Karel Capek
Almost 90 years ago, the word ‘robot’ was first used by science fiction writer Karel Capek in his 1922 play entitled ‘Rossum’s Universal Robots’. 20 years later, in the early 1940s, ‘robotics’ was coined by another writer, Isaac Asimov, who used it to describe the activities and disciplines that we in the modern times now associate with the word. Indeed, robotics has evolved much from the days where it was an imaginary field existing only in the pages of books, and in the imaginations of people. In this day and age, robots can be found everywhere – factories, hospitals, warehouses, laboratories, and even domestic households.
What is a robot? Over the last couple of decades, we have seen the rise of mechanical systems as aids in our day-to-day tasks. Such systems allow the movement of a rigid body (or an ‘end-effector’) with respect to a fixed base. There are various ways in which this body can move in space, whether the movement be translational or rotational, and the number of ways in which it can move is called its ‘degrees of freedom. ’ When an end-effector’s position is completely controlled by a mechanical system, then this system can now be called a robot.
Among the most useful of robots are those that we classify as industrial robots. These are primarily developed for use in manufacturing and assembly, and are used to perform tasks that humans shy away from and avoid. Such tasks include disposing of toxic materials and manipulating things in hazardous environments, which are potentially unsafe for people, but not so for robots. They also include tasks that are time-consuming and repetitive, requiring a lot of patience, but very little brain power, like moving materials from one point to another.
In general, these tasks form the so-called ‘4D tasks’, meaning dangerous, dull, dirty, or dumb. The first industrial robot was installed in 1961 by General Motors, for use in die casting, and today, industrial robots come in a wide variety of sizes, shapes, and design configurations. Given the broad spectrum that encompasses industrial robots, it thus becomes difficult to come up with a precise definition of what exactly they are. The International Organization of Standardization (ISO) broadly defines the industrial robot as ‘an automatically controlled, reprogrammable, multipurpose manipulator in three or more axes.
’ Differences in structure, size, degrees of freedom, and other such characteristics enable us to further classify industrial robots into sub-categories. Serial Robots The Springer Handbook of Robotics defines a serial robot, or serial chain robot, as ‘a sequence of links and joints that begins at a base and ends with an end-effector. ’ It is made up of rigid bodies that are connected by one-degree-of-freedom joints, and configuring these links and joints allows for different types of movements.
The first three joints are usually used to position the arm at a given point in space, while the succeeding joints are used to orient the arm around this point. The symmetry of a robot’s reachable workspace often serves as the basis for the design of that particular robot, and workspaces can generally be classified into three basic shapes: rectangular, cylindrical, and spherical. Cartesian Robots Robot arms operating in rectangular workspaces are called Cartesian Robots, and these use three prismatic joints to allow the movement of an arm along the three axes of the Cartesian plane.
A spherical joint attached to the arm acts as a wrist, allowing rotation about the arm’s central axis. Cartesian robots often come as gantries, which are characterised by their principal axes being supported by a framed structure. Such robots are modular, allowing for different configurations of their axes depending on the requirements of the task at hand. This makes them ideal for use in warehouses and transport bays, or for special tasks wherein large areas must be covered by the robots. While other kinds of industrial robots allow rotation about a fixed rotary axis, Cartesian robots do not.
This allows them to have relatively greater accuracies and consistencies, especially when they carry large loads. Cylindrical Robots Robot arms operating in cylindrical workspaces are what we call cylindrical robots. These workspaces can be formed using two ways: either by two prismatic joints combined with a rotary base joint, or by two rotary joints combined with a prismatic joint. Today, cylindrical robots take on a minor role in industry. Like Cartesian robots, they can also be used for loading and unloading machines. SCARA Robots
The SCARA, or Selective Compliance Assembly Robot Arm, is a special type of cylindrical robot that uses two parallel rotary joints to provide movement along a specific plane. It was developed by a team led by Hiroshi Makino, a professor at Yamanashi University in Japan, in the early 1980s. At that time, efforts were focused on developing complex, general purpose robots that could perform a variety of functions. However, Makino’s team realised that most assembly operations were actually made up of a finite number of simple motions, prompting them to deviate from convention by building a robot with more limited functions.
Due to the fact that the SCARA is kinematically simpler than other robot arms, and thus lower in cost, it allows for higher accuracy and precision, as well as operation at greater speeds. Polar Robots Polar robots are robot arms that use the polar coordinate system to operate in a spherical workspace. With the exception of their workspaces, they are very similar to cylindrical robots in both structure and function: they also use two rotary joints and a prismatic joint, and they are also used in loading and unloading operations. Jointed Robots
Unlike the previous classifications, jointed robots are classified not because of the workspace that they operate in, but rather, because of their structure. A jointed-arm robot, also known as an anthropomorphic robot, mimics the movements of a human arm. It uses rotary joints to connect rigid bars, analogous to how a shoulder, elbow, and wrist connect the torso (base) to the upper arm, forearm, and hand. Because of its configuration, it is ideal for use in tasks that would normally require the use of a human hand, such as the manipulation and positioning of small objects.
Parallel Robots Unlike serial robots, which use a single chain of arms and joints to support an end-effector, parallel robots can have two or more such chains working together to support such an effector. These will be discussed in detail in the succeeding sections. Table 1 Summary of typical arm and wrist configurations for various types of industrial robots. Types of Industrial Robots Axes Wrist Kinematic Chain Workspace DOF Examples Serial Cartesian Rectangular 1-3 Cylindrical Cylindrical 1-3 Spherical Spherical 1-3
SCARA Limited cylindrical 1-2 Jointed Limited spherical 2-3 Parallel Parallel 3-D space 3-6 — Parallel Robots Jean-Pierre Merlet defines a generalized parallel manipulator as ‘a closed-loop kinematic chain mechanism whose end-effector is linked to the base by several independent kinematic chains. ’ The connection degree of a link refers to the number of bodies that is attached to this link, and simple kinematic chains are chains where each body possesses a connection degree that is not greater than two.
