@techreport{TR-IC-PFG-17-07,
   number = {IC-PFG-17-07},
   author  =  {Luiz  Fernando  Cirigliano  Villela  \and  Esther  Luna 
                   Colombini},
   title   =  {{Humanoid  Robot  Walking  Optimization  using  Genetic 
                   Algorithms}},
   month = {July},
   year = {2017},
   institution = {Institute of Computing, University of Campinas},
   note = {In English, 15 pages.
    \par\selectlanguage{english}\textbf{Resumo}
       The  problem  of  creating fast and stable walking for humanoid
       robots is a very complex one due to the large degree of freedom
       and  the  external  variables  involved. This work tackles this
       problem  by  using  a  model  free approach where each joint is
       represented by a sinusoidal function of time. The parameters of
       the  functions  for  all  actuated joints are optimized using a
       genetic  algorithm. Experiments were performed with a NAO robot
       in a simulated environment under V-REP. The optimized robot was
       able to walk at a speed of $54 cm/s$ in a straight line and for
       up to 200 meters without falling. Experiments were also carried
       out  to evaluate the individuals capacity to adapt to different
       scenarios,  such  as  walking up and down ramps. Results showed
       different  movement  patterns,  a  slower pace and more upright
       positions for the robot walking uphill.
  }
}