{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Inverse dynamics (2D) for gait analysis \n", "\n", "> Marcos Duarte, Renato Naville Watanabe \n", "> [Laboratory of Biomechanics and Motor Control](http://pesquisa.ufabc.edu.br/bmclab/) \n", "> Federal University of ABC, Brazil" ] }, { "cell_type": "markdown", "metadata": { "toc": 1 }, "source": [ "

# Contents

\n", "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Python setup" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "ExecuteTime": { "end_time": "2020-05-13T00:42:50.784536Z", "start_time": "2020-05-13T00:42:49.980687Z" } }, "outputs": [], "source": [ "# import the necessary libraries\n", "import numpy as np\n", "from numpy import cross\n", "from IPython.display import IFrame\n", "%matplotlib inline\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "sns.set_context('notebook', font_scale=1.2, rc={\"lines.linewidth\": 2})\n", "import sys\n", "sys.path.insert(1, r'./../functions')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Forward and inverse dynamics\n", "\n", "With respect to the equations of motion to determine the dynamics of a system, there are two general approaches: forward (or direct) and inverse dynamics. For example, consider the solution of Newton's second law for a particle. If we know the force(s) and want to find the trajectory, this is **forward dynamics**. If instead, we know the trajectory and want to find the force(s), this is **inverse dynamics**:\n", "\n", "\n", "\n", "In Biomechanics, in a typical movement analysis of the human body using inverse dynamics, we would measure the positions of the segments and measure the external forces, calculate the segments' linear and angular acceleration, and find the internal net force and moment of force at the joint using the equations of motion. In addition, we could estimate the muscle forces (if we solve the redundancy problem of having more muscles than joints). \n", "Using forward dynamics, the muscle forces would be the inputs and the trajectories of the segments would be the outputs. The figure below compares the forward and inverse dynamics approaches.\n", "\n", "" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Estimation of joint force and moments of force by inverse dynamics\n", "\n", "Let's estimate the joint force and moments of force at the lower limb during locomotion using the inverse dynamics approach. \n", "We will model the lower limbs at the right side as composed by three rigid bodies (foot, leg, and thigh) articulated by three hinge joints (ankle, knee, and hip) and perform a two-dimensional analysis. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Free body diagrams\n", "\n", "The [free body diagrams](http://nbviewer.ipython.org/github/demotu/BMC/blob/master/notebooks/FreeBodyDiagram.ipynb) of the lower limbs are: \n", "
\n", " \n", " " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ " \n", "### Equations of motion\n", " \n", "The equilibrium equations for the forces and moments of force around the center of mass are: \n", "\n", "For body 1 (foot):\n", "\n", "\$$\n", "\\begin{array}{l l}\n", "\\mathbf{F}_1 + m_1\\mathbf{g} + \\mathbf{GRF} = m_1\\mathbf{a}_1 \\\\\n", "\\mathbf{M}_1 + \\mathbf{r}_{cmp1}\\times\\mathbf{F}_1 + \\mathbf{r}_{cmCOP}\\times\\mathbf{GRF} = I_1\\mathbf{\\alpha}_1 \n", "\\label{}\n", "\\end{array}\n", "\$$\n", "\n", "For body 2 (leg):\n", "\n", "\$$\n", "\\begin{array}{l}\n", "\\mathbf{F}_2 + m_2\\mathbf{g} - \\mathbf{F}_1 = m_2\\mathbf{a}_2 \\\\\n", "\\mathbf{M}_2 + \\mathbf{r}_{cmp2}\\times\\mathbf{F}_2 + \\mathbf{r}_{cmd2}\\times-\\mathbf{F}_{1} - \\mathbf{M}_1 = I_2\\mathbf{\\alpha}_2\n", "\\label{}\n", "\\end{array}\n", "\$$\n", "\n", "For body 3 (thigh):\n", "\n", "\$$\n", "\\begin{array}{l l}\n", "\\mathbf{F}_3 + m_3\\mathbf{g} - \\mathbf{F}_2 = m_3\\mathbf{a}_3 \\\\\n", "\\mathbf{M}_3 + \\mathbf{r}_{cmp3}\\times\\mathbf{F}_3 + \\mathbf{r}_{cmd3}\\times-\\mathbf{F}_{2} - \\mathbf{M}_2 = I_3\\mathbf{\\alpha}_3\n", "\\label{}\n", "\\end{array}\n", "\$$\n", "\n", "Where $p$ and $d$ stands for proximal and distal joints (with respect to the fixed extremity), $\\mathbf{r}_{cmji}$ is the position vector from the center of mass of body $i$ to the joint $j$, $COP$ is the center of pressure, the position of application of the resultant ground reaction force (GRF), $\\mathbf{\\alpha}$ is the angular acceleration, and $g$ is the acceleration of gravity\n", "\n", "Note that the pattern of the equations is the same for the three segments: distal and proximal forces and moments of force and the weight force are present in all segments. \n", "The only exception is with the foot in contact with the ground. As the ground only pushes the foot, it can not generate a moment of force over the foot. Because of that we model the interaction foot-ground as a resultant ground reaction force (GRF) applied on the foot at the COP position. \n", "\n", "Both GRF and COP quantities are measured with a force platform and are assumed as known quantities. \n", "Because of that the system of equations above is only solvable if we start by the body 1, from bottom to top. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The system of equations above is simple and straightforward to solve, it is just a matter of being systematic. \n", "We start by segment 1, find $\\mathbf{F}_1$ and $\\mathbf{M}_1$, substitute these values on the equations for segment 2, find $\\mathbf{F}_2$ and $\\mathbf{M}_2$, substitute them in the equations for segment 3 and find $\\mathbf{F}_3$ and $\\mathbf{M}_3\\:$: \n", "\n", "For body 1 (foot):\n", " \n", "\$$\n", "\\begin{array}{l l}\n", "\\mathbf{F}_1 &=& m_1\\mathbf{a}_1 - m_1\\mathbf{g} - \\mathbf{GRF} \\\\\n", "\\mathbf{M}_1 &=& I_1\\mathbf{\\alpha}_1 - \\mathbf{r}_{cmp1}\\times\\big(m_1\\mathbf{a}_1 - m_1\\mathbf{g} - \\mathbf{GRF}\\big) - \\mathbf{r}_{cmCOP}\\times\\mathbf{GRF}\n", "\\label{}\n", "\\end{array}\n", "\$$\n", "\n", "For body 2 (leg):\n", "\n", "\$$\n", "\\begin{array}{l l}\n", "\\mathbf{F}_2 &=& m_1\\mathbf{a}_1 + m_2\\mathbf{a}_2 - (m_1+m_2)\\mathbf{g} - \\mathbf{GRF} \\\\\n", "\\mathbf{M}_2 &=& I_1\\mathbf{\\alpha}_1 + I_2\\mathbf{\\alpha}_2 - \\mathbf{r}_{cmp2}\\times\\big(m_1\\mathbf{a}_1 + m_2\\mathbf{a}_2 - (m_1+m_2)\\mathbf{g} - \\mathbf{GRF}\\big)\\, + \\\\\n", "&\\phantom{=}& \\mathbf{r}_{cgd2}\\times\\big(m_1\\mathbf{a}_1 - m_1\\mathbf{g} - \\mathbf{GRF}\\big) - \\mathbf{r}_{cmp1}\\times\\big(m_1\\mathbf{a}_1 - m_1\\mathbf{g} - \\mathbf{GRF}\\big)\\, - \\\\\n", "&\\phantom{=}& \\mathbf{r}_{cmCOP}\\times\\mathbf{GRF}\n", "\\label{}\n", "\\end{array}\n", "\$$\n", "\n", "For body 3 (thigh):\n", "\n", "\$$\n", "\\begin{array}{l l}\n", "\\mathbf{F}_3 &=& m_1\\mathbf{a}_1 + m_2\\mathbf{a}_2 + m_3\\mathbf{a}_3 - (m_1+m_2+m_3)\\mathbf{g} - \\mathbf{GRF} \\\\\n", "\\mathbf{M}_3 &=& I_1\\mathbf{\\alpha}_1 + I_2\\mathbf{\\alpha}_2 + I_3\\mathbf{\\alpha}_3\\, - \\\\\n", "&\\phantom{=}& \\mathbf{r}_{cmp3}\\times\\big(m_1\\mathbf{a}_1 + m_2\\mathbf{a}_2 + m_3\\mathbf{a}_3\\, - (m_1+m_2+m_3)\\mathbf{g} - \\mathbf{GRF}\\big)\\, + \\\\\n", "&\\phantom{=}& \\mathbf{r}_{cmd3}\\times\\big(m_1\\mathbf{a}_1 + m_2\\mathbf{a}_2 - (m_1+m_2)\\mathbf{g} - \\mathbf{GRF}\\big)\\, - \\\\\n", "&\\phantom{=}& \\mathbf{r}_{cmp2}\\times\\big(m_1\\mathbf{a}_1 + m_2\\mathbf{a}_2 - (m_1+m_2)\\mathbf{g} - \\mathbf{GRF}\\big)\\, + \\\\\n", "&\\phantom{=}& \\mathbf{r}_{cmd2}\\times\\big(m_1\\mathbf{a}_1 - m_1\\mathbf{g} - \\mathbf{GRF}\\big)\\, - \\\\\n", "&\\phantom{=}& \\mathbf{r}_{cmp1}\\times\\big(m_1\\mathbf{a}_1 - m_1\\mathbf{g} - \\mathbf{GRF}\\big)\\, - \\\\\n", "&\\phantom{=}& \\mathbf{r}_{cmCOP}\\times\\mathbf{GRF}\n", "\\label{}\n", "\\end{array}\n", "\$$" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### The recursive approach for inverse dynamics of multi-body systems\n", "\n", "The calculation above is tedious, error prone, useless, and probably it's wrong. \n", "\n", "To make some use of it, we can clearly see that forces act on far segments, which are not directly in contact with these forces. In fact, this is true for all stuff happening on a segment: note that $\\mathbf{F}_1$ and $\\mathbf{M}_1$ are present in the expression for $\\mathbf{F}_3$ and $\\mathbf{M}_3$ and that the acceleration of segment 1 matters for the calculations of segment 3. \n", "\n", "Instead, we can use the power of computer programming (like this one right now!) and solve these equations recursively hence they have the same pattern. Let's do that.