Inverse Kinematics Analysis of Serial Manipulators Using Genetic Algorithms

This paper gives the insight into inverse kinematics analysis of industrial serial manipulators of different degree of freedom, utilizing homogeneous transformation matrices, Denavit-Hartenberg (D-H) parameters, and the forward kinematics. For a robot with 2/3/4 degrees of freedom, inverse kinematics can be determined by using simple geometry with (sin, cos, tan) trigonometry functions, first by solving the forward kinematics, and then creating simultaneous equations solving those for the joint angles. For a robot with six joints (spherical wrist) or more joints (redundant robots) the inverse kinematics becomes complex to solve manually, in such cases the inverse kinematics will be treated as an optimization problem which can be solved numerically. In the proposed approach, MATLAB's `ga' solver was used to solve the inverse kinematics of serial manipulators. The joint angles obtained from optimization methods (ga), (pso), and algebraic methods are compared for two degrees of freedom R-R manipulator. A case study on ABB IRB 1600-1.45 serial manipulator with 6-DOF is presented in this paper.