Home > Published Issues > 2016 > Volume 5, No. 2, April 2016 >

Proper Jacobian Pseudo Inverse Neural Network Matrix Method Applied to Robot Inverse Kinematics Controlling

Adrian D. Olaru 1, Serban A. Olaru 2, and Niculae N. Mihai 3
1. Machine and Manufacturing Systems, University Politehnica, Bucharest, Romania
2. Department of Mechatronics, RomSYS, Bucharest, Romania
3. Technoaccord, Quebec, Canada

Abstract—These in the controlling of the space movement of the end effecter and the robot’s joints one of the most important problem is to know with the extreme precision the joints relative displacement. Controlling this displacement must be done only by controlling with accuracy the movement of each motor in the joints by solving the forward and inverse kinematics problem. This problem could be solves on-line, but usually it is solved off line. One of the most precise method to solve the inverse kinematics problem in the robots with redundant chain is the complex coupled method, usually by coupling the neural network with Jacobian method. In this paper was proposed and used the proper coupled method Iterative Pseudo Inverse Jacobian Matrix Method (IPIJMM) with Sigmoid Bipolar Hyperbolic Tangent Neural Network with Time Delay and Recurrent Links (SBHTNN-TDRL). The solution of the inverse kinematics problem is very difficult to find when the degree of freedom increase and in many cases this is impossible because there are the redundant solutions. In all these cases must be used the numerical iterative approximation, like the proposed method, with artificial intelligence algorithm. The paper describe all virtual LabVIEW instrumentation, the needed steps in one case study, to obtain the space curves in different planes by using one arm type robot and the proposed algorithm. The errors of the space movement of the robot end-effecter, after applying the proposed method, was less than 0,001. The presented method and the Virtual Instrumentation (VI) are generally and they can be used in all other robots type application and for all other conventional and unconventional space curves.
Index Terms—iterative matrix method, forward and inverse kinematics, jacobianmatrix, pseudo inverse matrix, assisted research, neural network, LabVIEW instrumentation, gekodrivesystem

Cite: Adrian D. Olaru, Serban A. Olaru, and Niculae N. Mihai, "Proper Jacobian Pseudo Inverse Neural Network Matrix Method Applied to Robot Inverse Kinematics Controlling," International Journal of Mechanical Engineering and Robotics Research, Vol. 5, No. 2, pp. 120-123, April 2016. DOI: 10.18178/ijmerr.5.2.120-123