Ramakant Rana UNIT 1

Asst. Prof., MAE Deptt., MAIT Robotics

INDUSTRIAL AUTOMATION AND COMPUTERS
The term ‘Industrial automation’ is defined as the technology concerned with control of systems
in the process of achieving an end product. The process of achieving the end product has to be
with minimum or no human intervention. In these notes, the term ‘automation’ is used in place
of ‘industrial automation’ unless otherwise required. Automation fi rst became popular in
automobile industries. Automation helps the industrial manufacturers in achieving high
productivity, high level of accuracy, consistent quality and increased labour saving. Automation
is based on foundations of automatic control and obviously on feedback theory. It is applicable
to any branch of science, engineering and technology wherever and whenever a feedback is
required.

The use of feedback in order to control a system has a fascinating history. The first application of
feedback control is believed to be a float regulator for level control in Greece during the period
starting somewhere between 300 and 1 BC. This control mechanism is used in maintaining water
level in the water clocks and oil level in oil lamps. Pressure regulators for steam boilers (similar
to pressure cooker valve) were in use from 17 AD. James Watt’s fly ball governor, invented in
1769, is still in use in many industrial speed control systems.

Intuitive inventions were contributing to the development of automatic control and hence
automation till 1868 when Maxwell formulated the mathematical model (in terms of differential
equations) to describe a system. He demonstrated the effect of parameters on system
performance. The concepts of accuracy and stability were understood. The mathematical models
in different forms such as transfer function, pulse transfer function, describing function and state
variable modelling were considered as inevitable tools for analysing and designing of control
systems. Since Maxwell’s formulation of modelling, the automatic control and hence automation
have been accepted to be a science rather than an art.

World War II (during 1940s) was a milestone in the field of automation. In this period, the whole
area of scattered investigations of automation consolidated into a discipline. The compelling
necessity of designing military equipment, prototype production and testing and mass scale
manufacturing have contributed to a tremendous growth in the fi eld of automation. A push-
through into new insights and methods was achieved. Many automatic control theories were
formulated. The pioneering contributions of H. Nyquist, H. W. Bode, V. M. Popov and many
others have strengthened the discipline of automatic control and are still being used in
developing new technologies of designing, controlling and perfecting automation systems. The
idea of using computers in automation emerged during 1950s. Analogue computers were used
as on-line controllers in continuous processes such as steel and paper industries. Almost during
the same time, the power of rapid decision-making properties of digital computers was found
feasible in automation. A dramatic first application of digital computer control was established
when a polymerization unit went on-line. This pioneering work in computer control was noticed
by industrialists, manufacturers and research experts; the industrialists realized a potential tool
for increased automation, the manufacturers recognized a new market and the research experts

Page 1 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
saw diversified research fields. This has resulted in a tremendous growth in modern automation
methods. Automation without a computer is now hard to imagine. The cost of computers used
as controllers was the major concern in implementing automation. The cost of analogue
controllers increased linearly with increased control loops. On the other hand, even though initial
cost of digital computer was large, the cost of adding additional loops was small. Additional
advantages such as digital instrumentation panels, implementation of complex control
strategies, easy human interface, reduced human interaction and high reliability were the forcing
functions of increased use of digital computers in automation. The first microprocessor
developed by Intel Corporation, USA, during 1971, has initiated the microcomputer age. Since
then, the research efforts in microelectronics and development of computer software have been
offering powerful computers for automation at reduced cost. Since the necessities of computer
control are unlimited, one can safely guess that there will be a sustained growth in industrial
automation during many years to come.

INDUSTRIAL ROBOT
Automation and industrial robots are two closely related technologies. According to definition of
automation, an industrial robot can be considered itself as a form of automation. The term
‘robots’ will used in place of ‘industrial robots’ in the notes. A robot is a general-purpose
programmable machine which possesses the characteristics of a human arm. The robot can be
programmed by its computer to move its arm through sequences of motion in order to perform
some useful tasks. It repeats the motions over and over until it is reprogrammed to perform some
other task. Many industrial operations involve robots working together with other equipment.

