IMU-Based Walking Data

Citation Author(s):
Lefan
Wang
University of Cambridge
Sebastian
Pattinson
University of Cambridge
Submitted by:
Lefan Wang
Last updated:
Mon, 07/08/2024 - 15:58
DOI:
10.21227/sfy3-jn97
Data Format:
Research Article Link:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

IMUs have gained popularity for tracking joint kinematics due to their portability and versatility. However, challenges such as limited accuracy, lack of real-time data analysis, and complex sensor-to-segment calibration procedures have hindered their widespread use. To address these limitations, we developed a portable system that integrates four IMUs to collect treadmill walking data, with ground truth values obtained from a Motion Capture System. Our objective was to leverage ANNs to predict multi-plane ankle kinematics in real-time using a minimal number of IMUs, eliminating calibration requirements. This approach aims to provide reliable and practical tools for early prevention and rehabilitation of ankle injuries.

Instructions: 

Datafile description:

Each datafile has 68 columns.

Column 1: time information.

Column 2-16: the data from S1 which was located at the right shank. Specifically,

column 2-4: Acceleration data along the x, y, and z axes.

column 5-7: Gyroscope data along the x, y, and z axes.

Column 8-10: Linear acceleration (i.e. acceleration that is applied due to movement) data along the x, y, and z axes.

Column 11-14: quantization w, x, y, and z.

Column 15: temperature readings from BNO055

Column 16: IMU calibration status.

Column 17-31: the data from S2 which was located at the right thigh.

Column 32-46: the data from S3 which was located at the right heel.

Column 47-61: the data from S4 which was located below the right ankle.

Column 62-68: data obtained from the MoCAP VICON system, which is considered the ground truth values.

Column 62-64: knee kinematics along the x, y, and z axis, which are flexion, abduction, and rotation, respectively.

Column 65-67: ankle kinematics along the x, y, and z axis, which are flexion, rotation, and inversion/eversion, respectively.

Column 68: Gait events. 1: toe off; 2: heel strike, 0: all the others

Funding Agency: 
EPSRC
Grant Number: 
EP/V004654/1 and EP/V025279/1