Module No.

3: Robotic Cell Design and Programming

[Robotic cell layout, work cell design and control, error detection and recovery, robot path control
methods, robotic programming types, and languages, methods of robot programming, motion
interpolation, capabilities and limitations of lead through methods]

Robotic Cell
A robotic/robot cell/workcell is a complete system that includes the robot, controller, and
other peripherals such as a part positioner and safety environment.

Robot Work Cell Design:

– Physical design of work cells
– Control system to coordinate components
– Evaluation of anticipated performance
Considerations for Designing a Work Cell:
• Determine what job the work cell will be doing.
• Develop an outline of the tasks.
• Choose robots with the right applications to do the job.
• Select a positioner to move the product down the line.
• Create a layout to efficiently and safety complete the task.
Robot Work Cell Layouts

Generally, robot cell design comes in two cell materials: steel and aluminium. Aluminium is
lightweight, and it is easy to modify or extend with no need for welding or painting. Steel has a
rugged appearance, and it is used for heavy works, for example, the metals processing industries.

Types of Robot Work Cell

The robotic cell layouts can be classified into three basic type:
1. Robot-centered cell
2. In-line robot cell
3. Mobile robot cell

Robot-Centered Workcell
In the robot-centered cell, the robot is located at the approximate centre of the cell and the
equipment is arranged in a partial circle around it.
• Robot is positioned at approximate centre of
work cell.
• Other components, equipment are arranged
around it.
• This layout allows for high utilization of robot.
• Parts to be presented in known location and
orientation (usage of conveyors, part-feeders,
pallets).
In-Line Robot Cell
With the in-line cell arrangement, the robot is located along a moving conveyor or other handling
system and performs a task on the product as it travels past on the conveyor.

• One or more robots are located along an in-line conveyor or other material transport
system.
• Work is organized so that parts are presented to the robots by the transport system.
• Each robot performs some processing or assembly operation on each part.

Typical applications such as in welding lines used to spot-weld car body frames, usually utilizes
multiple robots.

Mobile Robot Cells

In mobile robot cell design, the robot is capable of moving to the various pieces of equipment
within the cell.

The robot is provided with a means of transport within the work cell like floor mounted tracks or
overhead railing system that allows the robot to be moved along linear paths and perform various
tasks at different locations.

Mobile robot work cells are suitable for installations where the robot must service more than one
station (production machine) that has long processing cycles, and the stations cannot be arranged
around the robot in a robot-centred cell arrangement.
Robot Path Control Methods:
The basic path control functions of a robot work cell are:
• Sequence Control
• Human (Operator) Interface
• Safety Monitoring

1. Sequence Control:
This is the basic function of the work cell controller. Sequence control includes:
– regulate the sequence of activities
– control of simultaneous activities
– making decisions to proceed/stop/delay work cycle

As almost all the activities occur sequentially and therefore,

- Controller must ensure activities occur in correct sequence and that each step is completed
before the next is started.
- Controller must communicate back and forth with the various equipment (machining
centre, conveyors and robot).
- Signals must be sent by the controller, and other signals must be received from the
components. These signals are called interlocks.

Additional functions within the sequence control:

- Logical decision making: To make decision on which action to activate first, giving a
certain situation.
- Computations: Each cycle places part in different slots, which requires computation of the
new position each time. In path planning, the robot may choose the minimum-time path to
go between some given points.
- Irregular elements: These are events that occurs irregularly.
- Exceptional events: Coping with exceptional events such as equipment breakdown and
recovering from a power failure.

2. Human (Operator) Interface:

Operator interface is to provide a means for operator to interact with robot work cell in order to:
- Program the robot, modify and update programs
- Let human operator participate in work cycle, such as man and robot each performing a
portion of work. Humans involve in tasks that require judgement and sensory capabilities
that the robot does not possess.
- Do data entry by human operator such as for part dimensions, part identifications.
- Do emergency stopping of activities.
3. Safety Monitoring:
Safety monitoring (or hazard monitoring) is a work cell control function where sensors are used
to monitor status and activities of the cell, to detect the unsafe or potentially unsafe conditions.

The safety monitoring is programmed to respond to various hazard conditions in different ways:
- Complete stoppage of cell activities
- Slowing down the robot speed to a safe level when human is present
- Warning buzzers to alert maintenance personnel of a safety hazard
- Specially programmed subroutines to permit the robot to detect errors and recover from a
particular unsafe event

Interlocks are essential for the coordination and synchronization of activities which allow for
variations in the times taken for certain elements in the work cycles.

