Gait Analysis Using Floor Markers and Inertial Sensors

Abstract

In this paper, a gait analysis system which estimates step length and foot angles is proposed. A measurement unit, which consists of a camera and inertial sensors, is installed on a shoe. When the foot touches the floor, markers are recognized by the camera to obtain the current position and attitude. A simple planar marker with 4,096 different codes is used. These markers printed on paper are placed on the floor. When the foot is moving off the floor, the position and attitude are estimated using an inertial navigation algorithm. For accurate estimation, a smoother is proposed, where vision information and inertial sensor data are combined. Through experiments, it is shown that the proposed system can both track foot motion and estimate step length.

Sensor Saturation Compensated Smoothing Algorithm for Inertial Sensor Based Motion Tracking

“…This type of movement can be found in walking [4,5], golf swing [17] and so on. The sensor data are sampled with the sampling period T .…”

Inertial Sensor-Based Two Feet Motion Tracking for Gait Analysis

“…In this paper, we use the inertial navigation algorithm in [9,15]. We only state an algorithm for r̂ 1 since the algorithm for r̂ 2 is exactly the same.…”

“…In [7,8], angles of leg joints are estimated, where attitude estimation algorithms are applied using inertial sensor data. In [9], an inertial navigation algorithm [10] is used to estimate a foot movement. By installing inertial sensors on a shoe, foot motion (position, velocity and attitude) can be estimated quantitatively.…”

“…Inertial navigation algorithm-based foot motion analysis is in most cases [9,11–13] done only for a single foot. However, two feet motion tracking provides more information for the gait analysis.…”

Estimation of Temporal Gait Parameters Using a Wearable Microphone-Sensor-Based System

“…Inertial sensors, which can be embedded in wearable and portable electronic devices, have been widely used in gait recognition for individual identification [11,12], gait action recognition [13] and gait analysis [14,15,16,17] because they are inexpensive, not limited by environmental conditions, low-power and tiny. Most wearable GA methods only use inertial sensor-based methods [14,15,16,17]. The temporal parameters of inertial sensor-based methods are mainly measured by threshold-based peak detection methods, so generally it is not easy to achieve high detection accuracy because it’s almost impossible to find a fixed threshold adaptable to many kinds of conditions.…”