Inverse Kinematic Study of Spatial Serial Manipulators using ANFIS Approach

Abstract

Inverse Kinematics (IK) is defined as the problem of determining a set of appropriate joint configurations for which the end effectors move to desired positions as smoothly, rapidly, and as accurately as possible. During the last decades, several methods and techniques, sophisticated or heuristic, such as Algebraic Method, Geometrical Method, Numerical Approach, Redundancy Resolution based concepts have been presented to produce fast and realistic solutions to the IK problem. However, most of the currently available methods suffer from high computational cost and production of unrealistic poses. This report reviews the most popular IK methods regarding their application functionality. Along with that, it enlightens the ongoing recent approach Adaptive Neuro-Fuzzy Logic Intelligent System (ANFIS), which is a combined effort of Neural Networks and Fuzzy logic. This approach ensures minimum drift to the values of Joint variables. In this report, a graphical comparison is drawn between the analytical solutions and ANFIS one. Also, desired path is planned by taking an example of SCARA manipulator having graphical comparison using MATLAB. It is noticed that finding joint variables through ANFIS is one of the most promising option, consequently for path planning too. Interestingly, it is found that without even finding the inverse kinematic equations for higher degree of freedom manipulators, the acceptable joint parameters can be obtained using ANFIS technique. Furthermore, the path planning for 5-DOFs Spatial Medical manipulator is presented for a certain path. Along with this, percentage error between desired value and ANFIS predicted value for x-, y- and z- coordinates is calculated. Thereafter, for a certain arm matrix corresponding to home position and orientation of spatial medical manipulator, a comparative analysis of inverse kinematic solution is explained. Finally, in the end of the thesis, obstacle avoidance using 3-DOFs manipulator with three obstacles is presented to show the effectiveness of proposed ANFIS approach