chainiksolverpos_lma_demo.cpp

Go to the documentation of this file.

 
/**************************************************************************
    begin                : May 2011
    copyright            : (C) 2011 Erwin Aertbelien
    email                : firstname.lastname@mech.kuleuven.ac.be
 
 History (only major changes)( AUTHOR-Description ) :
 
 ***************************************************************************
 *   This library is free software; you can redistribute it and/or         *
 *   modify it under the terms of the GNU Lesser General Public            *
 *   License as published by the Free Software Foundation; either          *
 *   version 2.1 of the License, or (at your option) any later version.    *
 *                                                                         *
 *   This library is distributed in the hope that it will be useful,       *
 *   but WITHOUT ANY WARRANTY; without even the implied warranty of        *
 *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
 *   Lesser General Public License for more details.                       *
 *                                                                         *
 *   You should have received a copy of the GNU Lesser General Public      *
 *   License along with this library; if not, write to the Free Software   *
 *   Foundation, Inc., 59 Temple Place,                                    *
 *   Suite 330, Boston, MA  02111-1307  USA                                *
 *                                                                         *
 ***************************************************************************/
 
#include <iostream>
#include <frames_io.hpp>
#include <models.hpp>
#include <chainiksolverpos_lma.hpp>
#include <chainfksolverpos_recursive.hpp>
#include <utilities/utility.h>
 
#include <boost/version.hpp>
#if BOOST_VERSION < 108300
#include <boost/timer.hpp>
#else
#include <boost/timer/timer.hpp>
#endif
 
void test_inverseposkin(KDL::Chain& chain) {
#if BOOST_VERSION < 108300
    boost::timer timer;
#else
    boost::timer::cpu_timer timer;
#endif
    int num_of_trials = 1000000;
    int total_number_of_iter = 0;
    int n = chain.getNrOfJoints();
    int nrofresult_ok = 0;
    int nrofresult_minus1=0;
    int nrofresult_minus2=0;
    int nrofresult_minus3=0;
    int min_num_of_iter = 10000000;
    int max_num_of_iter = 0;
    double min_diff = 1E10;
    double max_diff = 0;
    double min_trans_diff = 1E10;
    double max_trans_diff = 0;
    double min_rot_diff = 1E10;
    double max_rot_diff = 0;
    Eigen::Matrix<double,6,1> L;
    L(0)=1;L(1)=1;L(2)=1;
    L(3)=0.01;L(4)=0.01;L(5)=0.01;
    KDL::ChainFkSolverPos_recursive fwdkin(chain);
    KDL::ChainIkSolverPos_LMA solver(chain,L);
    KDL::JntArray q(n);
    KDL::JntArray q_init(n);
    KDL::JntArray q_sol(n);
    for (int trial=0;trial<num_of_trials;++trial) {
        q.data.setRandom();
        q.data *= KDL::PI;
        q_init.data.setRandom();
        q_init.data *= KDL::PI;
        KDL::Frame pos_goal,pos_reached;
        fwdkin.JntToCart(q,pos_goal);
        //solver.compute_fwdpos(q.data);
        //pos_goal = solver.T_base_head;
        int retval;
        retval = solver.CartToJnt(q_init,pos_goal,q_sol);
        switch (retval) {
        case 0:
            nrofresult_ok++;
            break;
        case -1:
            nrofresult_minus1++;
            break;
        case -2:
            nrofresult_minus2++;
            break;
        case -3:
            nrofresult_minus3++;
            break;
        }
        if (retval !=0) {
            std::cout << "-------------- failed ----------------------------" << std::endl;
            std::cout << "pos " << pos_goal << std::endl;
            std::cout << "reached pos " << solver.T_base_head << std::endl;
            std::cout << "TF from pos to head \n" << pos_goal.Inverse()*solver.T_base_head << std::endl;
            std::cout << "gradient " << solver.grad.transpose() << std::endl;
            std::cout << "q   " << q.data.transpose()/M_PI*180.0 << std::endl;
            std::cout << "q_sol " << q_sol.data.transpose()/M_PI*180.0 << std::endl;
            std::cout << "q_init " << q_init.data.transpose()/M_PI*180.0 << std::endl;
            std::cout << "return value " << retval << std::endl;
            std::cout << "last #iter " << solver.lastNrOfIter << std::endl;
            std::cout << "last diff  " << solver.lastDifference << std::endl;
            std::cout << "jacobian of goal values ";
            solver.display_jac(q);
            std::cout << "jacobian of solved values ";
            solver.display_jac(q_sol);
        }
        total_number_of_iter += solver.lastNrOfIter;
        if (solver.lastNrOfIter > max_num_of_iter) max_num_of_iter = solver.lastNrOfIter;
        if (solver.lastNrOfIter < min_num_of_iter) min_num_of_iter = solver.lastNrOfIter;
        if (retval!=-3) {
            if (solver.lastDifference > max_diff) max_diff = solver.lastDifference;
            if (solver.lastDifference < min_diff) min_diff = solver.lastDifference;
 