Sir Isaac Asimov Dr George Devol

The word ‘robot’ was introduced by Karel Capek a Czechoslovakian play writer in 1921. Robot
was derived from the Czech word ‘robota’ which means forced worker. Early writings and plays
during 1920s and 1930s pictured the robot as ferocious humanoid machine which was intended
for killing human beings. Sir Isaac Asimov1 (Figure 1) through his prodigious imagination
contributed a number of stories about robots starting from 1939. He coined the word ‘robotics’

1
Isaac Asimov (2 January, 1920 – 6 April, 1992) was an American writer and professor of biochemistry at Boston
University. He was known for his works of science fiction and popular science. Asimov was a prolific writer who wrote
or edited more than 500 books and an estimated 90,000 letters and postcards

Page 2 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
which means the science of robots. Later, Sir Isaac Asimov and Dr George Devol 2 started a
robotics industry named as Unimation Inc. They started developing fully hydraulic powered
UNIMATE robot. In later years, Ford Motors used UNIMATE successfully for die casting.

The Laws of Robotics

In 1942, Sir Isaac Asimov developed the famous three laws (Law one, Law two and Law three) of
robotics which still remain as worthy industrial design standard. However, the laws have been
extended and revised by him and others since 1985 to accommodate his creations, his attitude
to robotics and the modern requirements of humanity. The extended set of laws is as follows:

The Meta-Law
A robot may not act unless its actions are subject to the laws of robotics.

Law Zero
A robot may not injure humanity, or through inaction, allow a humanity to come to harm
(humanity is the family of all human beings and other biologically living things).

Law One
A robot may not injure a human being, or through inaction, allow a human being to come to
harm, unless this would violate a higher order (an earlier stated) law.
Law Two
A robot must obey orders given by human being, except where such orders would conflict with
a higher order law. A robot must obey orders given by subordinate robots, except where such
orders would conflict with a higher order law.
Law Three
A robot must protect the existence of a subordinate robot as long as such protection does not
conflict with a higher order law. A robot must protect its own existence as long as such
protection does not conflict with a higher order law.
Law Four
A robot must perform the duties for which it has been programmed, except where that would
conflict with a higher order law.

The Procreation Law

The robot may not take any part in the design or manufacture of a robot unless the new robot’s
actions are subject to the laws of robotics. The robots which are strictly manufactured in
accordance with the above rules do behave better than human beings.

2
George Charles Devol Jr. (20 February, 1912 – 11 August, 2011) was an American inventor, best known for creating
Unimate, the first industrial robot. Devol's invention earned him the title "Grandfather of Robotics". As recognized
by the National Inventors Hall of Fame, "Devol's patent for the first digitally operated programmable robotic arm
represents the foundation of the modern robotics industry."

Page 3 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics

When the concept of robot was introduced and strengthened, the necessity of industrial
automation was also deeply felt. Moreover, the technological progress in thermionic valve
(1904), hydraulic and pneumatic systems (1906), logic circuits (1943), digital computer (1946),
transistor (1947), microelectronics (1970) and microcomputer (1977) have all made automation
and robotics a reality. The first commercial robot, controlled by limit switches and cams, was
introduced in 1959. Since then, the development in robot technology has been in constant
growth. Nowadays, the service robot within industry and in other areas of applications has made
a breakthrough in robot applications.

Main Features of Robot

• A robot does not become fatigued or distracted
• It is capable of producing a job with consistent quality at a steady rate with practically
zero rework and wastage
• Continuous working throughout the year except during repairs and maintenance
• It can take up repeated and tiresome jobs even in unsafe and unhealthy environment
• It does not demand wage increase, fringe benefits, pension, holiday, sick pay, etc.
• Capital cost of robot is paid only once
• Robot’s upkeeping cost is increasing at a slower rate compared with labour maintaining
cost every year
• Robot can lift heavier weights, be more precise, work at higher speeds and can exert
larger force than is humanly possible

Robot Applications in Manufacturing

Robot applications are of wide variety in the manufacturing side. These include a large variety
of production operations. Here the operations are classified into the following categories.

