Mercurial > cortex

annotate org/movement.org @ 316:2c7fbcbd5ebb

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
replaced clojure.contrib.shell-out with clojure.java.shell
author Robert McIntyre <rlm@mit.edu>
date Tue, 28 Feb 2012 13:44:46 -0600
parents 5d448182c807
children bb3f8a4af87f
rev   line source
rlm@260 1 #+title: Simulated Muscles
rlm@158 2 #+author: Robert McIntyre
rlm@158 3 #+email: rlm@mit.edu
rlm@158 4 #+description: muscles for a simulated creature
rlm@158 5 #+keywords: simulation, jMonkeyEngine3, clojure
rlm@158 6 #+SETUPFILE: ../../aurellem/org/setup.org
rlm@158 7 #+INCLUDE: ../../aurellem/org/level-0.org
rlm@158 8
rlm@180 9
rlm@260 10 * Muscles
rlm@260 11
rlm@306 12 Surprisingly enough, terrestrial creatures only move by using torque
rlm@180 13 applied about their joints. There's not a single straight line of
rlm@180 14 force in the human body at all! (A straight line of force would
rlm@260 15 correspond to some sort of jet or rocket propulsion.)
rlm@180 16
rlm@278 17 In humans, muscles are composed of muscle fibers which can contract to
rlm@278 18 exert force. The muscle fibers which compose a muscle are partitioned
rlm@278 19 into discrete groups which are each controlled by a single alpha motor
rlm@306 20 neuron. A single alpha motor neuron might control as little as three
rlm@278 21 or as many as one thousand muscle fibers. When the alpha motor neuron
rlm@278 22 is engaged by the spinal cord, it activates all of the muscle fibers
rlm@278 23 to which it is attached. The spinal cord generally engages the alpha
rlm@278 24 motor neurons which control few muscle fibers before the motor neurons
rlm@306 25 which control many muscle fibers. This recruitment strategy allows
rlm@306 26 for precise movements at low strength. The collection of all motor
rlm@278 27 neurons that control a muscle is called the motor pool. The brain
rlm@278 28 essentially says "activate 30% of the motor pool" and the spinal cord
rlm@306 29 recruits motor neurons until 30% are activated. Since the
rlm@278 30 distribution of power among motor neurons is unequal and recruitment
rlm@278 31 goes from weakest to strongest, the first 30% of the motor pool might
rlm@278 32 be 5% of the strength of the muscle.
rlm@260 33
rlm@306 34 My simulated muscles follow a similar design: Each muscle is defined
rlm@267 35 by a 1-D array of numbers (the "motor pool"). Each entry in the array
rlm@306 36 represents a motor neuron which controls a number of muscle fibers
rlm@278 37 equal to the value of the entry. Each muscle has a scalar strength
rlm@278 38 factor which determines the total force the muscle can exert when all
rlm@278 39 motor neurons are activated. The effector function for a muscle takes
rlm@278 40 a number to index into the motor pool, and then "activates" all the
rlm@278 41 motor neurons whose index is lower or equal to the number. Each
rlm@267 42 motor-neuron will apply force in proportion to its value in the array.
rlm@267 43 Lower values cause less force. The lower values can be put at the
rlm@267 44 "beginning" of the 1-D array to simulate the layout of actual human
rlm@306 45 muscles, which are capable of more precise movements when exerting
rlm@306 46 less force. Or, the motor pool can simulate more exotic recruitment
rlm@306 47 strategies which do not correspond to human muscles.
rlm@260 48
rlm@260 49 This 1D array is defined in an image file for ease of
rlm@260 50 creation/visualization. Here is an example muscle profile image.
rlm@260 51
rlm@260 52 #+caption: A muscle profile image that describes the strengths of each motor neuron in a muscle. White is weakest and dark red is strongest. This particular pattern has weaker motor neurons at the beginning, just like human muscle.
rlm@260 53 [[../images/basic-muscle.png]]
rlm@260 54
rlm@260 55 * Blender Meta-data
rlm@260 56
rlm@260 57 In blender, each muscle is an empty node whose top level parent is
rlm@260 58 named "muscles", just like eyes, ears, and joints.
rlm@260 59
rlm@260 60 These functions define the expected meta-data for a muscle node.
rlm@180 61
rlm@261 62 #+name: muscle-meta-data
rlm@158 63 #+begin_src clojure
rlm@260 64 (in-ns 'cortex.movement)
rlm@158 65
rlm@180 66 (defvar
rlm@180 67 ^{:arglists '([creature])}
rlm@180 68 muscles
rlm@180 69 (sense-nodes "muscles")
rlm@180 70 "Return the children of the creature's \"muscles\" node.")
