Design, Analysis, Simulation, and Virtual Reality Verified Intelligent Controller for Industrial Application SCARA Robot

Abstract

Robot manipulators are designed to execute required movements. Their controller design is equally important. Selective Compliant Assembly Robot Arm (SCARA) has four degrees of freedom (DOFs), with three (shoulder, elbow, wrist) controlled by servo motors and one by pneumatics. Presented here is the development of a complete mathematical model of an industrial-application SCARA robot including the servomotor dynamics and simulation of the dynamics, also the analytical inverse kinematic problem (IKP) and the forward kinematic solution by D-H parameters. The DC servomotor driving each of the robot-arm joint is studied and modeled. The robot arm is built for trajectories of drilling, manufacture, assembly, etc. It is realized by a 3D virtual reality (VR) model, which builds and receives commands through a MATLAB/Simulink link for the design‟s simulation on MATLAB Version R2012a. The control scheme is Adaptive Neuro Fuzzy Inference Strategy (ANFIS). A neural network with fuzzy logic controller (FLC) selects the proper rule base through back propagation algorithm. The integrated approach improves system performance, cost-effectiveness, efficiency, dynamism, and controller reliability. The method is effective, and the response (settling) is fast. The SCARA here is VR-verified. Real-time application is possible through interface cards.