ENM4MR – Modern Robotics 1

1 Chapter 1 – Introduction to Robotics

1.1 Introduction
The advent of modern industrial robotics has been driven by the need to increase product quality and reduce
manufacturing costs. In the past, the quality of products depended on the skill of the craftsman. Consistent
product quality was difficult to achieve by manual methods. The production costs were high and the speed of
production was slow.

Robot manipulators become more economical than manual production methods when the production volume
is high and product variations are low. In recent times, robot manipulators have been introduced in
manufacturing systems to perform part handling sorting, spray painting, welding and assembly processes. They
can repeat manufacturing operations quickly, reliably and accurately. Robot manipulators are attractive to
manufacturers because they are flexible and can be reprogrammed to perform different operations. Robot
manipulators are therefore classified as flexible automation.

The Oxford Dictionary defines the word robot as "a machine resembling a human being and able to replicate
certain human movements and functions automatically." The industrial robots of today rarely resemble a human
being in appearance. An industrial robot is comprised of a robot manipulator and a digital control system. Robot
manipulators can perform repetitive tasks at speeds and accuracies that far exceed human capabilities.

Robotics is the art, knowledge and skill base that allows one to design, apply and use robots in human
endeavours. Robotic systems, in addition to robots, include devices and all other systems that are employed to
ensure the that the robot functions as intended. The design and manufacture of a robot is an interdisciplinary
task involving the knowledge bases of mechanical, electrical and electronic engineering, computer science,
cognitive sciences and a few others.
1.2 Advantages and Disadvantages of Robots
1.2.1 Advantages
In many circumstances, robots can increase productivity, safety and efficiency. Robots can also ensure high
quality and consistency of products. Concerning working environment and safety, robots are capable of working
in environments that are exposed to radiation, darkness, high and low temperatures, underwater, etc. There is
no requirement for life support or comfort concerning the robot. Robots also do not require air conditioning,
ventilation or noise protection. Robots do not experience fatigue and can continue performing tasks with high
precision although the tasks may be boring and monotonous. Sensory systems fitted to robots exceed the
sensory capabilities of humans. Robots can easily process tasks simultaneously.

1.2.2 Disadvantages
The implementation and adoption of robotic systems can create resentment and dissatisfaction from workers
when it results in job losses. Robots are incapable of understanding or responding to emergencies. The robot
can only respond appropriately if a certain type of emergency protocols has been programmed into it. In
circumstance when a robot is unable to respond appropriately to an emergency, this may result in the robot
injuring workers or damaging other equipment. Despite the advantages, robots lack cognitive abilities, lack
creativity and experience some limitations concerning their vision and sensory systems. The adoption of robotic
systems is commonly hindered by high investment costs and the cost of training the end-user. There is also a
need for skilled labour to program the robot. Moreover, servicing, maintenance and repairs are usually a task
that can only be done by the company that supplied the robotic system. The process of obtaining on-site
assistance poses challenges and risks of production downtime.

1.3 Mechanisms
Robot manipulators are created from a combination of links and joints, which form a mechanism. Links are the
rigid bodies that allow forces and torques to be transmitted by the mechanism. In more advanced robotic
applications, links can include flexible members. The connection between two links is called a joint.

A kinematic pair is a connection between two links or mechanical elements that imposes constraints on their
relative movement. A kinematic pair is called a lower pair if there is surface area contact between the two
elements, for example a screw and nut or a universal joint. When the two elements have point or line contact
while in motion then the pair is known as a higher pair. Examples include chain drives, gears and cam joints.

A kinematic chain is a collection of links and joints that from an assembly. If a kinematic chain forms a closed
loop it is called a closed loop chain. In closed loop chains there are at least two paths of connectivity between
one link and every other link. In chains where there is only one path of connectivity between links, it is called
an open loop chain. If a mechanism contains both open and closed loop chains, it is called a hybrid kinematic
chain.
1.4 Degrees of Freedom
Degrees of Freedom (DOF) are the number of independent variables that would have to be specified to locate
all parts of a mechanism in three-dimensional space.

