Gait phase identification of various horse gaits using both wearable and video-based technologies

Abstract

Healthy gait consists of locomotion, balance and the ability to adapt to the environment. The human gait cycle has been studied thoroughly and gait analysis has been used for various applications. The biomechanics of the horse is well-known, but there is a need to develop systems that can quantify the quality of equine gait, especially in the real world. Inertial Measurement Units (IMU’s) are small, lightweight, and ambulant sensors which can determine the position and orientation of the body segments they have been placed on. The aim of this study was to investigate the possibility of using an Inertial Measurement Unit (IMU) System, especially placed at the cannons, to identify the gait phases for walk and trot, and to use an Optical Motion Capture System to validate these with information from wearable sensors. The positions of the hooves and cannons were tracked using Qualisys Track Manager. The resultant acceleration of each position was calculated. The resultant acceleration of Movebeat’s acceleration data, was computed in each sample. Hoof Strikes and Toe Offs were detected using Tracker from video recordings from Qualisys. The acceleration patterns were similar for Movebeat and Qualisys for both hooves and cannons and the acceleration peaks almost coincided with Hoof Strike and Toe Off. However, there were some parts of the pattern that differed, where one probable explanation could be that the acceleration from Qualisys was estimated, while the acceleration from Movebeat was measured. The conclusions are that IMU’s placed on the cannons are able to identify the gait events for walk and trot with accuracy, and that Hoof Strike and Toe Off almost coincides with the acceleration peaks. However, further investigations of the accelerations and the gyros in the different directions, x, y and z, need to be done to simplify the identification of the gait phases. More studies need to be performed to further validate the consistency of the identifications. This study was a collaboration project between Chalmers University of Technology, Swedish University of Agricultural Sciences, and VectorizeMove AB.