Detecting Walking Challenges in Gait Patterns Using a Capacitive Sensor Floor and Recurrent Neural Networks

Abstract

Gait patterns are a result of the complex kinematics that enable human two-legged locomotion, and they can reveal a lot about a person’s state and health. Analysing them is useful for researchers to get new insights into the course of diseases, and for physicians to track the progress after healing from injuries. When a person walks and is interfered with in any way, the resulting disturbance can show up and be found in the gait patterns. This paper describes an experimental setup for capturing gait patterns with a capacitive sensor floor, which can detect the time and position of foot contacts on the floor. With this setup, a dataset was recorded where 42 participants walked over a sensor floor in different modes, inter alia, normal pace, closed eyes, and dual-task. A recurrent neural network based on Long Short-Term Memory units was trained and evaluated for the classification task of recognising the walking mode solely from the floor sensor data. Furthermore, participants were asked to do the Unilateral Heel-Rise Test, and their gait was recorded before and after doing the test. Another neural network instance was trained to predict the number of repetitions participants were able to do on the test. As the results of the classification tasks turned out to be promising, the combination of this sensor floor and the recurrent neural network architecture seems like a good system for further investigation leading to applications in health and care.

Recent Advances in Flexible Sensors and Their Applications

“…When force is applied to a capacitive sensor, the distance or the area of the plates changes, and the capacitance changes accordingly. Employing this principle, wearable sensors can be made using nanomaterials that measure capacitive strain [73][74][75][76][77][78][79][80]. A novel capacitive strain sensor using CNTs demonstrates good durability with a strain range between 1% and 300% [81].…”

Deep Neural Networks for Human’s Fall-risk Prediction using Force-Plate Time Series Signal

“…They experimented with a capacitive sensor floor to record walking kinematics of subjects. According to their result, using combination of sensor-based data and NN is a promising approach to be applied in in health and care (Hoffmann et al, 2021).…”

Identity and Gender Recognition Using a Capacitive Sensing Floor and Neural Networks

“…While wearable sensor [ 6 ] and computer vision-based gait analysis techniques [ 7 , 8 ] show good performance, they can be deemed as intrusive. The literature shows that it might be possible to extract gait information in an unobtrusive manner using floor-based sensing [ 9 , 10 ].…”

“…An issue with this capacitive approach is that it requires pressure-based floor deformation to operate, introducing fatigue-based longevity concerns similar to pressure-based sensor implementations. Hoffmann et al [ 9 ] explored gait mode classification using a capacitive sensing floor and an LSTM network, Other measurement methods have been used to detect: gender [ 39 , 40 ]; gait on steps [ 41 ]; emotional, height, and criminal detection [ 42 ]; fatigue [ 43 ]; identity [ 44 ]; and footsteps [ 21 , 45 ] and spatio-temporal gait parameters [ 46 ]. However, these approaches either required the subject to be tagged with a device, required the detection of floor vibration which would vary as the flooring aged, or suffered from low accuracy.…”