(i) Materials transfer and machine loading and un-loading

These are applications in which the robot grasps and moves a work part from one location
to another. This category includes applications of robots as they transfer parts into and
out of a production machine. These include metal machining operations, die casting,
plastic moulding and certain forging operations.
(ii) Processing operations
Here the robot uses a tool as an end effector. This tool will accomplish some processing
operations on a work part positioned for the robot during the work cycle. Spot welding,
arc welding, spray coating and some machining operations fall in this category.
(iii) Assembly and inspection operations
A set of new applications such as robot-based assembly and inspections is the final set of
robot applications. The robot is used to put components together into an assembly and/or
to perform some automated inspection operations. More robots are equipped with vision
capability facilitating the performance of these operations. The applications of robot
cameras are to inspect welding, brazing and bonding with glues and other fasteners.
Page 4 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics

DEFINITIONS OF INDUSTRIAL ROBOT

Any automatic machine cannot be considered as a robot. Robot is to have a specific set of
characteristics. Interestingly, a 3-axis computer numerical control (CNC) milling machine may
have a very similar configuration and control system of a robot arm. However, the CNC machine
is just a machine. It cannot do jobs other than milling. But the robot must do something more.
That is why the definitions are proposed for a machine to be a robot.

Different countries have different definitions for a robot. The Robot Institute of America (RIA,
1985) defines the robot as
A robot is a reprogrammable multi-function manipulator designed to move materials,
parts, or specialized devices through variable programmable motions for the
performance of a variety of tasks.
This definition restricts robots in industrial applications. The two important key words are
‘reprogrammable’ and ‘multi-functional’. If the machine is single functional, it cannot be
reprogrammable. Reprogrammable means that
(i) The robot motion is controlled by a written program and
(ii) The program can be modified to change significantly the robot motion.

Multi-functional implies that a robot is able to perform many different tasks depending on the
program in the memory and tooling at the end of arm. This means that the robot can be
programmed for welding with a welding tool at the end of arm and can be reprogrammed if the
end of arm has a totally new facility such as for gripping. Another, a little broader definition is
proposed by McKerrow (1986) as:

A robot is a machine which can be programmed to do a variety of tasks in the same way
that a computer is an electronic circuit which can be programmed to do a variety of tasks.

The International Standards Organization (ISO 8373) defines a robot in a similar way as follows:
A robot is an automatically controlled, reprogrammable, multi-purpose, manipulative
machine with several reprogrammable axes, which may be either fixed in place or mobile
for use in industrial automation applications.

CONFIGURATION AND WORK VOLUME

We can compare the configuration of a human body, its work volume and its other features with
those of an industrial robot. They are similar in many aspects. Human is natural and the robot is
man-made to work for his/her objectives. The objectives can be faster, higher accuracy and/or
more powerful compared with those of human being. Both configurations have their own
advantages and disadvantages. In the following descriptions, we compare the configuration of a
human body with that of an industrial robot.

Page 5 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
Configuration of Human Body
A human body with arms (especially one arm) mimics a stationary industrial robot. Every human
being has a similar structure but with varied sizes and power. He (she inclusive) can stand firmly
on the ground. He has a waist, body, shoulder, forearm, elbow, lower arm, wrist and fingers and
each one of these can have its own articulations. Articulations are rotations about or translations
along several axes of the joints. Typically, the waist (a joint between body and legs) can rotate
about three axes, the shoulder (a joint between body and forearm) about three axes, the elbow
(a joint between forearm and lower arm) about two axes and the wrist (a joint between lower
arm and the fingers) about two axes.

