1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432
|
/* -------------------------------------------------------------------------- *
* Simbody(tm) *
* -------------------------------------------------------------------------- *
* This is part of the SimTK biosimulation toolkit originating from *
* Simbios, the NIH National Center for Physics-Based Simulation of *
* Biological Structures at Stanford, funded under the NIH Roadmap for *
* Medical Research, grant U54 GM072970. See https://simtk.org/home/simbody. *
* *
* Portions copyright (c) 2014 Stanford University and the Authors. *
* Authors: Chris Dembia *
* Contributors: Michael Sherman *
* *
* Licensed under the Apache License, Version 2.0 (the "License"); you may *
* not use this file except in compliance with the License. You may obtain a *
* copy of the License at http://www.apache.org/licenses/LICENSE-2.0. *
* *
* Unless required by applicable law or agreed to in writing, software *
* distributed under the License is distributed on an "AS IS" BASIS, *
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. *
* See the License for the specific language governing permissions and *
* limitations under the License. *
* -------------------------------------------------------------------------- */
#include "Simbody.h"
#include "TaskSpace.h"
using namespace SimTK;
//==============================================================================
// Jacobian
//==============================================================================
void TaskSpace::Jacobian::updateCache(Matrix& cache) const
{
m_tspace->getMatterSubsystem().calcStationJacobian(
getState(),
m_tspace->getMobilizedBodyIndices(),
m_tspace->getStations(),
cache);
}
const TaskSpace::JacobianTranspose& TaskSpace::Jacobian::transpose() const
{
return m_tspace->getJacobianTranspose(getState());
}
Vector TaskSpace::Jacobian::operator*(const Vector& u) const
{
Vector_<Vec3> Ju;
m_tspace->getMatterSubsystem().multiplyByStationJacobian(getState(),
m_tspace->getMobilizedBodyIndices(), m_tspace->getStations(),
u, Ju);
// Convert to a Vector.
Vector out(3 * Ju.size());
for (int i = 0; i < Ju.size(); ++i) {
out[3 * i] = Ju[i][0];
out[3 * i + 1] = Ju[i][1];
out[3 * i + 2] = Ju[i][2];
}
return out;
}
//==============================================================================
// JacobianTranspose
//==============================================================================
void TaskSpace::JacobianTranspose::updateCache(Matrix& cache) const
{
cache = transpose().value().transpose();
}
const TaskSpace::Jacobian& TaskSpace::JacobianTranspose::transpose() const
{
return m_tspace->getJacobian(getState());
}
Vector TaskSpace::JacobianTranspose::operator*(const Vector_<Vec3>& f_GP) const
{
Vector f;
m_tspace->getMatterSubsystem().multiplyByStationJacobianTranspose(
getState(),
m_tspace->getMobilizedBodyIndices(),
m_tspace->getStations(),
f_GP,
f);
return f;
}
Vector TaskSpace::JacobianTranspose::operator*(const Vector& f_GP) const
{
unsigned int nIn = f_GP.size();
SimTK_APIARGCHECK1_ALWAYS(nIn % 3 == 0,
"TaskSpace::JacobianTranspose", "operator*",
"Length of f_GP, which is %i, is not divisible by 3.", nIn);
unsigned int nOut = nIn / 3;
// Create the Vector_<Vec3>.
// TODO debug, or look for methods that already do this.
Vector_<Vec3> my_f_GP(nOut);
for (unsigned int i = 0; i < nOut; ++i)
{
// getAs is just a recast; doesn't copy.
my_f_GP[i] = Vec3::getAs(&f_GP[3 * i]);
}
// Perform the multiplication.
return operator*(my_f_GP);
}
Vector TaskSpace::JacobianTranspose::operator*(const Vec3& f_GP) const
{
return operator*(Vector_<Vec3>(1, f_GP));
}
Matrix TaskSpace::JacobianTranspose::operator*(const Matrix& f_GP) const
{
unsigned int nrow = getState().getNU();
unsigned int ncol = f_GP.ncol();
Matrix out(nrow, ncol);
for (unsigned int j = 0; j < ncol; ++j)
{
// TODO is this cast inefficient? Is it copying?
out(j) = operator*(Vector(f_GP(j)));
}
return out;
}
Matrix TaskSpace::JacobianTranspose::operator*(
const TaskSpace::Inertia& Lambda) const
{
// TOOD could be more efficient.
