Kinematic, dynamic and workspace analysis of a novel 6-DOF parallel manipulator
Shah, Hrishi L.
MetadataShow full item record
A parallel manipulator is a closed loop kinematic chain mechanism that is connected to the base via multiple independent kinematic chains. Parallel manipulators have found numerous applications in industries ranging from airplane simulators to high-speed pick and place robots due to their higher load-carrying capacities and improved stiffness as compared to their serial counterparts. These advantages arise due to the presence of kinematic-closed loops within the manipulator that allow for the load to be transmitted to the ground via multiple chains. However, these kinematic closed loops also cause the workspace of parallel manipulators to be severely limited and also pose a major challenge to their analysis and control. Another problem is the evaluation of the workspace that is simultaneous affected by so many geometric parameters of the parallel manipulator. Parallel platform manipulators consist of a central platform attached to the base via multiple legs. One of the most popular parallel manipulators is the 6-DOF Stewart-Gough platform. It has 6-UPS architecture and has been the focus of much work till date. However, the main problem is the power consumption of the manipulator because of the bulky prismatic actuators lying along the links. Hence, we explore a novel design of a parallel manipulator with 6PUS architecture that has the prismatic actuators attached to the base, thereby making the legs lighter and the device energy efficient on the whole. We will refer to this novel design as the "Hexapod" for the rest of this work. The goal of this work is to look into the following aspects of parallel manipulators with special reference to the Hexapod: (1) Workspace analysis; (2) Modeling, Equation of Motion (EOM) generation and simulation In this work, we dwell into the workspace analysis of parallel platform manipulators. Since the workspace of a parallel manipulator is quite complex, it takes a considerable amount of time to compute it. However, in order to optimize a parallel mechanism for workspace, an efficient and universal way of workspace analysis is necessary. We propose an improved method of calculating the constant orientation workspace of any parallel platform manipulator. Finally, we use a CAD package to speed up the process to make a case for CAD-enhanced workspace analysis and showcase its general application to any parallel platform manipulator. Another study explores a novel method for computing the equations of motion automatically by using a symbolic computation engine called Maple. We also explore how this method of symbolic equation generation, coupled with an easy to use interface called Maplesim; can be beneficial in augmenting the learning in robotics courses. The next task is to employ these EOM's to formulate control strategies and test them in simulations.