Each joint connects two links (limbs). Waist connects the leg and the body. The shoulder connects
the body and the forearm, the elbow connects forearm and lower arm, and the wrist connects
the lower arm and the fingers. These joint rotations and link lengths create a work space (work
volume) around the human. The work volume is a set of all points which the arm can reach. The
fingers can give several independent rotations in each of its joints. The human has a number of
in-built sensing abilities for touch, pressure, olfaction, vision, hearing, heat, force, torque and
compliance with which the human body does almost unlimited tasks. The compliance is an ability
to perform little changes in articulations when unexpected structural disturbance is experienced.
A small momentary thrust on the human body shakes the body a little without modifying the
rotational angles. Above all, the brain controls all such articulations based on the sensory
outputs. A sensor creates a level of intelligence. Existence of countless sensors with an analytic
brain exclusively differentiates a human as highly intelligent than the machines, especially the
robots.

The human, with his vast sensing abilities, can be faster and accurate than a robot in performing
several tasks. He cannot be consistent in speed and accuracy through a longer period of time due
to distractions and tiredness. However, a robot can maintain consistency over a long period of
time.

Page 6 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics

Robot Anatomy

Industrial Robot Configurations

An industrial robot is constructed for a specific purpose. It has to be economical and be able to
do varieties of industrial jobs. Hence, it has to be of varied structures so as to suit specific
requirements. For example, for spray painting, a robot has to carry the weight of spray painting
gun. The robot for this job can be of light weight. However, a robot which is intended to stack
cement bags in a lorry has to be powerful, strong in structure, with long arms and less
articulations. Hence, the configurations of these robots have to be different. The configurations
of industrial robots are many to suit several job requirements. To indicate a few, the job
requirements can be of picking and placing an object in a machine tool, spray painting a finished
product, welding two metal plates, assembling a number of pieces together and palletizing
(stacking the products within a specified space).

Figure 3 illustrates one of the configurations of an industrial robot manufactured by Cincinnati

Milacron, USA. This combines a variety of several robots. This is called T3 (twist power 3) robot,
that is, a twist exerted at the base and another two twists exerted at the body. This has a base
firmly attached to the ground or a platform. The robot articulation is possible due to rotations of
several joints such as body rotation (about a vertical axis), shoulder swivel (about an horizontal
axis), elbow rotation (about another horizontal axis) and wrist rotations named as yaw, pitch and
roll in three different axes. An end effector (similar to our palm and fingers, not shown in the
figure) can be mounted to the tool mounting plate. The centre point of the end effector can have
its movements due to wrist rotations. The end effector can have a gripper to grip and hold
objects. Several varieties of grippers are available for various applications. A gripper can have
two or more fingers. The end effector can be a tool to perform a process on a product. Drilling

Page 7 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
certain locations is a process. A welding torch is a tool fitted to the tool mounting plate to perform
welding processes.

Figure 3

The robot has its brain in terms of a robot computer. The computer has the set of software
required for the functioning of the robot. The computer has facilities to interface with number of
sensor systems such as a vision camera that can be attached to the robot or placed elsewhere
near the robot. The computer can have facilities to interface with other industrial equipment
typically a conveyor system which brings the product to the robot so that the robot can perform
handling or processing operations.

The robot is also connected to a power supply unit that gives the required power – electrical or
hydraulic or pneumatic or combination of these. Electric supply is required to operate the
internal motors coupled to various joints. Hydraulic and pneumatic powers are useful in
operating cylinder–piston mechanisms built in the robot structure. There is a provision to teach
a robot on how to perform a task. This can be done in various ways. The usual method is through
a teach pendant.

Page 8 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
All these components of a robotic system described above are also illustrated in Figure 4.

