Human-Machine Interaction in a Self-Paced Treadmill: Comparison of Control Strategies and Development of Serious Games
Name: BRAYAN SNEIDER MORENO AREVALO
Publication date: 09/09/2024
Examining board:
Name |
Role |
|---|---|
| ANSELMO FRIZERA NETO | Presidente |
| CAMILO ARTURO RODRIGUEZ DIAZ | Examinador Interno |
| EDUARDO ROCON DE LIMA | Examinador Externo |
| RICARDO CARMINATI DE MELLO | Coorientador |
Summary: Mobility is crucial for quality of life, especially for individuals with motor limitations. This research explores a motor rehabilitation model integrating self-paced treadmills and serious games. The objective was to develop and validate a self-paced treadmill for gait rehabilitation, utilizing sensor-based technology and adaptive algorithms to adjust speed according to the user's walking rhythm. Human-machine interaction strategies using serious games were implemented to enhance treatment adherence and rehabilitation outcomes. The UFES SmartTreadmill was developed with LRF sensors, a motion detection camera, and a game projection system. Control strategies employing adaptive WFLC and FLC algorithms allowed precise estimation of walking speed and dynamic adaptation to user needs. Furthermore, serious games provided an interactive and motivating environment, facilitating treatment adherence and improving rehabilitation outcomes. Three experiments with young, healthy participants demonstrated effective control strategies for gait parameters, even during dual tasks. Stride length and cadence increased with walking speed, following logarithmic and linear trends respectively, akin to ground walking dynamics, with high R2 values (0.84 for stride length, 0.82 for cadence). Implementing serious games in rehabilitation significantly improved participant immersion and motivation. The Player Experience Inventory (PXI) questionnaire showed increased autonomy perception and progress feedback with task complexity, crucial for maintaining interest and adherence. In conclusion, this study establishes a foundation for technology-assisted rehabilitation, validating the effectiveness of sensor systems and serious games in enhancing motor rehabilitation, patient motivation, and treatment adherence. Future research should expand these findings to diverse populations and integrate technologies like augmented reality and AI algorithms. Longitudinal studies are crucial for assessing long-term effects, sustainability, and effectiveness.
