• DocumentCode
    2100418
  • Title

    Optimum method of working assignment in some welding robots with genetic algorithm

  • Author

    Kaneko, Yasuyoshi ; Hosoda, Tatsuo ; Ohshima, K.

  • Author_Institution
    Dept. of Electr. & Electron. Syst., Saitama Univ., Japan
  • Volume
    1
  • fYear
    2001
  • fDate
    2001
  • Firstpage
    232
  • Abstract
    Deals with the optimisation of the task assignment of welding robots working cooperatively. In shipbuilding, in order to weld an assembling part to the bridge beam, etc. by using many welding robots efficiently in a short time, it is important to optimize the task assignment of these welding robots. In this paper, we propose a method to optimize the task assignment with a genetic algorithm. In order to obtain not only the efficient task assignment of the welding robot but also a good welding result, the knowledge of welding in order to prevent heat deformation and union defects are added to the fitness parameters of the genetic algorithm, and this welding knowledge is used to create the initial population of the genetic algorithm, because the time to search for the optimum answer is short. The validity of this proposed method was verified by performing some simulated experiments of simple work with two robots
  • Keywords
    assembling; genetic algorithms; industrial robots; multi-robot systems; scheduling; search problems; ships; welding; assembling; bridge beam; cooperatively working welding robots; efficiency; fitness parameters; genetic algorithm; heat deformation; initial population; optimum work assignment method; search time; shipbuilding; task assignment optimisation; union defects; Bridges; Genetic algorithms; Marine vehicles; Optimization methods; Production facilities; Proposals; Robotic assembly; Service robots; Shipbuilding industry; Welding;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Industrial Electronics Society, 2001. IECON '01. The 27th Annual Conference of the IEEE
  • Conference_Location
    Denver, CO
  • Print_ISBN
    0-7803-7108-9
  • Type

    conf

  • DOI
    10.1109/IECON.2001.976485
  • Filename
    976485