Figure 4
A robotic system basically consists of robot mechanical structure, a robot control computer and
power sources. The robot structure has all necessary electro-mechanical, pneumatic and
hydraulic elements such as electrical actuators (motors for rotary operations) and to control the
speed and position of actuators and interfacing for teaching and learning a specified task and for
robot safety to avoid any possible damage to robot structure. In addition, the computer has its
own operating system for operating its internal electronics, peripheral devices such as key board
and display, disk drives, mass storage and associated equipment.

Figure 5 shows another variety of robot, FANUC S-420i, six axes and electrically powered. This
has a tool mounting plate to which an end effector has to be fixed. Generally, robots are sold
without the end effector and the user has to design or buy the end effector suitable for his
applications. This S-420i family consists of five members. This family of robot is normally used in
automotive industry for spot welding, body assembly and material handling. Table 2.1 lists the
specifications of three of the S-420i family. Each axis has its own range of rotation and speed.
Maximum reach is the farthest point which can be reached by the centre of the tool mounting
plate (or by the centre of the end effector, if end effector is mounted). Centre of the end effector
is generally known as tool centre point irrespective of gripper or tool at the tool mounting plate.
The maximum reach is a function of robot’s configuration and its link lengths including obstacles
on its articulations. Load capacity (or pay load) is the maximum load (kg) that a robot can lift,
hold and position by the end effector repeatedly without affecting other specifications. The S-
420i family does have automatic braking of all axes when power supply fails.

Page 9 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics

Figure 5

Structural Configuration
Industrial robots have many structural configurations that suit several applications in industry.
Figure 6 describes some of the configurations of robots that can have varieties of joints.
Configurations generally follow coordinate frames (Cartesian, cylindrical and spherical) with
which the robots are referred to. Figure 6 describes combinations of linear (prismatic or
translational) and rotational joints. Figure 7 describes some of the usually adopted industrial
robots. A robot has varieties of joints so that its end effector can be manipulated for any specific
task. Figure 8 describes four types of joints. The linear (L) joint gives a link the translational
motion. The rotational joint ® makes a link move about an axis perpendicular to the previous link.
Twist joint (T) rotates a link by itself while it is aligned with the earlier link. However, in the
revolute joint (V), the link has an angular position with respect to the earlier link and rotates at
an axis along the earlier link. With this definition of joints, the robots of Figure 6 may be
designated as having its configuration in terms as L, T, V and R. A Cartesian coordinate robot is
termed as LLL, cylindrical as TLL, spherical as TVL, articulated as TRR and the SCARA as TRL. SCARA
is a special variety, meaning ‘selective compliance assembly robot arm’. SCARA is especially
useful in electronic component assembly for picking and inserting Ics in their respective sockets
much faster and with greater accuracy and repeatability compared with other types of robots.
Its specific characteristic is that it is more compliant in the horizontal plane but is stiff in the
vertical plane and thus has its selective compliance.

Page 10 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics

Figure 6: Configurations of Industrial Robots

Figure 8: Types of Joints: (a) Linear or Translating (L), (b) Rotating (R), (c) Twisting (T) and (d)
Revolving (V)

Page 11 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics

Figure 7: Commonly Known Industrial Robots

Page 12 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics

Robot Work Volume

Work volume (work envelope) is the space in which the robot can manipulate its wrist end. This
convention of wrist end of robot to defi ne the work volume is considered better than adding
different sizes of robot end effectors attached to their individual wrists. The general objective for
the robot manufacturers is of getting a spherical work volume; however, the work volume cannot
be an ideal sphere due to the characteristics of robot such as physical configuration, size of the
body, length of links and mechanical limits on joint movements. All manufacturers give detailed
specification of work volume of their robots in their pamphlets.