A rigid body in space has six Degrees of Freedom. Three variables are required to describe the location of the
body, i.e. x, y and z. Three variables are required to describe the orientation of the rigid body, i.e. 𝜃𝜃𝑥𝑥 , 𝜃𝜃𝑦𝑦 , 𝜃𝜃𝑧𝑧 .
Figure 1-1 depicts the six degrees of freedom of a body in space.

𝑧𝑧

𝑦𝑦

𝑥𝑥

Figure 1-1: Depiction of the six degrees of freedom of a body in space

A body moving on a plane has three Degrees of Freedom. Two variables are required to describe the location of
the body, i.e. x and y. One variable, 𝜃𝜃𝑧𝑧 , is required to describe the orientation of the body. A system with seven
or more DOF is said to be a redundant system. Figure 1-2 illustrates the three DOFs associated with a body
moving on a plane.

𝑦𝑦

𝑥𝑥

Figure 1-2: Moving body on a plane with 3 Degrees of Freedom

1.4.1 Lower Pairs
There are six fundamental types of lower pairs as follows:
• A prismatic joint, P, allows the two paired elements to slide relative to each other along an axis defined
by the geometry of the joint. A prismatic joint imposes five constraints and allows 1-dof between the
paired elements. A prismatic joint is also called a sliding pair. Figure 1-3 illustrates a prismatic joint.
• A revolute joint, R, permits two paired elements to rotate with respect to each other about an axis that
is defined by the geometry of the joint as shown in Figure 1-4. A revolute joint imposes five constraints
and allows 1-dof between the paired elements. A revolute joint is also called a turning pair.
• A cylindrical joint, C, permits independent rotation and translation about an axis defined by the
geometry of the joint. It imposes four constraints and allows 2-dof between paired elements. Figure
1-5 depicts a cylindrical joint.
• A helical joint, H, permits the two paired elements to rotate relative to each other, coupled with a
translation that is dependent on the amount of rotation. The rotation and translation are about an axis
defined by the geometry of the joint. Because the translation is dependent on the rotation, or vice
versa, a helical joint allows 1-dof between paired elements and imposes five constraints. A helical joint
is also called a screw pair. An example of a helical joint is shown in Figure 1-6.
• A spherical joint, S, permits the two paired elements to rotate freely with respect to each other in all
possible orientations. The rotation is about the centre of the sphere, which defines the geometry of the
joint. A spherical joint imposes three constraints and allows 3-dof between paired elements. A spherical
joint is also called a spherical pair and is shown in Figure 1-7.
• A Planar joint, E, allows translation along 2 axes (the plane of contact) and rotation about 1 axis that is
normal to the plane of contact. An example of a planar joint is illustrated in Figure 1-8.

A combination of two intersecting revolute joints forms a Universal joint, U, and allows 2-DOF between
paired elements. Figure 1-9 depicts a universal joint. A universal joint imposes four constraints.

Figure 1-3: Prismatic (P) Joint

Figure 1-4: Revolute (R) Joint

Figure 1-5: Cylindrical (C) Joint

Figure 1-6: Helical (H) Joint

Figure 1-7: Spherical (S) Joint

Figure 1-8: Planar (E) Joint

 Figure 1-9: Universal (U) Joint

1.4.2 Degrees of Freedom of a Mechanism

When designing a mechanism, it is important to first establish the number of degrees of freedom of the
mechanism. The number of degrees of freedom is the number of parameters or inputs that are required to
completely specify the mechanism’s position and orientation in space. The degrees of freedom of a mechanism
is influenced by the number of links, number of joints and types of joints. An equation can be developed to
calculate the number of degrees of freedom of a mechanism. This equation is known as the Grübler/Kutzbach
criterion.