            if (solver.lastTransDiff > max_trans_diff) max_trans_diff = solver.lastTransDiff;
            if (solver.lastTransDiff < min_trans_diff) min_trans_diff = solver.lastTransDiff;
 
            if (solver.lastRotDiff > max_trans_diff) max_rot_diff = solver.lastRotDiff;
            if (solver.lastRotDiff < min_trans_diff) min_rot_diff = solver.lastRotDiff;
        }
        fwdkin.JntToCart(q_sol,pos_reached);
    }
    std::cout << "------------------ statistics ------------------------------"<<std::endl;
    std::cout << "#times successful " << nrofresult_ok << std::endl;
    std::cout << "#times -1 result " << nrofresult_minus1 << std::endl;
    std::cout << "#times -2 result " << nrofresult_minus2 << std::endl;
    std::cout << "#times -3 result " << nrofresult_minus3 << std::endl;
    std::cout << "average number of iter " << (double)total_number_of_iter/(double)num_of_trials << std::endl;
    std::cout << "min. nr of iter " << min_num_of_iter << std::endl;
    std::cout << "max. nr of iter " << max_num_of_iter << std::endl;
    std::cout << "min. difference after solving " << min_diff << std::endl;
    std::cout << "max. difference after solving " << max_diff << std::endl;
    std::cout << "min. trans. difference after solving " << min_trans_diff << std::endl;
    std::cout << "max. trans. difference after solving " << max_trans_diff << std::endl;
    std::cout << "min. rot. difference after solving " << min_rot_diff << std::endl;
    std::cout << "max. rot. difference after solving " << max_rot_diff << std::endl;
#if BOOST_VERSION < 108300
    double el = timer.elapsed();
#else
    boost::timer::cpu_times const ct(timer.elapsed());
    double el = ct.user / 1e9;
#endif
    std::cout << "elapsed time " << el << std::endl;
    std::cout << "estimate of average time per invposkin (ms)" << el/num_of_trials*1000 << std::endl;
    std::cout << "estimate of longest time per invposkin (ms) " << el/total_number_of_iter*max_num_of_iter *1000 << std::endl;
    std::cout << "estimate of shortest time per invposkin (ms) " << el/total_number_of_iter*min_num_of_iter *1000 << std::endl;
}
 
int main(int argc,char* argv[]) {
    std::cout << " This example generates random joint positions, applies a forward kinematic transformation,\n"
              <<  " and calls ChainIkSolverPos_LMA on the resulting pose.  In this way we are sure that\n"
              <<  " the resulting pose is reachable.  However, some percentage of the trials will be at\n"
              <<  " near singular position, where it is more difficult to achieve convergence and accuracy\n"
              <<  " The initial position given to the algorithm is also a random joint position\n"
              <<  " This routine asks for an accuracy of 10 micrometer, one order of magnitude better than a\n"
              <<  " a typical industrial robot.\n"
              <<  " This routine can take more then 6 minutes to execute. It then gives statistics on execution times\n"
              <<  " and failures.\n"
              <<  " Typically when executed 1 000 000 times, you will still see some small amount of failures\n"
              <<  " Typically these failures are in the neighbourhood of singularities.  Most failures of type -2 still\n"
              <<  " reach an accuracy better than 1E-4.\n"
              <<  " This is much better than ChainIkSolverPos_NR, which fails a few times per 100 trials.\n";
 
    KDL::Chain chain;
    chain = KDL::Puma560();
    //chain = KDL::KukaLWR_DHnew();
 
    KDL::ChainIkSolverPos_LMA solver(chain);
 
    test_inverseposkin(chain);
 
    return 0;
}