rlm@158 71
rlm@260 72 (defn muscle-profile-image
rlm@260 73 "Get the muscle-profile image from the node's blender meta-data."
rlm@260 74 [#^Node muscle]
rlm@260 75 (if-let [image (meta-data muscle "muscle")]
rlm@260 76 (load-image image)))
rlm@260 77
rlm@260 78 (defn muscle-strength
rlm@260 79 "Return the strength of this muscle, or 1 if it is not defined."
rlm@260 80 [#^Node muscle]
rlm@260 81 (if-let [strength (meta-data muscle "strength")]
rlm@260 82 strength 1))
rlm@260 83
rlm@260 84 (defn motor-pool
rlm@260 85 "Return a vector where each entry is the strength of the \"motor
rlm@260 86 neuron\" at that part in the muscle."
rlm@260 87 [#^Node muscle]
rlm@260 88 (let [profile (muscle-profile-image muscle)]
rlm@260 89 (vec
rlm@260 90 (let [width (.getWidth profile)]
rlm@260 91 (for [x (range width)]
rlm@260 92 (- 255
rlm@260 93 (bit-and
rlm@260 94 0x0000FF
rlm@260 95 (.getRGB profile x 0))))))))
rlm@260 96 #+end_src
rlm@260 97
rlm@273 98 Of note here is =motor-pool= which interprets the muscle-profile
rlm@260 99 image in a way that allows me to use gradients between white and red,
rlm@260 100 instead of shades of gray as I've been using for all the other
rlm@260 101 senses. This is purely an aesthetic touch.
rlm@260 102
rlm@260 103 * Creating Muscles
rlm@261 104 #+name: muscle-kernel
rlm@260 105 #+begin_src clojure
rlm@261 106 (in-ns 'cortex.movement)
rlm@261 107
rlm@260 108 (defn movement-kernel
rlm@180 109 "Returns a function which when called with a integer value inside a
rlm@191 110 running simulation will cause movement in the creature according
rlm@191 111 to the muscle's position and strength profile. Each function
rlm@191 112 returns the amount of force applied / max force."
rlm@260 113 [#^Node creature #^Node muscle]
rlm@260 114 (let [target (closest-node creature muscle)
rlm@158 115 axis
rlm@158 116 (.mult (.getWorldRotation muscle) Vector3f/UNIT_Y)
rlm@260 117 strength (muscle-strength muscle)
rlm@260 118
rlm@260 119 pool (motor-pool muscle)
rlm@260 120 pool-integral (reductions + pool)
rlm@296 121 forces
rlm@260 122 (vec (map #(float (* strength (/ % (last pool-integral))))
rlm@260 123 pool-integral))
rlm@158 124 control (.getControl target RigidBodyControl)]
rlm@296 125 (println-repl (.getName target) axis)
rlm@158 126 (fn [n]
rlm@260 127 (let [pool-index (max 0 (min n (dec (count pool))))
rlm@296 128 force (forces pool-index)]
rlm@191 129 (.applyTorque control (.mult axis force))
rlm@191 130 (float (/ force strength))))))
rlm@158 131
rlm@180 132 (defn movement!
rlm@180 133 "Endow the creature with the power of movement. Returns a sequence
rlm@180 134 of functions, each of which accept an integer value and will
rlm@180 135 activate their corresponding muscle."
rlm@158 136 [#^Node creature]
rlm@180 137 (for [muscle (muscles creature)]
rlm@260 138 (movement-kernel creature muscle)))
rlm@260 139 #+end_src
rlm@158 140
rlm@273 141 =movement-kernel= creates a function that will move the nearest
rlm@260 142 physical object to the muscle node. The muscle exerts a rotational
rlm@306 143 force dependent on it's orientation to the object in the blender
rlm@273 144 file. The function returned by =movement-kernel= is also a sense
rlm@260 145 function: it returns the percent of the total muscle strength that is
rlm@260 146 currently being employed. This is analogous to muscle tension in
rlm@260 147 humans and completes the sense of proprioception begun in the last
rlm@260 148 post.
rlm@260 149
rlm@260 150 * Visualizing Muscle Tension
rlm@306 151 Muscle exertion is a percent of a total, so the visualization is just a
rlm@260 152 simple percent bar.
rlm@260 153
rlm@261 154 #+name: visualization
rlm@260 155 #+begin_src clojure
rlm@191 156 (defn movement-display-kernel
rlm@191 157 "Display muscle exertion data as a bar filling up with red."