1.4.3 The Grübler/Kutzbach criterion

The following variables are used to develop the equations:
• 𝑐𝑐𝑖𝑖 = The number of constraints that joint 𝑖𝑖 imposes
• 𝐹𝐹 = The number of degrees of freedom of the mechanism
• 𝑓𝑓𝑖𝑖 = Degrees of relative motion that is permitted by joint 𝑖𝑖
• 𝑗𝑗 = The number of joints of the mechanism. This assumes that all joints are binary. See the paragraph
below for the description of a binary joint.
• 𝑗𝑗𝑖𝑖 = The number of joints with 𝑖𝑖 degrees of freedom
• 𝐿𝐿 = The number of independent loops that the mechanism possesses
• 𝑛𝑛 = The number of links of the mechanism (inclusive of the fixed/ground link)
• 𝜆𝜆 = The degrees of freedom of the space in which the mechanism was designed to function

A binary joint is formed when two links are joined at the same connection. A ternary joint is created when three
links are joined at the same connection and counts as two binary joints. A quaternary joint is created when four
links are joined at the same connection and is equivalent to 3 binary joints. In general, for 𝐼𝐼 number of links
joined at the same connection, the equivalent number of binary joints is equal to (𝐼𝐼 − 1).
Figure 1-10 illustrates binary, ternary and quaternary joints. There is only one binary joint which is joint A.
Ternary joints are seen at joint B, C, E and F. Quaternary joints are seen at joint D and G.

B
4 11
2 C
3
D 6
5
G
8
7 10

9
E F

Figure 1-10: Diagram depicting binary, ternary and quaternary joints

Concerning the variable 𝜆𝜆, a single value applies to the motions of all moving links. That is, all the links operate
in the same working space therefore 𝜆𝜆 = 6 for spatial mechanisms and 𝜆𝜆 = 3 for planar and spherical
mechanisms.

When determining the degrees of freedom of a mechanism, the degrees of freedom associated with all the
moving links and the number of constraints that the joints impose must be first considered. Hence, the degrees
of freedom would equal to the degrees of freedom associated with all the moving links minus the number of
constraints that the joints impose. Suppose that all the links of a mechanism are free of constraints, the degrees
of freedom of an n-link mechanism, with one link fixed to the ground, would be equal to 𝜆𝜆(𝑛𝑛 − 1). Now,
𝑗𝑗
considering the total number of constraints imposed by the joints (equal to ∑𝑖𝑖=1 𝑐𝑐𝑖𝑖 ) the degrees of freedom of
a mechanism is usually given by:
𝑗𝑗

𝐹𝐹 = 𝜆𝜆(𝑛𝑛 − 1) − � 𝑐𝑐𝑖𝑖 (1.1)

𝑖𝑖=1

The motion parameter, 𝜆𝜆, is equal to the degrees of freedom permitted by the joint and the number of
constraints imposed by the joint. This is illustrated by the following equation:
𝜆𝜆 = 𝑐𝑐𝑖𝑖 + 𝑓𝑓𝑖𝑖
(1.2)
The total number of constraints that the all joints impose is:
𝑗𝑗 𝑗𝑗 𝑗𝑗

� 𝑐𝑐𝑖𝑖 = �(𝜆𝜆 − 𝑓𝑓𝑖𝑖 ) = 𝑗𝑗𝑗𝑗 − � 𝑓𝑓𝑖𝑖 (1.3)

𝑖𝑖=1 𝑖𝑖=1 𝑖𝑖=1

By substituting Equation (1.3) into (1.1), the Grübler/Kutzbach criterion is defined:

𝐹𝐹 = 𝜆𝜆(𝑛𝑛 − 𝑗𝑗 − 1) + � 𝑓𝑓𝑖𝑖
(1.4)
𝑖𝑖
Example 1 – 1

The parameters of the planar mechanism, seen in Figure 1-11 are as follows:

• 𝜆𝜆 = 3 as this is a planar manipulator.

• 𝑛𝑛 = 4 as this is the total number of links.
• 𝑗𝑗 = 𝑗𝑗1 = 4 as the manipulator possesses 4 revolute joints.
• 𝑓𝑓𝑖𝑖 = 4 as since each revolute joint permit 1 DOF (4 × 1 = 4).
• The DOFs can be calculated using the Grübler/Kutzbach equation.

𝐹𝐹 = 3(4 − 4 − 1) + 4 = 1

Figure 1-11: A four-bar linkage mechanism

Example 1 – 2

The parameters of the Five-bar, 5R manipulator, seen in Figure 1-12, are as follows:

• 𝜆𝜆 = 3 as this is a planar manipulator.