\n", "\n", "For body 1 (foot):\n", "\n", "\$$\n", "\\begin{array}{l l}\n", "\\mathbf{F}_1 = m_1\\mathbf{a}_1 - m_1\\mathbf{g} - \\mathbf{GRF} \\\\\n", "\\mathbf{M}_1 = I_1\\mathbf{\\alpha}_1 - \\mathbf{r}_{cmp1}\\times\\mathbf{F}_1 - \\mathbf{r}_{cmCOP}\\times\\mathbf{GRF}\n", "\\label{}\n", "\\end{array}\n", "\$$\n", "\n", "For body 2 (leg):\n", "\n", "\$$\n", "\\begin{array}{l}\n", "\\mathbf{F}_2 = m_2\\mathbf{a}_2 - m_2\\mathbf{g} + \\mathbf{F}_1\\\\\n", "\\mathbf{M}_2 = I_2\\mathbf{\\alpha}_2 - \\mathbf{r}_{cmp2}\\times\\mathbf{F}_2 +\\mathbf{r}_{cmd2}\\times\\mathbf{F}_{1} + \\mathbf{M}_1\n", "\\label{}\n", "\\end{array}\n", "\$$\n", "\n", "For body 3 (thigh):\n", "\n", "\$$\n", "\\begin{array}{l l}\n", "\\mathbf{F}_3 = m_3\\mathbf{a}_3 - m_3\\mathbf{g} + \\mathbf{F}_2\\\\\n", "\\mathbf{M}_3 = I_3\\mathbf{\\alpha}_3 - \\mathbf{r}_{cmp3}\\times\\mathbf{F}_3 + \\mathbf{r}_{cmd3}\\times\\mathbf{F}_{2} + \\mathbf{M}_2\n", "\\label{}\n", "\\end{array}\n", "\$$" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Python function invdyn2d.py\n", "\n", "We could write a function that it would have as inputs the body-segment parameters, the kinematic data, and the distal joint force and moment of force and output the proximal joint force and moment of force. \n", "Then, we would call this function for each segment, starting with the segment that has a free extremity or that has the force and moment of force measured by some instrument (i,e, use a force plate for the foot-ground interface). \n", "This function would be called in the following manner: \n", "\n", "python\n", " Fp, Mp = invdyn2d(rcm, rd, rp, acm, alfa, mass, Icm, Fd, Md)\n", " \n", "\n", "So, here is such function:" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "ExecuteTime": { "end_time": "2020-05-13T00:40:07.354464Z", "start_time": "2020-05-13T00:40:07.342326Z" } }, "outputs": [], "source": [ "# %load ./../functions/invdyn2d.py\n", "\"\"\"Two-dimensional inverse-dynamics calculations of one segment.\"\"\"\n", "\n", "__author__ = 'Marcos Duarte, https://github.com/demotu/BMC'\n", "__version__ = 'invdyn2d.py v.2 2015/11/13'\n", "\n", "\n", "def invdyn2d(rcm, rd, rp, acm, alpha, mass, Icm, Fd, Md):\n", " \"\"\"Two-dimensional inverse-dynamics calculations of one segment\n", "\n", " Parameters\n", " ----------\n", " rcm : array_like [x,y]\n", " center of mass position (y is vertical)\n", " rd : array_like [x,y]\n", " distal joint position\n", " rp : array_like [x,y]\n", " proximal joint position\n", " acm : array_like [x,y]\n", " center of mass acceleration\n", " alpha : array_like [x,y]\n", " segment angular acceleration\n", " mass : number\n", " mass of the segment \n", " Icm : number\n", " rotational inertia around the center of mass of the segment\n", " Fd : array_like [x,y]\n", " force on the distal joint of the segment\n", " Md : array_like [x,y]\n", " moment of force on the distal joint of the segment\n", " \n", " Returns\n", " -------\n", " Fp : array_like [x,y]\n", " force on the proximal joint of the segment (y is vertical)\n", " Mp : array_like [x,y]\n", " moment of force on the proximal joint of the segment\n", "\n", " Notes\n", " -----\n", " To use this function recursevely, the outputs [Fp, Mp] must be inputed as \n", " [-Fp, -Mp] on the next call to represent [Fd, Md] on the distal joint of the\n", " next segment (action-reaction).\n", " \n", " This code was inspired by a similar code written by Ton van den Bogert [1]_.\n", " See this notebook [2]_.