- Input interlocks. Input interlocks make use of signals sent from the components in the cell
to the controller.

- Output interlocks. Makes use of signals sent from the controller to other devices or
machines in the work cell.

In some cases, the interlock signals can be generated by the electronic controllers for the machines.

In other cases, the applications engineer must design the interlocks using sensors to generate the
required signals.

ERROR DETECTION AND RECOVERY

Hardware malfunctions and unexpected events will cause costly delays and loss of production.
Usually, in automated processes, human assistance is required to intervene, diagnose and make
repairs and then restart the system.

When a computer is used to detect and correct errors, this is known as “error detection and
recovery”.

The implementation involves the use of sensors and programming.

Error Detection
The sensor signals must be interpreted so that errors can be recognized and classified. This needs
some form of intelligence in processing the error signals.
In general, errors in manufacturing can be classified into:
– Random errors: result from stochastic phenomena and characterized by their statistical nature.
[An example of this randomly varying part sizes in a machining operation.]
– Systematic errors: not determined by chance, but by some bias that exists in the process.
[For example, incorrect machine setting or fixture setting will likely result in systematic error in
the product.]
– Illegitimate error: result from an outright mistake, either by equipment or human error.
[An example of this error in the robot program.]

Error Source Particular Malfunction or Error

Category
Tooling Tool wear-out, breakage, vibration, tool not present, wrong tool loaded.
Work Work part not present, wrong work part, defective work part.
Process Wrong part program, wrong part, chip fouling, no coolant when there should
be, vibration, excessive force, cutting temperature is too high.
Fixture Part not in fixture, part located improperly, part deflection, part breakage,
chips causing location errors, hydraulic or pneumatic failure.
Machine Tool Vibration loss of power, power overload, thermal deflection, mechanical
failure, hydraulic or electrical failure.
Robot/ Improper grasping of work part, no part present at pickup, hydraulic or
end-effector electrical failure, loss of positioning accuracy, improper handling.

Error Recovery
This is concerned with defining and implementing the strategies that can be employed by the robot
to correct or compensate for the malfunction that has occurred. The recovery strategies can be
grouped into some general categories:
- Adjustments at the end of current cycle: At the end of current cycle, the robot program
would branch to a subroutine to make the required corrections, then branch back to the
main program.
[Example, the robot dropped a part and the logical action is to go for the next part]
- Adjustments during current cycle: Error is sufficiently serious, so corrective action must
be taken during the current cycle of operation without stopping it.
[Example, an oversized part is present, therefore invoke additional program to machine the
extra material]
 - Stop the process and invoke corrective algorithm: The error requires the process to be
stopped, and a subroutine is called to correct the error. At the end of the correction
algorithm, the process can be resumed or restarted.
[Example, tool failure, therefore change tool and replace the damaged part.]
- Stop the process and call for help: This is when the malfunction cannot be corrected by
the robot or due to an unclassified error to which no corrective algorithm has been
designed. Here, human assistance is needed to restore the system.
[Example, hydraulic system break-down, which means automatic recovery is not possible,
therefore needs human assistance.]
The ‘error detection and recovery system’ is implemented by means of the sensors used in the
work cell together with the robot programming system.

WORK CELL CONTROLLER

The work cell controller deals with the coordination of the robot’s activities with those of other
equipment in the cell.
There are a number of candidates for the work cell controller:
- robot controller
- relays
- programmable logic controllers (PLC)
- small stand-alone computers
The choice among them depends on the following complexity of the cell:
- number of separate control actions,
- number of separate pieces of equipment,
- number of robots.

ROBOT PROGRAMMING TYPES

According to the consistent performance by the robots in industries, the robot programming can
be divided in two common types such as:
1. Leadthrough Programming Method
2. Textual Robot Languages

Leadthrough Programming Method:

During this programming method, the traveling of robots is based on the desired movements, and
it is stored in the external controller memory. There are two modes of a control system in this
method such as a run mode and teach mode. The program is taught in the teach mode, and it is
executed in the run mode.

This system of programming was initially popular but has now almost disappeared. It is still
however used by many paint spraying robots. The robot is programmed by being physically moved
through the task by an operator. This is exceedingly difficult where large robots are being used
and sometimes a smaller version of the robot is u ed for this purpose. Any hesitations or
inaccuracies that are introduced into the program cannot be edited out easily without
reprogramming the whole task. The robot con roller simply records the joint positions at a fixed
time interval and then plays this back.