• 𝑛𝑛 = 5 as this is the total number of links.
• 𝑗𝑗 = 𝑗𝑗1 = 5 as the manipulator possesses 5 revolute joints.
• 𝑓𝑓𝑖𝑖 = 5 as since each revolute joint permit 1 DOF (5 × 1 = 5).

𝐹𝐹 = 3(5 − 5 − 1) + 5 = 2

The Five-bar, 5R manipulator therefore has 2 degrees of freedom. Link 2 and Link 3 are the inputs and can be
used to control the position of point Q anywhere on the plane.
 Point Q

5 4

Figure 1-12: The Five-bar, 5R manipulator

The Grübler/Kutzbach criterion assumes that the constraints imposed by the joints are independent of each
other and that they do not introduce redundancies. Consider a revolute-spherical binary link chain in which the
revolute joint axis passes through the center of the spherical joint. In this case, the link can is allowed to rotate
about an axis allowed by both the revolute and spherical joint. Hence, a degree of freedom is redundant. This is
called a passive degree of freedom. This combination of joints allows a connecting link to rotate about an axis
defined by both joints. The connecting link possesses the ability to transmit force or torques, hence, motion
about some other axes but it cannot transmit torque about the passive axis.

Generally, binary links with S-S, S-E and E-E pairs possess a passive degree of freedom.

• S-S binary link: The passive degree of freedom is the rotation about an axis through centers of the ball
joints.
• S-E binary link: The passive degree of freedom is the rotation about an axis through the center of the
ball joint and perpendicular to the plane of the plane pair.
• E-E binary link: The passive degree of freedom is the sliding along an axis parallel to the line of
intersection of the planes of the plane pairs. Figure 1-13 illustrates the line of intersection of two planes.
If the two planes are parallel, then there exists 3 passive degrees of freedom.
 Plane 2

Plane 1
Line of
intersection

Figure 1-13: Intersecting planes

A sequence of binary links possessing S-S, S-E and E-E pairs as their terminal joints also have a passive degree
of freedom.

A passive degree of freedom cannot transmit torque therefore it cannot transmit motion about the passive axis.
This influences the Grübler/Kutzbach criterion. When such joint pairs exist in a mechanism, 1 degree of freedom
must be subtracted from the degree of freedom equation. The number of passive degrees of freedom is given
by 𝑓𝑓𝑝𝑝 . The equation is altered to the following:

𝐹𝐹 = 𝜆𝜆(𝑛𝑛 − 𝑗𝑗 − 1) + � 𝑓𝑓𝑖𝑖 − 𝑓𝑓𝑝𝑝

(1.5)
𝑖𝑖

Example 1 – 3

Consider the Stewart-Gough Platform in Figure 1-14. Each limb possesses two binary links connected by a
prismatic joint. The construction of the limb is an S-P-S limb.

S End Effector

S Base

Figure 1-14: Stewart-Gough Platform with passive degrees of freedom

Due to the S-P-S limb combination, there is a passive degree of freedom associated with each limb. Therefore,
the parameters are as follows:

• 𝜆𝜆 = 6 as this is a spatial manipulator.

• 𝑛𝑛 = 14 as this is the total number of links (2 per limb, the end effector and the base).
• 𝑗𝑗1 = 6 as the manipulator possesses 6 prismatic joints (each permits 1 DOF).
• 𝑗𝑗3 = 12 as the manipulator possesses 12 spherical joints (each permits 3 DOFs).
• 𝑗𝑗 = 𝑗𝑗1 + 𝑗𝑗3 = 18 as this is the total number of joints.
• 𝑓𝑓𝑝𝑝 = 6 since there is a passive DOF associated with each limb
• 𝑓𝑓𝑖𝑖 = 42 since each prismatic joint permits 1 DOF (6 × 1 = 6) and each spherical joints permits 3 DOFs
(3 × 12 = 36).

