09-10-2014, 11:54 AM
Abstracts: Mankind has always strived to give life-like qualities to its artifacts is an attempt to find substitutes for himself to carry out his orders and also to work in a hostile environment. The popular concept of a robot is of a machine that looks and works like a human being. This humanoid concept has been inspired by science fiction stories and films in the twentieth century. The industrial robots have been used in carrying out various task automatically with desired accuracy and precision and also repeatedly. Robotic manipulators nowadays are very widely used in industry for carrying out various tasks like pick and place type activities, welding of various parts, carrying out assembly, spray painting, inspection etc. During the working cycle of manipulator, it is required that end-effector should move through set of specific points in space to follow required trajectory. For robotic manipulator to perform its desired task accurately, it is required to control actuators of the joints of the manipulator effectively so that end-effectors move through desired path in the reachable work space of manipulator. Control of robotic manipulator is very difficult task. Difficulty arises from the complexity of dynamic model of the manipulator arm. One of the methods is to use computed torque control scheme which is an approach that makes direct use of complete dynamic model of the manipulator to cancel out various effects such as gravity, Coriolis and centrifugal force, friction, and manipulator inertia tensor which includes non-linearity in control problem. Aim of the project is to develop complete dynamic model of 3-DOF manipulator and use this dynamic model in designing control parameters so that desired task can be accomplished by the 2-DOF planar manipulator. For designing control circuit for manipulator MATLAB R2013a software by MathWorks is used. Simulink feature of the MATLAB is used for development of complete control scheme which includes plant (manipulator), circuits for inertia, centrifugal and gravity components, feedback elements and PD control blocks. Desired input is then given in joint space to be followed by the manipulator. Variation in desired input and actual output is presented which provides information about accuracy of the control system.