Artificial Neural Networks Based Inverse Kinematics Solution and Simulation of a Six Degree of Freedom Lighting Manipulator

Abstract

Inverse kinematics problem of a robot manipulator is calculation of joint variables of the manipulator. Inverse kinematic is a challenging problem of robots. The inverse kinematics in robotic is the determination of joint angles for a desired position of the end effector. This paper proposes an Artificial Neural Network (ANN) model to find Inverse kinematic solution of a 6- DOF manipulator, which is designed instead of a manually lighting system that is used in surgery rooms. After obtaining forward kinematic equations, these equations were used to derive training data of ANN. The manipulator was designed in a computer aided design (CAD) program and this CAD model transferred to Simulink environment for a realistic and visual simulation purpose. To evaluate the trained ANN performance visually, this Simulink model were used. A test input set is introduced to the trained ANN. Results are discussed and demonstrated graphically. It was observed that obtained results are satisfactory, and the error of ANN for a reference position is acceptable.

Keywords

• 6-DOF Manipulator
• Kinematic Analysis
• Inverse Kinematic
• Artificial Neural Networks