Research Data supporting "Real-Time Prediction of Lower Limb Joint Kinematics, Kinetics, and Ground Reaction Force using Wearable Sensors and Machine Learning"

This dataset contains the FSR, IMU, ground reaction force (GRF), inverse kinematics (IK), and inverse dynamics (ID) data (Input data) by subject an side (right/left) used to train the models in the associated paper. /Data/ └── GRF Model/ ├── Subject1/ │ ├──file.txt │ ├──... ├── Subject2/ ├── ... └── Angle Models/ ├── Subject1/ │ ├── IK/ │ │ ├── Right │ │ │├──file.txt │ │ ├── Left │ │ │├──file.txt │├── Sensors/ │ │ ├── Right │ │ │├──file.txt │ │ ├── Left │ │ │├──file.txt ├── Subject2/ ├── ... └── Moment Models/ ├── Subject1/ │ ├── IK/ │ │ ├── Right │ │ │├──file.txt │ │ ├── Left │ │ │├──file.txt │├── ID/ │ │ ├── Right │ │ │├──file.txt │ │ ├── Left │ │ │├──file.txt │├── GRF/ │ │ ├── Right │ │ │├──file.txt │ │ ├── Left │ │ │├──file.txt ├── Subject2/ ├── ... For GRF Model, the files include the aligned FSR (FSR1 to 8) and vGRF (GRF) data. For Angle Models, the IK files contain all the joint angles in the model (with time and gait percentage). The Sesnors files contain the UNIX-timestamped IMu data (accelerometer AX, AY, AZ. magnetometer MX, MY, MZ, and gyroscope GX, GY, GZ) for the 4 IMUs in order: right shank, left shank, right foot, left foot (the remaining columns for EMG1 to 6 and metab are not used in this work). For Moment Models, the IK files follow the same structure, in addition to the ID files which have the joint moments for all the model joints. The GRF files have the timed vGRF values for the right and left feet.