The lead through programming method can be done by two methods namely:
Powered Leadthrough Method
Manual Leadthrough Method

a) Powered Leadthrough Method:

The powered leadthrough is the common programming method in the industries. A teach pendant
is incorporated in this method for controlling the motors available in the joints. The logic for the
program can be generated either using a menu based system or simply using a text editor but the
main characteristic of this method is the means by which the robot is taught the positional data. A
teach pendant with controls to drive the robot in a number of different co-ordinate systems is used
to manually drive the robot to the desired locations. These locations are then stored with names
that can be used within the robot program.

It is also used to operate the robot wrist and arm through a sequence of points. The playback of an
operation is done by recording these points. The control of complex geometric moves is difficult
to perform in the teach pendant. As a result, this method is good for point to point movements.
Some of the key applications are spot welding, machine loading & unloading, and part transfer
process.

b) Manual Leadthrough Method:

In this method, the robot’s end effector is moved physically by the programmer at the desired
movements. Sometimes, it may be difficult to move large r b t arm manually. To get rid of it ateach
button is implemented in the wrist for special programming. The manual leadthrough method is
also known as Walk Through method. It is mainly used to perform continuous path movements.
This method is best for spray painting and arc welding operations

Textual Robot Languages:

In 1973, WAVE language was developed, and it is the first textual robot language as well. It is
used to interface the machine vision system with the robot.

Then AL language was introduced in 1974 for controlling multiple robot arms during arm
coordination.

VAL was invented in 1979, and it is the common textual robot language. Later, this language was
dated in 1984, and called as VAL II.

The IBM Corporation has established their two own languages such as AML and AUTOPASS,
which is used for the assembly operations.

Other important textual robot languages are:

- Manufacturing Control Language (MCL),
- RAIL, and
- Automatic Programmed Tooling (APT) languages.
ROBOT PROGRAMMING METHODS
There are three basic methods for programming industrial robots but currently over 90% are
programmed using the teach method.

Teach Method

The logic for the program can be generated either using a menu based system or simply using a
text editor but the main characteristic of this method is the means by which the robot is taught the
positional data. A teach pendant with controls to drive the robot in a number of different coordinate
systems is used to manually drive the robot to the desired locations.

These locations are then stored with names that can be used within the robot program. The
coordinate systems available on a standard jointed arm robot are :

Joint Co-ordinates
The robot joints are driven independently in either direction.

Global Co-ordinates

The tool centre point of the robot can be driven along the X, Y or Z axes of the robots
global axis system. Rotations of the tool around these axes can also be performed

Tool Co-ordinates
Similar to the global co-ordinate system but the axes of this one are attached to the tool centre
point of the robot and therefore move with it. This system is especially useful when the tool is
near to the workpiece.

Workpiece Co-ordinates

With many robots it is possible to set up a co-ordinate system at any point within the
working area. These can be especially useful where small adjustments to the program are required
as it is easier to make them along a major axis of the co-ordinate system than along a general line.
The effect of this is similar to moving the position and orientation of the global co-ordinate system.

This method of programming is very simple to use where simple movements are required.
It does have the disadvantage that the robot can be out of production for a long time during
reprogramming. While this is not a problem where robots do the same task for their entire life, this
is becoming less common and some robotic welding systems are performing tasks only a few
times before being reprogrammed.

Lead Through

In this system, the robot is programmed by being physically moved through the task by an
operator. This is exceedingly difficult where large robots are being used and sometimes a smaller
version of the robot is used for this purpose.
Off-line Programming

The program structure is built up where intelligent tools are available and useful to generate
sequences of location and process information. The benefits of this form of programming are:

· Reduced down time for programming.

· Programming tools make programming easier.
· Enables concurrent engineering and reduces product lead time.

PROGRAMMING LANGUAGES FOR ROBOTICS

URBI: URBI stands for Universal Real-time Behavior Interface. It is a client/server based
interpreted language in which Robot works as a client and controller as a server. It makes us to
learn about the commands which we give to Robots and receive messages from them. The
interpreter and wrapped server are called as "URBI Engine".

PYTHON: Python is an object-oriented language which is used to access and control Robots.
Python is an interpreted language; this language has an application in working with mobile robots,
particularly those manufactured by different companies. With python it is possible to use a single
program for controlling many different robots.