Figure 9: Configurations and Work Volumes

Figure 9 illustrates varieties of robot structures and their respective work volumes. Cartesian
coordinate robot develops a full rectangular form of work volume, whereas cylindrical robot can
produce only a hollow cylindrical work volume due its inherent structural limitations. Polar

Page 13 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
coordinate robot and jointed arm horizontal axes robot present nearly spherical work volumes.
Jointed arm vertical axis is the SCARA robot which offers a near cylindrical form of work volume.
The gantry pendulum arm’s work volume is a perfect section of a hollow sphere. The multiple
jointed arm robot is named as spine robot manufactured by Komatsu Ltd, Japan. This provides
an excellent flexibility in reaching a location in its work volume. It is specially designed for spray
painting, spot welding and seam (or continuous) welding. Its arm can reach intrinsic locations
where other conventional robots cannot reach; hence, they are very suitable for car spot welding,
painting of irregular shaped objects and for performing inspections in nuclear reactors. It is
interesting to note that it produces near perfect spherical work volume.

PRECISION OF MOVEMENT
Robot’s capability to reach a point within its work volume depends on the precision of the robot.
The precision of a robot is a function of three features:
(i) Spatial resolution
(ii) Accuracy
(iii) Repeatability.

In order to describe the precision of a robot, it is necessary that the robot is at rest and works
only when commanded. In addition, the robot is assumed without end effector. The precision is
not usually defined with an end effector since the end effector can have different sizes and
weights. In order to defi ne precision, we assume a robot of having one prismatic joint and we
later formulate how the error is magnified by all other joints to affect the precision.

Spatial Resolution
Spatial resolution of a robot is the smallest increment of movement in which the robot divides
its entire work volume. The smallest increment in movement depends on two factors – the
robot’s mechanical inaccuracies and the system control resolution.

Mechanical inaccuracies mainly come from robot joints. The joints generally have gears and
chains to move the links. The gear and chain drives have their nonlinearities such as back clash
and hysteresis which create errors in the movement of robot link. In the case of hydraulic and
pneumatic drives at the joints, additional inaccuracies are created due to fluid leakage and
compressibility. These inaccuracies are added and magnified if the robot has several joints, thus
making the measurement of spatial resolution uncertain.

The control resolution is determined by the robot’s position control system and its position
measuring devices. Position measuring devices are usually potentiometers and shaft encoders.
Inaccuracies in the measurement by either of these components affect the spatial resolution. The
robot control system creates word data for each of the locations in the work volume. Locations
are known as addressable points. It is always advantageous when the work volume is divided into
numerous addressable points in order that the robot can perform minute manipulations. The

Page 14 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
work volume is created by the range of joint motions (rotational or translational). Reaching any
addressable point within the work volume by the end effector depends on the number of
increments within a restricted joint range. These increments are created by robot computer. For
example, if the robot has 8-bit word storage, it can create 256 discrete increments within a joint
range.

EXAMPLE 1
Consider a robot having a sliding joint which can extend the link maximum to 1 m. The robot
control memory has 12-bit word storage. What is the control resolution of the axis motion?
Solution:
Maximum number of increments within the joint range is 212 = 4096.
The minimum increment in sliding joint motion is 1 m/4096 = 0.24414 mm.
Therefore, the control resolution is 0.244 mm (to three digits accuracy).

In Example 1, if the robot is required to reach 2.5 mm, a compromise has to be made in the
control software on whether to go for 10 increments or 11 increments. A robot has several joints,
some of which are rotational and others translational. Each joint has its own joint range and
hence its control resolution. Hence, the total control resolution of the entire robot cannot be
obtained by algebraic addition of all individual resolutions. This is a complex procedure in which
each resolution has to be considered as a vector in a three-dimensional space; the total resolution
is, then, the vector addition of all joint resolutions.

Accuracy
Accuracy refers to a robot’s ability to position its wrist end at a desired target point within its
work volume. The accuracy is related to spatial resolution because the accuracy depends on how
closely the neighboring points are defined within the work volume and on how accurately the
mechanical structure is produced. For example, consider the robot of Example 2.1. Figure 10
illustrates the definition of accuracy. A and B are the nearest points as determined by control
resolution. T is a point halfway between A and B.

