A Method for the Approximation of the Multiple IK Solutions of Regular Manipulators Based on the Uniqueness Domains and Using MLP

  • Vassilis C. Moulianitis
  • Evgenios M. Kokkinopoulos
  • Nikos A. Aspragathos
Conference paper
Part of the Mechanisms and Machine Science book series (Mechan. Machine Science, volume 37)


In this paper, a method for dividing the training data as well as training MLP systems to obtain the multiple solutions of IKP of regular manipulators is presented. The sets of training data for each system are strictly defined using the concept of uniqueness domains. The training data are obtained by the forward kinematics and the sign of the determinant of the manipulator Jacobian is used for the determination of the uniqueness domains. An illustrative example with a 3 dof robot with known IK solutions is presented for the verification of the proposed approach.


Inverse kinematics Multiple solutions MLP Regular manipulators 


  1. 1.
    de Angulo, V., Torras, C.: Learning inverse kinematics: reduced sampling through decomposition into virtual robots. IEEE Trans. Syst. Man Cybern. Part B: Cybern. 38(6), 1571–1577 (2008). doi: 10.1109/TSMCB.2008.928232
  2. 2.
    Burdick, J.W.: A classification of 3R regional manipulator singularities and geometries. Mech. Mach. Theory 30(1), 71–89 (1995). doi: 10.1016/0094-114X(94)00043-K
  3. 3.
    Gotlih, K., Troch, I.: Base invariance of the manipulability index. Robotica 22(4), 455–462 (2004). doi: 10.1017/S0263574704000220
  4. 4.
    Kalra, P., Mahapatra, P., Aggarwal, D.: An evolutionary approach for solving the multimodal inverse kinematics problem of industrial robots. Mech. Mach. Theory 41(10), 1213–1229 (2006). doi:http://dx.doi.org/10.1016/j.mechmachtheory.2005.11.005. http://www.sciencedirect.com/science/article/pii/S0094114X05002053
  5. 5.
    Lazarevska, E.: A neuro-fuzzy model of the inverse kinematics of a 4 dof robotic arm. In: Proceedings of 14th International Conference on Computer Modelling and Simulation (UKSim), 2012 UKSim, pp. 306–311 (2012). doi: 10.1109/UKSim.2012.51
  6. 6.
    Mavroidis, C., Roth, B.: Structural parameters which reduce the number of manipulator configurations. J. Mech. Des. 116(1), 3–10 (1994). doi:http://dx.doi.org/10.1115/1.2919373
  7. 7.
    Nielsen, J., Roth, B.: On the kinematic analysis of robotic mechanisms. Int. J. Robot. Res. 18(12), 1147–1160 (1999). doi: 10.1177/02783649922067771. http://ijr.sagepub.com/content/18/12/1147.abstract
  8. 8.
    Raptis, S.N., Tzafestas, E.S.: Robot inverse kinematics via neural and neurofuzzy networks: architectural and computational aspects for improved performance. J. Inf. Optim. Sci. 28(6), 905–933 (2007). doi: 10.1080/02522667.2007.10699780. http://dx.doi.org/10.1080/02522667.2007.10699780
  9. 9.
    Srinivasan, A., Nigam, M.: Neuro-fuzzy based approach for inverse kinematics solution of industrial robot manipulators. Int. J. Comput. Commun. Control 3(3), 224–234 (2008)Google Scholar
  10. 10.
    Tabandeh, S., Clark, C., Melek, W.: A genetic algorithm approach to solve for multiple solutions of inverse kinematics using adaptive niching and clustering. In: Proceedings of the IEEE Congress on Evolutionary Computation, 2006. CEC 2006, pp. 1815–1822 (2006). doi: 10.1109/CEC.2006.1688527
  11. 11.
    Wenger, P.: Uniqueness domains and regions of feasible paths for cuspidal manipulators. IEEE Trans. Robot. 20(4), 745–750 (2004). doi: 10.1109/TRO.2004.829467

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Vassilis C. Moulianitis
    • 1
  • Evgenios M. Kokkinopoulos
    • 2
  • Nikos A. Aspragathos
    • 2
  1. 1.Department of Product and Systems Design EngineeringUniversity of the AegeanErmoupolisGreece
  2. 2.Mechanical Engineering and Aeronautics DepartmentUniversity of PatrasRio, PatrasGreece

Personalised recommendations