rlm@191 158 [exertion]
rlm@191 159 (let [height 20
rlm@191 160 width 300
rlm@191 161 image (BufferedImage. width height
rlm@191 162 BufferedImage/TYPE_INT_RGB)
rlm@191 163 fill (min (int (* width exertion)) width)]
rlm@191 164 (dorun
rlm@191 165 (for [x (range fill)
rlm@191 166 y (range height)]
rlm@191 167 (.setRGB image x y 0xFF0000)))
rlm@191 168 image))
rlm@191 169
rlm@191 170 (defn view-movement
rlm@191 171 "Creates a function which accepts a list of muscle-exertion data and
rlm@191 172 displays each element of the list to the screen."
rlm@191 173 []
rlm@191 174 (view-sense movement-display-kernel))
rlm@158 175 #+end_src
rlm@158 176
rlm@260 177 * Adding Touch to the Worm
rlm@158 178
rlm@278 179 To the worm, I add two new nodes which describe a single muscle.
rlm@277 180
rlm@277 181 #+attr_html: width=755
rlm@277 182 #+caption: The node highlighted in orange is the parent node of all muscles in the worm. The arrow highlighted in yellow represents the creature's single muscle, which moves the top segment. The other nodes which are not highlighted are joints, eyes, and ears.
rlm@277 183 [[../images/worm-with-muscle.png]]
rlm@277 184
rlm@283 185 #+name: test-movement
rlm@261 186 #+begin_src clojure
rlm@283 187 (defn test-worm-movement
rlm@283 188 ([] (test-worm-movement false))
rlm@261 189 ([record?]
rlm@261 190 (let [creature (doto (worm) (body!))
rlm@261 191
rlm@261 192 muscle-exertion (atom 0)
rlm@261 193 muscles (movement! creature)
rlm@261 194 muscle-display (view-movement)]
rlm@261 195 (.setMass
rlm@261 196 (.getControl (.getChild creature "worm-11") RigidBodyControl)
rlm@261 197 (float 0))
rlm@261 198 (world
rlm@261 199 (nodify [creature (floor)])
rlm@261 200 (merge standard-debug-controls
rlm@261 201 {"key-h"
rlm@261 202 (fn [_ value]
rlm@261 203 (if value
rlm@261 204 (swap! muscle-exertion (partial + 20))))
rlm@261 205 "key-n"
rlm@261 206 (fn [_ value]
rlm@261 207 (if value
rlm@261 208 (swap! muscle-exertion (fn [v] (- v 20)))))})
rlm@261 209 (fn [world]
rlm@261 210 (if record?
rlm@261 211 (Capture/captureVideo
rlm@261 212 world
rlm@261 213 (File. "/home/r/proj/cortex/render/worm-muscles/main-view")))
rlm@261 214 (light-up-everything world)
rlm@261 215 (enable-debug world)
rlm@261 216 (.setTimer world (RatchetTimer. 60))
rlm@261 217 (set-gravity world (Vector3f. 0 0 0))
rlm@261 218 (.setLocation (.getCamera world)
rlm@261 219 (Vector3f. -4.912815, 2.004171, 0.15710819))
rlm@261 220 (.setRotation (.getCamera world)
rlm@261 221 (Quaternion. 0.13828252, 0.65516764,
rlm@278 222 -0.12370994, 0.7323449)))
rlm@261 223 (fn [world tpf]
rlm@261 224 (muscle-display
rlm@261 225 (map #(% @muscle-exertion) muscles)
rlm@261 226 (if record?
