces that depend on the angular orientation error between the walker and the desired path. Regarding the navigation strategy, it involves social conventions defined by proxemics, and haptic signals generated through the spatial modulation of the admittance controller for a safe navigation within confined spaces.
The USW uses a multimodal cognitive interaction composed of a haptic feedback and a visual interface with two LEDs to indicate the correct/desired direction when necessary. The proposed control strategies are suitable for a natural HREI as demonstrated in the experimental validation. Moreover, this Ph.D. thesis presents a strategy to obtain navigation commands for the USW based on multi-axial force sensors, in addition to a study of the admittance control parameters and its influence on the maneuverability of the USW, in order to improve its HREI.