Using the Grübler/Kutzbach criterion we calculate the degrees of freedom of the mechanism as follows:

𝐹𝐹 = 6(14 − 18 − 1) + (3 × 12 + 6) − 6 = 6

1.4.4 Considerations when using the Grübler/Kutzbach Criterion

The Grübler/Kutzbach Criterion considers a single value for 𝜆𝜆 for all the moving links. If there are instances where
a spatial mechanism possesses a planar or spherical linkage as a subsystem, these subsystems must be counted
with care. The 𝜆𝜆 = 3 for the planar or spherical subsystem and 𝜆𝜆 = 6 for the spatial portion of the mechanism.
An example of such a mechanism is a robotic arm, shown in Figure 1-15 and Figure 1-16, with a spherical wrist
mechanism and a planar four-bar subsystem that provides additional support to the robot arm. The four-bar
linkage subsystem provides 1 DOF, the spherical wrist portion provides 3 DOFs and the open loop chain provides
2 DOFs.

Planar
four-bar
subsystem
Spherical
subsystem

Figure 1-15: A 6-DOF robotic arm with planar and spherical subsystems
 Axes of revolution
of the joints of the
spherical subsystem

Planar
four-bar
subsystem

Figure 1-16: Side view of a 6-DOF robotic arm with planar and spherical subsystems

The degrees of freedom, 𝐹𝐹, value can be interpreted as follows:

• If 𝐹𝐹 > 0, then the mechanism possesses 𝐹𝐹 degrees of freedom

• If 𝐹𝐹 = 0, then the mechanism cannot move and becomes a structure with zero degrees of freedom
• If 𝐹𝐹 < 0, then the mechanism is statically indeterminate and with redundant constraints.

There are also mechanisms that do not obey the Grübler/Kutzbach Criterion. Mechanisms that require special
link lengths to achieve mobility are called overconstrained mechanisms.

Concerning closed-loop mechanisms and mechanical manipulators, the number and the locations of the
actuated joints should be selected carefully so that the end effector can be controlled at will. Generally, the
number of actuated joints should be equal to the number of degrees of freedom of the mechanism. The locations
of the actuated joints should be selected so that they form a set of independent coordinates. Should the number
of actuated joints be less than the number of degrees of freedom, then the end effector cannot be controlled at
will. If the number of actuators is greater than the number of degrees of freedom, then the motion of these
actuators must be coordinated according to their kinematic constraints. A redundantly driven manipulator can
be designed for eliminating gear backlash and clearances concerned with the geometry of the joints.

When a robot possesses 7 DOFs, there is no unique solution to the position and orientation of the end effector.
If for instance the robot is performing a part handling task and possesses 7 DOFs, there are infinite ways to
position or orientate the part at the intended location. To assist the robot in its choice, an additional decision-
making routine needs to be included to select one of the infinite solutions. This can be achieved using an
optimisation algorithm that focusses on one or multiple objectives such as finding the shortest distance,
selecting the fastest robotic movements, performing the operation with the stiffest robot pose, etc. For these
reasons, 7 DOF robots are usually not implemented for industrial tasks.

There are instances when a robot can gain or lose a DOF. These circumstances require the study of a
phenomenon known as singularities. In robotic systems, joints are actuated which result in the movement of
the end effector. Therefore, the joint velocities influence the end effector velocity. The mapping of the joint
space velocities to the end effector velocities can be achieved by formulating a Jacobian matrix. Mathematically,
a singularity occurs when the mapping between the joint space and the end effector space cannot be
accomplished. The robot can become dangerous or unstable at or in the neighbourhood of singular points.

When a manipulator gains a DOF, it is unable to resist force or torques in some directions which results in the
manipulator losing its stiffness. Here, there is no movement of the joints yet the end effector is still capable of
movement. When a manipulator loses a DOF, there is movement at the joints that results in no movement of
the end effector. The manipulator can passively resist forces in certain directions without any actuator forces or
torques.

1.4.5 Loop Mobility Criterion

When links are connected by joints and assembled together, this establishes a kinematic chain. This chain forms
one or more closed loops if every link is connected to at least two other links. The number of independent loops
is the total number of loops excluding the external loop. A four-bar linkage mechanism is a kinematic chain
possessing one independent loop. Here the four-bar linkage mechanism is comprised of 4 links and 4 joints. For
a single loop kinematic chain, the number of links equals the number of joints (𝑛𝑛 = 𝑗𝑗).