rlm@261 227 (File. "/home/r/proj/cortex/render/worm-muscles/muscles"))))))))
rlm@261 228 #+end_src
rlm@261 229
rlm@261 230 * Video Demonstration
rlm@261 231
rlm@261 232 #+begin_html
rlm@261 233 <div class="figure">
rlm@261 234 <center>
rlm@261 235 <video controls="controls" width="550">
rlm@261 236 <source src="../video/worm-muscles.ogg" type="video/ogg"
rlm@261 237 preload="none" poster="../images/aurellem-1280x480.png" />
rlm@261 238 </video>
rlm@309 239 <br> <a href="http://youtu.be/8Rp4jEGMDWU"> YouTube </a>
rlm@261 240 </center>
rlm@261 241 <p>The worm is now able to move. The bar in the lower right displays
rlm@261 242 the power output of the muscle . Each jump causes 20 more motor neurons to
rlm@261 243 be recruited. Notice that the power output increases non-linearly
rlm@306 244 with motor neuron recruitment, similar to a human muscle.</p>
rlm@261 245 </div>
rlm@261 246 #+end_html
rlm@261 247
rlm@261 248 ** Making the Worm Muscles Video
rlm@261 249 #+name: magick7
rlm@261 250 #+begin_src clojure
rlm@261 251 (ns cortex.video.magick7
rlm@261 252 (:import java.io.File)
rlm@316 253 (:use clojure.java.shell))
rlm@261 254
rlm@261 255 (defn images [path]
rlm@261 256 (sort (rest (file-seq (File. path)))))
rlm@261 257
rlm@261 258 (def base "/home/r/proj/cortex/render/worm-muscles/")
rlm@261 259
rlm@261 260 (defn pics [file]
rlm@261 261 (images (str base file)))
rlm@261 262
rlm@261 263 (defn combine-images []
rlm@261 264 (let [main-view (pics "main-view")
rlm@261 265 muscles (pics "muscles/0")
rlm@261 266 targets (map
rlm@261 267 #(File. (str base "out/" (format "%07d.png" %)))
rlm@261 268 (range 0 (count main-view)))]
rlm@261 269 (dorun
rlm@261 270 (pmap
rlm@261 271 (comp
rlm@261 272 (fn [[ main-view muscles target]]
rlm@261 273 (println target)
rlm@261 274 (sh "convert"
rlm@261 275 main-view
rlm@261 276 muscles "-geometry" "+320+440" "-composite"
rlm@261 277 target))
rlm@261 278 (fn [& args] (map #(.getCanonicalPath %) args)))
rlm@261 279 main-view muscles targets))))
rlm@261 280 #+end_src
rlm@261 281
rlm@261 282 #+begin_src sh :results silent
rlm@261 283 cd ~/proj/cortex/render/worm-muscles
rlm@261 284 ffmpeg -r 60 -i out/%07d.png -b:v 9000k -c:v libtheora worm-muscles.ogg
rlm@261 285 #+end_src
rlm@158 286
rlm@260 287 * Headers
rlm@260 288 #+name: muscle-header
rlm@260 289 #+begin_src clojure
rlm@260 290 (ns cortex.movement
rlm@260 291 "Give simulated creatures defined in special blender files the power
rlm@260 292 to move around in a simulated environment."
rlm@260 293 {:author "Robert McIntyre"}
rlm@260 294 (:use (cortex world util sense body))
rlm@260 295 (:use clojure.contrib.def)
rlm@260 296 (:import java.awt.image.BufferedImage)
rlm@260 297 (:import com.jme3.scene.Node)
rlm@260 298 (:import com.jme3.math.Vector3f)
rlm@260 299 (:import com.jme3.bullet.control.RigidBodyControl))
rlm@260 300 #+end_src
rlm@260 301
rlm@261 302 #+name: test-header
rlm@261 303 #+begin_src clojure
rlm@261 304 (ns cortex.test.movement
rlm@261 305 (:use (cortex world util sense body movement))
rlm@261 306 (:use cortex.test.body)
rlm@261 307 (:use clojure.contrib.def)
rlm@261 308 (:import java.io.File)
rlm@261 309 (:import java.awt.image.BufferedImage)
rlm@261 310 (:import com.jme3.scene.Node)
rlm@283 311 (:import (com.jme3.math Quaternion Vector3f))
rlm@261 312 (:import (com.aurellem.capture Capture RatchetTimer))
rlm@261 313 (:import com.jme3.bullet.control.RigidBodyControl))
rlm@261 314 #+end_src
rlm@261 315
rlm@261 316 * Source Listing
rlm@261 317 - [[../src/cortex/movement.clj][cortex.movement]]
rlm@261 318 - [[../src/cortex/test/movement.clj][cortex.test.movement]]
rlm@261 319 - [[../src/cortex/video/magick7.clj][cortex.video.magick7]]
rlm@261 320 #+html: <ul> <li> <a href="../org/movement.org">This org file</a> </li> </ul>
rlm@261 321 - [[http://hg.bortreb.com ][source-repository]]
rlm@158 322
rlm@158 323 * COMMENT code generation
rlm@158 324 #+begin_src clojure :tangle ../src/cortex/movement.clj
rlm@261 325 <<muscle-header>>
rlm@261 326 <<muscle-meta-data>>
rlm@261 327 <<muscle-kernel>>
rlm@261 328 <<visualization>>
rlm@158 329 #+end_src
rlm@261 330
rlm@261 331 #+begin_src clojure :tangle ../src/cortex/test/movement.clj
rlm@261 332 <<test-header>>
rlm@261 333 <<test-movement>>
rlm@261 334 #+end_src
rlm@261 335
rlm@261 336 #+begin_src clojure :tangle ../src/cortex/video/magick7.clj
rlm@261 337 <<magick7>>
rlm@261 338 #+end_src