Gait analysis using floor markers and inertial sensors

Area of Science:

• Biomechanics
• Robotics
• Computer Vision

Background:

• Gait analysis is crucial for understanding human locomotion and diagnosing movement disorders.
• Accurate estimation of step length and foot angles presents a significant challenge in current gait analysis systems.
• Existing methods often rely on complex laboratory setups or invasive markers.

Purpose of the Study:

• To propose a novel gait analysis system for estimating step length and foot angles.
• To integrate vision and inertial sensor data for enhanced motion tracking accuracy.
• To develop a practical and efficient system for real-world gait assessment.

Main Methods:

• A measurement unit comprising a camera and inertial sensors was developed and attached to a shoe.
• A planar marker with 4,096 unique codes was used for floor-based position and attitude detection.
• An inertial navigation algorithm combined with a smoother integrated vision and sensor data for accurate foot motion estimation.

Main Results:

• The proposed system successfully tracked foot motion during gait.
• Accurate estimation of step length and foot angles was achieved through experimental validation.
• The integrated smoother significantly improved the accuracy of position and attitude estimation.

Conclusions:

• The developed gait analysis system offers a promising solution for accurate and non-invasive foot motion tracking.
• Combining camera and inertial sensor data provides a robust approach for gait parameter estimation.
• This system has potential applications in sports science, rehabilitation, and clinical diagnostics.