A single loop chain can be extended into a two-loop chain by joining an open loop chain to a single loop chain.
The two ends of the open loop chain are attached to the single loop chain by two joints as depicted in Figure
1-17. When one loop is extended to two loops, the number of joints exceeds the number of links by one. An
open loop chain can also be added to a two-loop chain to form a three-loop chain and so on.

B B’

Loop 2
Loop 1

A A’

Figure 1-17: Forming a multiloop chain

When a kinematic loop is extended from one loop to 𝐿𝐿 loops, the difference between the number of joints and
the number of links is increased by 𝐿𝐿 − 1. The following equation can be developed:

𝑗𝑗 = 𝑛𝑛 + 𝐿𝐿 − 1 (1.6)

Or we can rewrite the equation as follows:

𝐿𝐿 = 𝑗𝑗 − 𝑛𝑛 + 1 (1.7)

The equation above is known as Euler’s Equation. This equation indicates that the number of independent loops
exceeds the difference between the number of joints and the number of links by a value of 1.

The loop mobility criterion equation can be formed by combining Euler’s Equation and the Grübler/Kutzbach
Equation:

� 𝑓𝑓𝑖𝑖 = 𝐹𝐹 + 𝜆𝜆𝜆𝜆 (1.8)

𝑖𝑖

Usually, the number of loop closure equations that can be developed for a mechanism is equal to the number
of independent loops in the mechanism.

1.5 Robotic Systems

A robotic manipulator does not exist alone. It needs to be actuated by some means which also must be
controlled. The robotic manipulator must also be monitored if it is moving accurately in space and accurately
performing its intended task. Figure 1-18 shows the different elements of a typical robotic system and their

relation to each other.

Workpiece Vision or Laser

Robot
tracking system

Motor
Feedback
Control

Feedback Feedback
Controller Computer
Command

Figure 1-18: A typical robotic system

The typical robotic system is comprised of a mechanical manipulator, a workpiece, controller, computer and
sensing devices such as vision systems, etc. Concerning the mechanical manipulator, one link is fixed to the
ground and another link is identified as the output link. End effectors are usually attached to the output link.
The end effector interacts with the workpiece. The end effector can be understood as the mechanical interface
between the mechanical manipulator and its environment. End effectors are designed to grasp, lift and
manipulate workpieces that are similar in shape, size and weight. These tasks are generally repetitive and it is
not necessary for the end effector to possess a large range of dexterity or versatility. These types of end effectors
are sometimes known as grippers. Most simple grippers are designed as 1 degree of freedom grippers.

Some tasks require the gripper to manipulate a variety of objects with different shapes, sizes and materials. An
end effector, that requires a certain level of dexterity and versatility, are known as a universal gripper or a
dexterous hand. The controller of a mechanical manipulator may range from a low-level PID controller to a high-
level model based intelligent controller. Feedback control may be achieved by employing encoders and/or
tachometers at the manipulator’s joints to measure the relative movement between the two connected links.
Some advanced controllers may require the manipulators to be fitted with force sensors, touch sensors, a vision
system, etc. Teach pendants may also be fitted to a manipulator. The teach pendant can teach position, editing
and running programs and to interface with the computer. The computer may be microprocessor-based
hardware, a personal computer or a mainframe computer with high-speed real-time reasoning, computation
and issuing control commands.

1.6 Classification of Robots

1.6.1 Classifying Robots by Degrees of Freedom
If a manipulator possesses 6 DOFs, then the manipulator can freely manipulate a workpiece’s location and
orientation in space. From this point of view, such type of robotic manipulators is called general-purpose robots.
If a manipulator possesses more than 6 DOFs, then the robot is known as a redundant manipulator. Lastly, if a
manipulator possesses less than 6 DOFs, then it is known as a deficient robot. There are uses for each type of
robot. A redundant robot can move around objects more freely and can operate in tightly confined workspaces.
On the other hand, certain manipulator tasks do not require all 6 DOFs hence a deficient robot would be well
suited. An example of an application is the assembly of components on a plane. In this instance, only 4 DOFs
would be required.