return operator*(Lambda.value());
}
Matrix TaskSpace::JacobianTranspose::operator*(
const TaskSpace::DynamicallyConsistentJacobianInverseTranspose& JBarT) const
{
return operator*(JBarT.value());
}
//==============================================================================
// Inertia
//==============================================================================
void TaskSpace::Inertia::updateCache(Matrix& cache) const
{
FactorLU inertiaInverse(m_tspace->getInertiaInverse(getState()).value());
inertiaInverse.inverse(cache);
}
const TaskSpace::InertiaInverse& TaskSpace::Inertia::inverse() const
{
return m_tspace->getInertiaInverse(getState());
}
Vector TaskSpace::Inertia::operator*(const Vector& a) const
{
return value() * a;
}
Vector TaskSpace::Inertia::operator*(const Vec3& a) const
{
return operator*(Vector(a));
}
//==============================================================================
// InertiaInverse
//==============================================================================
void TaskSpace::InertiaInverse::updateCache(Matrix& cache) const
{
const SimbodyMatterSubsystem& matter = m_tspace->getMatterSubsystem();
const JacobianTranspose& JT = m_tspace->JT(getState());
// TODO const Matrix& JT = m_tspace.JT().value();
const Matrix& J = m_tspace->J(getState()).value();
/* TODO
// TODO cache the result.
unsigned int nst = m_tspace.getNumScalarTasks();
unsigned int nu = m_tspace.getState().getNU();
Matrix J = m_tspace.getJacobian().value();
Matrix MInvJt(nu, nst);
for (unsigned int j = 0; j < nst; ++j)
{
matter.multiplyByMInv(m_tspace.getState(), J.transpose()(j), MInvJt(j));
}
updCacheValue() = J * MInvJt;
*/
unsigned int nt = m_tspace->getNumTasks();
unsigned int nst = m_tspace->getNumScalarTasks();
unsigned int nu = getState().getNU();
Matrix& inertiaInverse = cache;
inertiaInverse.resize(nst, nst);
// Create temporary variables.
Vector Jtcol(nu);
Vector MInvJtcol(nu);
Vector_<Vec3> JMInvJt_j(nt);
// f_GP is used to pluck out one column at a time of Jt. Exactly one
// element at a time of f_GP will be 1, the rest are 0.
Vector f_GP(nst, Real(0));
for (unsigned int j = 0; j < nst; ++j)
{
f_GP[j] = 1;
Jtcol = JT * f_GP;
f_GP[j] = 0;
matter.multiplyByMInv(getState(), Jtcol, MInvJtcol);
// TODO replace with operator.
inertiaInverse(j) = J * MInvJtcol;
/* TODO
matter.multiplyByStationJacobian(m_tspace.getState(),
m_tspace.getMobilizedBodyIndices(), m_tspace.getStations(),
MInvJtcol, JMInvJt_j);
inertiaInverse(j) = JMInvJt_j;
*/
}
}
const TaskSpace::Inertia& TaskSpace::InertiaInverse::inverse() const
{
return m_tspace->getInertia(getState());
}
//==============================================================================
// DynamicallyConsistentJacobianInverse
//==============================================================================
void TaskSpace::DynamicallyConsistentJacobianInverse::updateCache(Matrix& cache)
const
{
const JacobianTranspose& JT = m_tspace->getJacobianTranspose(getState());
const Inertia& Lambda = m_tspace->getInertia(getState());
// TODO inefficient?
Matrix JtLambda = JT * Lambda;
unsigned int nst = m_tspace->getNumScalarTasks();
unsigned int nu = getState().getNU();
Matrix& Jbar = cache;
Jbar.resize(nu, nst);
for (unsigned int j = 0; j < nst; ++j)
{
m_tspace->getMatterSubsystem().multiplyByMInv(getState(),
JtLambda(j), Jbar(j));
}
}
const TaskSpace::DynamicallyConsistentJacobianInverseTranspose&
TaskSpace::DynamicallyConsistentJacobianInverse::transpose() const
{
return m_tspace->getDynamicallyConsistentJacobianInverseTranspose(getState());
}
Vector TaskSpace::DynamicallyConsistentJacobianInverse::operator*(
const Vector& vec) const
{
const JacobianTranspose& JT = m_tspace->getJacobianTranspose(getState());
const Inertia& Lambda = m_tspace->getInertia(getState());
// TODO where is this even used? TODO test this.