A closed-loop kinematic chain exists when there is at least one closed loop that is formed by the links and joints of the chain, and this happens when one of the bodies has a connection degree of three or greater. Now, generalised parallel manipulators are not the same as parallel robots. The former encompasses a broad spectrum of parallel manipulators, including those with redundant mechanisms. ‘Parallel robots’ are thus defined as ‘made up of an end-effector with n degrees of freedom, and a fixed base, linked together by at least two independent kinematic chains.
Actuation takes place through n simple actuators. ’ Such robots are especially interesting for a number of reasons. First, with a minimum of two chains, we can distribute the load on the chains. Next, there is no redundancy, and the number of actuators needed, as well as the number of sensors used to control actuation, is kept to a minimum. Finally, when the actuators are locked, the mobility of the system is zero, which makes it safe and ideal for use in a wide variety of applications.
In general, we get parallel robots when we combine the ‘automatically controlled, reprogrammable, multipurpose’ characteristics of robots with parallel mechanisms. Now, the concept of parallel mechanisms is actually even older than the notion of a robot. James E. Gwinnett designed and patented the very first parallel mechanism way back in 1928, to be used as a platform in a movie theater. However, the basic principles of such a mechanism, a mechanism with a closed-loop kinematic structure, were only established in 1947, by a man named Eric Gough.
He used these principles to design a moving platform to test tire wear and tear. Ball-and-socket joints were used to connect links to a hexagonal platform, with the other ends of the links attached to a base, and linear actuators allowed for the modification of link lengths. The Gough platform is one of the most popular parallel platforms ever built, although it was finally retired in 2000. Subsequent platform designs have been produced by others such as Stewart, Cappel, and Clavel. To interconnect the various components of a parallel mechanism, a wide variety of kinematic joints can be used.
Simply speaking, these joints are formed when two bodies come into contact with each other, and their connection ends up constraining the movements of the bodies relative to each other. There are six types of simple kinematic joints: revolute (R) or rotary joints, prismatic (P) or sliding joints, helical (H) or screw joints, cylindrical (C) joints, which are like R joints but they also allow sliding along the axis of rotation, spherical (S) or ball-and-socket joints, and planar joints. Sometimes, it may also be necessary to make use of more complex joints.
These can be higher-pair joints, wherein two bodies make contact only along points and lines (as opposed to simple joints, where contact occurs over surfaces), compound joints, which consist of chains of simple joints and other members, like the universal (U) joints, and the 6-DOF joint, which is a theoretical concept that is used to model two bodies which have no contact with each other. Kinematic joints can be either actuated or passive. Actuated joints are actively controlled by motors, or actuators, while passive joints are free to move according to the forces acting upon it.
Structure and Performance of Parallel Robots Parallel robots can be configured in a myriad of ways. In fact, because there are so many combinations of joint types and movements, kinematic chains, and actuators, it is virtually impossible to categorise all the different structures of parallel robots. Traditionally, parallel robots had six degrees of freedom. However, in the recent years, an increasing number of designs for new structures of parallel robots, especially those with less than six degrees of freedom, have emerged.
The reason behind this is simple practicality: many applications don’t need movement in all six degrees. Still, having less than six degrees of freedom (and thus less actuators) does not necessarily translate to a cheaper design. The need for different parts and actuators may increase manufacturing costs, and unwanted movements produced by the design irregularity may lead to poorer performance. Thus, in evaluating the structures of parallel robots, it is important to consider basic issues like kinematics and workspaces.
Merlet proposes a general system of classification based on the type of movements that can be made. Using this, two main groups emerge: planar robots and spatial robots. Planar Robots Planar robots have their movement restricted to a single plane. A fully parallel planar robot has an end-effector with three degrees of freedom – translation along the x- and y-axes, and rotation about the z-axis. It is possible to have planar robots with only two degrees of freedom; however, these are used for very specific tasks only, and their use is not as widespread.
Planar robots with three degrees of freedom consist of three actuators and three independent kinematic chains. One end of each chain is attached to a base (the ground), while the other end is attached to a moving platform. By using either P joints or R joints, it is possible to come up with eight different configurations for the kinematic chains (RRR, RRP, RPR, PPR, PRP, PPP). Note that the PPP joint configuration, however, should be excluded, as it does not allow for independent joint motion, reducing our number of configurations to seven (Figure 1).
Robots which have different configurations per kinematic chain are also possible. Figure 1 The different configurations for a parallel planar robot with three degrees of freedom and identical kinematic chains. Extensive studies have been done on fully parallel planar robots, especially in relation to dimensional synthesis, singularity analysis, workspace analysis, stiffness, and balancing. Applications of these robots have been used for educational purposes, which is what the company Googol Technology is doing, as well as for high speed robot functions, as proposed by Hunt.
Other studies done on planar robots include the possibility of replacing rigid links with wires, exploring different arrangements of actuators, adding redundancy, and building micro-robots with flexible joints. Spatial Robots Unlike planar robots, spatial robots can move in all of 3-dimensional space. However, as mentioned earlier, not all parallel spatial robots make use of the full six degrees of freedom. Different combinations of revolute, prismatic, universal, and spherical joints determine the type of motions that a moving platform can make, and spatial robots can have anywhere from three to six degrees of freedom.
Sample Essay of Eduzaurus.com