1.6.2 Classifying Robots by Kinematic Structure

Robots can also be classified according to their structural topologies. A robot is classified as a serial robot or
open-loop manipulator if the kinematic structure of the mechanism is in the form of an open-loop chain. A robot
is classified as a parallel manipulator if it possesses a close-loop chain. If a manipulator possesses both open-
loop and closed-loop chains, then the manipulator is known as a hybrid manipulator.

1.6.3 Classifying Robots by Drive Technology

The three most popular drive technologies, that robots can be classified by, are electric, pneumatic and
hydraulic. Most manipulators employ electric DC servomotors or stepper motors because they are clean and are
relatively easily to control. When high-speed or high-payload applications are involved, pneumatic and hydraulic
drive technologies are preferred. The disadvantages of hydraulic drive technologies are the possibilities oil leaks
and the system are flexible due to the bulk modulus of oil. Pneumatic drives are clean and fast however, it is
difficult to control since air is a compressible fluid.
In a serial manipulator, one actuator is usually used to control the motion of each joint. Every link that moves is
driven by one actuator that is mounted on its preceding link through a gear reduction unit. The joint
displacements are kinematically independent of each other. This type of manipulator is called a conventional
serial manipulator. If each joint is driven by an actuator that does not possess speed reducer, the manipulator
is called a direct-drive manipulator.

The advantages of using a gear reduction system are the use of smaller motors which consequently reduces the
inertia of the manipulator. The drawback of gear reduction system is that the robotic manipulator would be
prone to position errors of the end effector due to gear backlash. Direct drive motors are larger and heavier and
are commonly used for the first joint of the serial robot whereby the motor can be installed on the base. The
second and third joints can also be driven by a motor mounted on the base using a push-rod linkage or a metal
belt.

Some manipulators make use of other types of drives such as gear trains, tendons and chains and sprockets to
drive their joints. When such transmission devices are employed across several joints of a manipulator, the joint
displacements are no longer independent of each other.

1.6.4 Classifying Robots by Workspace Geometry

Robot workspace is defined as the volume of space that the end effector can reach. There are two different
types of workspaces that are frequently referenced. These are the reachable workspace and the dexterous
workspace. The reachable workspace is described as the volume of space that the end effector can reach in at
least one orientation. The dexterous workspace is defined as the volume of space that the end effector can reach
in all possible orientations. In accordance with these definitions, the dexterous workspace is a subset of the
reachable workspace.

Serial manipulators are usually designed with their first 3 links longer than the remaining links. The first three
links are primarily used to manipulate the position and the remaining links are used to manipulate the
orientation of the end effector. The subassembly associated with the first three links of a serial manipulator is
referred to as the arm and the subassembly associated with the remaining links are referred to as the wrist.
Except for redundant manipulators, the arm usually possesses 3 degrees of freedom and the wrist may possess
1 to 3 degrees of freedom. The wrist is usually designed such that the joint axes intersect at a common point
known as the wrist center. The arm can assume different kinematic structures and can generate different work
envelopes which are referred to as regional workspaces. Robotic manufacturers usually provide information on
the regional workspace.

A robot arm that is comprised of three mutually perpendicular prismatic joints is known as a Cartesian robot.
This type of robot exhibits a rectangular workspace. The wrist center position of a cartesian robot can be
described by the three cartesian coordinates that is related to the three prismatic joints. Cartesian robots are
referred to as gantry robots when the robot is mounted on rails above its workspace.
A robot can be classified as a cylindrical robot if the first or second joint of a cartesian robot is replaced by a
revolute joint. The wrist center position of a cylindrical robot can be defined by a set of cylindrical coordinates
that are associated with the three joint variables. Due to the mechanical limits of the prismatic joint, the
workspace of a cylindrical robot is confined by two concentric cylinders of finite length.

A robot is classified as a spherical robot if the first two joints are two intersecting revolute joins and the third
joint is prismatic. Usually, the prismatic joint is not parallel to the second joint axis. A set of spherical coordinates,
associated with the three joint variables, is used to describe the wrist center position. The workspace of a
spherical robot is confined by two concentric spheres.

A robotic arm with all three joints as revolute joints is said to be an articulated robot. The workspace is very
complex and generally exhibiting a crescent shaped cross section. Many of the commonly found industrial robots
are of the articulated type.