Vector JBarvec;
m_tspace->getMatterSubsystem().multiplyByMInv(getState(),
JT * (Lambda * vec), JBarvec);
return JBarvec;
}
Matrix TaskSpace::DynamicallyConsistentJacobianInverse::operator*(
const Matrix& mat) const
{
unsigned int nrow = getState().getNU();
unsigned int ncol = mat.ncol();
Matrix out(nrow, ncol);
for (unsigned int j = 0; j < ncol; ++j)
{
out(j) = operator*(mat(j).getAsVector());
}
return out;
}
//==============================================================================
// DynamicallyConsistentJacobianInverseTranspose
//==============================================================================
void TaskSpace::DynamicallyConsistentJacobianInverseTranspose::updateCache(
Matrix& cache) const
{
cache = transpose().value().transpose();
}
const TaskSpace::DynamicallyConsistentJacobianInverse&
TaskSpace::DynamicallyConsistentJacobianInverseTranspose::transpose() const
{
return m_tspace->getDynamicallyConsistentJacobianInverse(getState());
}
Vector TaskSpace::DynamicallyConsistentJacobianInverseTranspose::operator*(
const Vector& g) const
{
// TODO inefficient. can we have an MInvT operator??
return value() * g;
}
//==============================================================================
// InertialForces
//==============================================================================
void TaskSpace::InertialForces::updateCache(Vector& cache) const
{
Vector jointSpaceInertialForces;
m_tspace->getMatterSubsystem().calcResidualForceIgnoringConstraints(
getState(), Vector(0), Vector_<SpatialVec>(0), Vector(0),
jointSpaceInertialForces);
Vector JDotu;
m_tspace->getMatterSubsystem().calcBiasForStationJacobian(
getState(),
m_tspace->getMobilizedBodyIndices(), m_tspace->getStations(),
JDotu);
const DynamicallyConsistentJacobianInverseTranspose& JBarT =
m_tspace->JBarT(getState());
const Vector& b = jointSpaceInertialForces;
const Inertia& Lambda = m_tspace->Lambda(getState());
cache = JBarT * b - Lambda * JDotu;
}
Vector TaskSpace::InertialForces::operator+(const Vector& f) const
{
return value() + f;
}
//==============================================================================
// Gravity
//==============================================================================
void TaskSpace::Gravity::updateCache(Vector& cache) const
{
cache = m_tspace->JBarT(getState()) * systemGravity();
}
Vector TaskSpace::Gravity::operator+(const Vector& f) const
{
return value() + f;
}
Vector TaskSpace::Gravity::systemGravity() const
{
// TODO where does this go? Make a separate class?
Vector g;
m_tspace->getMatterSubsystem().multiplyBySystemJacobianTranspose(
getState(),
m_tspace->m_gravityForce.getBodyForces(getState()),
g);
// Negate, since we want the 'g' that appears on the same side of the
// equations of motion as does the mass matrix. That is, M udot + C + g = F
return -g;
}
//==============================================================================
// NullspaceProjection
//==============================================================================
void TaskSpace::NullspaceProjection::updateCache(Matrix& cache) const
{
cache = transpose().value().transpose();
}
const TaskSpace::NullspaceProjectionTranspose&
TaskSpace::NullspaceProjection::transpose() const
{
return m_tspace->getNullspaceProjectionTranspose(getState());
}
Vector TaskSpace::NullspaceProjection::operator*(const Vector& vec)
const
{
return vec - (m_tspace->JBar(getState()) * (m_tspace->J(getState()) * vec));
}
//==============================================================================
// NullspaceProjectionTranspose
//==============================================================================
void TaskSpace::NullspaceProjectionTranspose::updateCache(Matrix& cache) const
{
cache = 1 - (m_tspace->JT(getState()) * m_tspace->JBarT(getState()));
}
const TaskSpace::NullspaceProjection&
TaskSpace::NullspaceProjectionTranspose::transpose() const
{
return m_tspace->getNullspaceProjection(getState());
}
Vector TaskSpace::NullspaceProjectionTranspose::operator*(const Vector& vec)
const
{
return vec - (m_tspace->JT(getState()) * (m_tspace->JBarT(getState()) * vec));
}
//==============================================================================
// TaskSpace
//==============================================================================
// TODO account for applied forces? velocities?
Vector TaskSpace::calcInverseDynamics(const State& s, const Vector& taskAccelerations) const
{
return Lambda(s) * taskAccelerations + mu(s) + p(s);
}
|