Figure 10: Robot Accuracy, Spatial Resolution and Control Resolution

Page 15 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
If the mechanical structure is precise, then the robot can reach A or B exactly, but not at a point
in between them. Spatial resolution and the control resolution are the same if the mechanical
structure is precise. The robot accuracy is then defined as half of control resolution from either
side (±0.122 mm). However, the mechanical structure usually has its own inaccuracies. Due to
this, even when the robot is commanded to reach A or B, it will not reach them exactly, but will
reach any point around A or B.

In Figure 10, the mechanical inaccuracies are represented as points along the bold short lines
around A or B. They can be regarded as having the statistical normal distribution along the bold
lines. With the mechanical inaccuracies, the spatial resolution is defined as the distance between
two farthest points that a robot can reach when it is commanded to go to any point between A
and B. Then the accuracy is defined as one-half of the spatial resolution. This definition seems to
imply that the accuracies are the same everywhere within the work volume. In fact, the
accuracies mainly depend on two factors. One when the robot works at the extreme areas near
to the borders of work volume. Here the robot has to extend its arm. The other factor in defining
accuracy is the load being carried by the robot end effector when the load is attached to the wrist
end. Larger load with extended arm produces greater deflections of the mechanical links, thus
reducing the accuracy. The robot accuracy, when arm is fully extended, is worse than when the
arm is close to the centre of the work volume or close to the robot’s base. This is similar to the
accuracy with which we place a 20 kg weight at a location reachable by our extended arm
compared with a 20 kg weight placed near to our legs.

Repeatability
The repeatability and accuracy of a human in throwing darts in a dart game are illustrated in
Figure 11. A person is said to be accurate and repeatable if he is able to throw the darts every
time very close to and around the target point T (Figure 11(d)). If the darts always fall around and
close to a point that is not T, then the person has a capability of repeating with no accuracy
(Figure 11(c)). Figure 11(b) shows all the darts thrown close to and around T, showing accuracy
but not repeatable. If the darts fall everywhere on the board (Figure 11(a)), the person is
understood as having no repeatability and no accuracy. The same configurations of repeatability
and accuracy in dart game will also happen to a robot. Assume a robot holds a pin at its gripper
and is commanded to go to the target location on a board, kept horizontal, again and again and
prick. By studying the test (pricked) points, we can estimate the robots ability of repeatability and
accuracy.

In Figure 12, T is the location to which the robot is commanded to go. However, due to its errors
and errors in control resolution, the robot will not reach T but it reaches the point P. If
repeatability is the highest, the robot has to reach the point P whenever it is commanded to go
to T. Due to its inability to reach point P over and over again, it goes to any other point along the
bold short line around P. One such point is R which shows the repeatability error. The
repeatability is now defined as ±r where 2r is the length of the bold short line which carries all
the test points. Figure 12 shows one-dimensional aspects. It is noted that, during the

Page 16 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
experiments, the robot can have test points on a two-dimensional plane. In this case, r will be
the radius of a circle encompassing all the test points in the two-dimensional plane.

Figure 11: Accuracy and Repeatability: (a) Not Accurate Neither Repeatable, (b) Accurate but
Not Repeatable, (c) Not Accurate but Repeatable and (d) Accurate and Repeatable

Figure 12: Measure of Repeatability

Page 17 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics

Figure 13: Joints and DoF

Page 18 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
Repeatability

Robot Drive Systems

Types of Drive Systems

Page 19 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics

Page 20 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
THE TEXTUAL ROBOT LANGUAGES

GENERATIONS OF ROBOT PROGRAMMING LANGUAGES

Page 21 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics
First Generation Languages

Second Generation Languages

Page 22 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics

Future Generation Languages

Page 23 of 24 [Link]
Ramakant Rana UNIT 1
Asst. Prof., MAE Deptt., MAIT Robotics

Robot Language Elements and Functions

Page 24 of 24 [Link]