Authors : Wesam M. Jasim and Esam T. Yassen
Abstract: Finding the reliable and precise solution for inverse kinematics is considered as one of the most challenging problems in robotics manipulator. This is because the geometry of robot and the equations of nonlinear trigonometric that characterize the relationship between the cartesian space and the joint space. Thus, it has been a fertile area of research which has drawn researching efforts from various communities such as operation research and artificial intelligence. Additionally, solving this problem is necessary in real-time control. In this study, we investigate the performance of a new hybrid technique in solving the problem of the industrial manipulators inverse kinematics solution. The proposed hybrid technique combines an Artificial Neural Network (ANN) with fuzzy logic system (ANFIS). In order to evaluate the performance of the proposed model, the simulation to identify the joints of 3 and 6-DOF robots are utilized. The simulation results have showed the effectiveness of the hybrid model in solving the problem of the inverse kinematics. This demonstrates that the integration of the ANN and the fuzzy logic can fit the actual manipulator joints with an acceptable error.
Wesam M. Jasim and Esam T. Yassen, 2017. High Order Robotics Arm Modelling Based on ANFIS Technique. Journal of Engineering and Applied Sciences, 12: 8689-8694.