The SCARA (selective compliance assembly robot arm) robot consists of two revolute joints followed by a
prismatic joint. All three joint axes are parallel to each other and normally point along the direction of gravity.
The design of the SCARA robot relieves the first two actuators from having to work against gravitational forces
of the links and the payload. The SCARA robot has 4 degrees of freedom and finds applications in part assembly
along planes.

1.6.5 Classifying Robots by Motion Characteristics

A robot’s motion profiles are influenced by the types of joints, joint layout, number of joints, number of links,
etc. A rigid body moving in space displays planar motion if all points on the body describe plane curves that lie
in parallel planes. A mechanism is classified as a planar mechanism when all moving links of the mechanism
perform motions that are parallel to one another. The loci of all points in all links can be drawn on a plane. Planar
mechanisms that only employ lower-pair joints are called planar linkages. The only permissible lower-pairs for
planar linkages are revolute and prismatic joints. In planar linkages, the axes of the revolute joints must be
normal to the plane of motion and the direction of translation of a prismatic joint must be parallel to the plane
of motion. A manipulator is a described as a planar manipulator if its mechanism is a planar mechanism. These
manipulators are employed to manipulate objects along a plane.

If all particles in a rigid body describe curves that lie on concentric spheres, then the rigid body displays spherical
motion. For a rigid body to perform spherical motion, at least one point must be stationary. Therefore, when a
rigid body rotates about a fixed axis, it can be considered a special case of spherical motion since any point that
lies on the axis of revolution can be treated as the stationary point. A mechanism can be classified as a spherical
mechanism if all the moving links perform spherical motion about a common stationary point. The motions of
all particles can be determined by their radial projections on the surface of a unit sphere. The only lower-pair
joint that can be used in the construction of spherical linkages is the revolute joint. All joint axes constituting a
spherical linkage must intersect at a common point. A manipulator is classified as a spherical manipulator if is
comprised of a spherical mechanism. Spherical manipulators are suitable as pointing devices.
If a manipulator’s motion cannot be described as planar or spherical, then it is classified as spatial motion. A
manipulator is classified as a spatial manipulator if at least of one of the moving links exhibits a general spatial
motion. Unique motion characteristics cannot be associated with a spatial manipulator. Planar and spherical
manipulators can be considered as special cases of spatial mechanisms. The orientations of the joint axes,
concerning planar and spherical manipulators, lead to a special geometry of the manipulators. These special
geometries allow the synthesis and analysis of planar and spherical manipulators to be greatly simplified. The
selection of the type of manipulator depends on the application, working environment and other factors.

1.7 Robot Characteristics

Robots possess different characteristics and specifications that are influenced by design, material, size,
manufacturing methods, type of actuation, etc. Below are a few robot specifications that are among the most
important robot specifications.

1.7.1 Payload
Payload refers to the maximum weight that the robot can carry and still maintain all other specifications.
Although a robot’s maximum payload capacity may be larger than the specified payload, at the maximum
payload capacity, the robot may become less accurate, deviate from its intended trajectory and incur excessive
deflections.

1.7.2 Reach
Reach is defined as the maximum distance that the robot can reach within its workspace. There are points within
the robot’s workspace that can be reached with all possible orientations, however, at points closer to the
extreme limits of the workspace, the orientations that can be achieved by the end effector is limited. This
limitation is due to the dependence on the robot’s joints, link lengths and its configuration.

1.7.3 Accuracy
Accuracy describes how close the end effector is to its intended location (or true value). This is influenced by the
resolution of the actuators and the capabilities of the robot’s feedback system. The accuracy value is also
dependent on the number of positions and orientations used during testing, load, speed and other aspects of
the robot.

1.7.4 Repeatability
Repeatability refers to how accurately the robot can reach the same position if the trajectory is repeated several
times. The robot may not reach the same point for all repetitions but should reach a point in close proximity
which is a certain radius away from the intended point. The radius of the circle that is formed by the repeated
motions is recognised as the repeatability. Repeatability is also a function of the pose of the robot. If the robot
exhibits a consistent error, such as 0.5 mm to the left of the intended point, this can be corrected through
programming. However, if the error cannot be predicted and is instead random, it cannot be easily eliminated.