Real Time Parallel Robot Direct Kinematic Problem Computation Using Neural Networks

The calculation of the Direct Kinematic Problem (DKP) is one of the main issues in real-world applications of Parallel Robots, as iterative procedures have to be applied to compute the pose of the robot. Being this issue critical to robot Real-Time control, in this work a methodology to use Artificial Neural Networks to approximate the DKP is proposed and a comprehensive study is carried out to demonstrate experimentally the Real-Time performance benefits of the approach in a 3PRS parallel robot.

[1]  Jooyoung Park,et al.  Universal Approximation Using Radial-Basis-Function Networks , 1991, Neural Computation.

[2]  Mohammad Eghtesad,et al.  Neural network solution for forward kinematics problem of HEXA parallel robot , 2008, ACC.

[3]  Yunfeng Wang,et al.  A direct numerical solution to forward kinematics of general Stewart–Gough platforms , 2006, Robotica.

[4]  Clément Gosselin,et al.  On the computation of the direct kinematics of parallel manipulators using polynomial networks , 1998, IEEE Trans. Syst. Man Cybern. Part A.

[5]  Hamid D. Taghirad,et al.  Comparison of Different Methods for Computing the Forward Kinematics of a Redundant Parallel Manipulator , 2005, J. Intell. Robotic Syst..

[6]  Jorge Angeles,et al.  The direct kinematics of parallel manipulators under joint-sensor redundancy , 2000, IEEE Trans. Robotics Autom..

[7]  Junwei Han,et al.  Real-time state estimation for spatial six-degree-of-freedom linearly actuated parallel robots , 2009 .

[8]  Tae-Young Lee,et al.  Algebraic elimination-based real-time forward kinematics of the 6-6 Stewart platform with planar base and platform , 2001, Proceedings 2001 ICRA. IEEE International Conference on Robotics and Automation (Cat. No.01CH37164).

[9]  Jianhe Lei,et al.  Kinematic analysis of a novel 3-DOF actuation redundant parallel manipulator using artificial intelligence approach , 2011 .

[10]  Carlo Innocenti,et al.  Forward Kinematics in Polynomial Form of the General Stewart Platform , 2001 .

[11]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[12]  Shimin Wei,et al.  Closed-form forward kinematics for a symmetrical 6-6 Stewart platform using algebraic elimination , 2010 .

[13]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[14]  Myung-Chul Han,et al.  The estimation for forward kinematic solution of Stewart platform using the neural network , 1999, Proceedings 1999 IEEE/RSJ International Conference on Intelligent Robots and Systems. Human and Environment Friendly Robots with High Intelligence and Emotional Quotients (Cat. No.99CH36289).

[15]  Jeha Ryu,et al.  A closed-form solution to the direct kinematics of nearly general parallel manipulators with optimally located three linear extra sensors , 2001, IEEE Trans. Robotics Autom..

[16]  Clément Gosselin,et al.  Kinematic Analysis and Optimization of a New Three Degree-of-Freedom Spatial Parallel Manipulator , 2000 .

[17]  Jean-Pierre Merlet,et al.  Parallel Robots , 2000 .

[18]  Kah Bin Lim,et al.  Forward kinematics solution of Stewart platform using neural networks , 1997, Neurocomputing.