\n", "\n", " References\n", " ----------\n", " .. [1] http://isbweb.org/data/invdyn/index.html\n", " .. [2] http://nbviewer.ipython.org/github/demotu/BMC/blob/master/notebooks/GaitAnalysis2D.ipynb\n", "\n", " \"\"\"\n", " \n", " g = 9.80665 # m/s2, standard acceleration of free fall (ISO 80000-3:2006)\n", " # Force and moment of force on the proximal joint\n", " Fp = mass*acm - Fd - [0, -g*mass]\n", " Mp = Icm*alpha - Md - cross(rd-rcm, Fd) - cross(rp-rcm, Fp)\n", " \n", " return Fp, Mp" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The inverse dynamics calculations are implemented in only two lines of code at the end, the first part of the code is the help on how to use the function. The help is long because it's supposed to be helpful :), see the [style guide for NumPy/SciPy documentation](https://numpydoc.readthedocs.io/en/latest/format.html#docstring-standard). \n", "\n", "The real problem is to measure or estimate the experimental variables: the body-segment parameters, the ground reaction forces, and the kinematics of each segment. For such, it is necessary some expensive equipments, but they are typical in a biomechanics laboratory, such the the [BMClab](http://pesquisa.ufabc.edu.br/bmclab). " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Experimental data\n", "\n", "Let's work with some data of kinematic position of the segments and ground reaction forces in order to compute the joint forces and moments of force. \n", "The data we will work are in fact from a computer simulation of running created by Ton van den Bogert. The nice thing about these data is that as a simulation, the true joint forces and moments of force are known and we will be able to compare our estimation with these true values. \n", "All the data can be downloaded from a page at the [ISB website](http://isbweb.org/data/invdyn/index.html):" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "ExecuteTime": { "end_time": "2020-05-13T00:40:09.564573Z", "start_time": "2020-05-13T00:40:09.551269Z" } }, "outputs": [ { "data": { "text/html": [ "\n", " \n", " " ], "text/plain": [ "" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "IFrame('http://isbweb.org/data/invdyn/index.html', width='100%', height=400)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Load data file" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "ExecuteTime": { "end_time": "2020-05-13T00:43:50.414470Z", "start_time": "2020-05-13T00:43:50.210449Z" } }, "outputs": [], "source": [ "# load file with ground reaction force data\n", "grf = np.loadtxt('./../data/all.frc') # [Fx, Fy, COPx]\n", "# load file with kinematic data\n", "kin = np.loadtxt('./../data/all.kin') # [Hip(x,y), knee(x,y), ankle(x,y), toe(x,y)] \n", "freq = 10000\n", "time = np.linspace(0, grf.shape[0]/freq, grf.shape[0])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Data filtering" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "ExecuteTime": { "end_time": "2020-05-13T00:46:23.243568Z", "start_time": "2020-05-13T00:46:23.224289Z" } }, "outputs": [], "source": [ "# this is simulated data with no noise, filtering doesn't matter\n", "if False:\n", " # filter data\n", " from scipy.signal import butter, filtfilt\n", " # Butterworth filter\n", " b, a = butter(2, (10/(freq/2)))\n", " for col in np.arange(grf.shape[1]-1):\n", " grf[:, col] = filtfilt(b, a, grf[:, col])\n", " b, a = butter(2, (10/(freq/2)))\n", " for col in np.arange(kin.shape[1]): \n", " kin[:, col] = filtfilt(b, a, kin[:, col])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Data selection" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "ExecuteTime": { "end_time": "2020-05-13T00:46:44.028090Z", "start_time": "2020-05-13T00:46:44.021429Z" } }, "outputs": [], "source": [ "# heel strike occurs at sample 3001\n", "time = time[3001 - int(freq/40):-int(freq/20)]\n", "grf = grf[3001 - int(freq/40):-int(freq/20), :]\n", "kin = kin[3001 - int(freq/40):-int(freq/20), :]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Plot file data" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "ExecuteTime": { "end_time": "2020-05-13T00:46:58.045556Z", "start_time": "2020-05-13T00:46:56.899820Z" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "