A range of body heights is assumed for human-beings and the corresponding body masses are calculated based on body mass index (BMI). Masses, moments of inertia, lengths and centers of masses of all body limbs are computed. Inverse kinematics is conducted to find out the required joint angles to achieve a suitable gait cycle. Based on these angles, forward kinematics and inverse dynamics are carried out in order to determine torques and subsequently, power consumptions at knee and hip joints for flexion/extension. By taking body heights and masses as inputs and power consumptions at knee and hip joints as the outputs, multilayer feed-forward neural network architecture is developed. The neural network is trained using back-propagation algorithm. After the training, an input-output relationship between body heights, body masses and power consumption at knee and hip joints, is established. From this data bank, it is possible to predict the required power consumptions at knee and hip joints for a variety of patients of different body heights and masses, on-line.
CITATION STYLE
Pal, A. R., Mundada, A. O., & Pratihar, D. K. (2016). Predictions of hip and knee power consumptions of patients having different body heights and masses during normal walking. Lecture Notes in Mechanical Engineering, 149–157. https://doi.org/10.1007/978-81-322-2740-3_16
Mendeley helps you to discover research relevant for your work.