ROBOTC: Other Languages which we use are C,C++ and C # etc. or their implementation, like
ROBOTC, ROBOTC is an implementation of C language. It is actually a text based programming
language. The commands which we want to give to our Robot, first written on the screen in the
form of simple text.

Although commands are given in text form (called as codes) but this language is very specific
about the commands which is provided as instruction.

ROBOTICS.NXT: ROBOTICS.NXT has a support for a simple message-based control. It direct

commands, nxt-upload is one of its programs which is used to upload any file. It works on Linux.

Microsoft Robotics Developer Studio

Microsoft Robotics Developer Studio is an environment given for simulation purpose of Robots.
It is based on a .net library concurrent implementation.
WAVE and AL: WAVE, developed at Stanford, demonstrated a robot hand—eye coordination.
Later a powerful language AL was developed to control robot arms. WAVE was incorporated with
path calculations through coordination of joint movements, end-effector positions and touch
sensing.

AML: A manufacturing language, AML was developed by IBM. AML is very useful for assembly
operations as different user—robot programming interfaces are possible. The programming
language AML is also used in other automated manufacturing systems.
The advantage of using AML is that integers, real numbers and strings can be specified in the
same aggregate which is said to be an ordered set of constants or variables.
MCL: US Air force ICAM project led to the development of another manufacturing control
language known as MCL by McDonnel—Douglas. This is a modification of the popular APT
(Automatically Programmed Tooling) language used in CNC machine tools as many similar
commands are used to control.

RAIL: It was developed by Automatic for robotic assembly, inspection, arc Welding and
machine vision. A variety of data types as used in PASCAL can be used. It uses Motorola 68000
type microcomputer system; It supports many commands and control of the vision system.

HELP: HELP was developed by General Electric Company. It acts more or less like RAIL. It has
the capability to control two robot arms at the same time. The structure of the language is like
PASCAL.

JARS: JARS was developed by NASA JPL. The base of the language is PASCAL. JARS can be
interfaced with PUMA 6000 robot for running robotic programs.

RPL: RPL was developed at SRI International. A compiler is used to convert a program into the
codes that can be interpreted by an interpreter. Unimation PUMA 500 can be controlled with the
help of RPL. The basic ideas of LISP (an Al language) have been organized into a FORTRAN-
like syntax in RPL. It is modular and flexible.
Besides these, there are some other languages like PAL, ADA etc. PAL has been written by
Richard Paul by modifying WAVE and incorporating features of PASCAL. But the
representations of syntaxes used in the program are difficult to handle. ADA developed by the
Department of Defense (DOD) in USA is a real-time system that can be run on several
microcomputers like Zilog, VAX, Motorola 68000, etc. ADA is convenient for controlling the
robots used in a manufacturing cell.
MOTION INTERPOLATION
Interpolated motion is a type of motion exhibited by point-to-point robots.

In many instances, the path will not be completely specified. Knot points such as intermediate via
points and endpoints will be specified. Interpolated motion is called for when the path that an
object takes through space is important.

The path planning interpolates between the knot points to produce a smooth the path that can be
executed using continuous-path motion control techniques.

Motion interpolation provides a much faster alternative to creating new motions through the same
means.

Using interpolation, motions can be generated in real time while preserving the realistic qualities
of the example motions.

The high-level languages to program robots are based on three types of movement:
1. joint interpolation (MOVEJ),
2. linear interpolation (MOVES), and
3. circular arcs (MOVEC).
CAPABILITIES AND LIMITATIONS OF LEAD THROUGH METHODS
Capabilities:
i) During this programming method, the traveling of robots is based on the desired
movements, and it is stored in the external controller memory.
ii) There are two modes of a control system in this method such as a run mode and teach
mode. The program is taught in the teach mode, and it is executed in the run mode.
iii) In powered leadthrough method, a teach pendant is incorporated used to operate the
robot wrist and arm through a sequence of points. This method is good for point-to-
point movements. Some of the key applications are spot welding, machine loading &
unloading, and part transfer process.
iv) In manual leadthrough method, the robot has end effectors with desired movements.
This method is also known as Walk Through method. It is mainly used to perform
continuous path movements. This method is best for spray painting and arc welding
operations.
Limitations:
- Lead through programming is not readily compatible with modern computer based
technology.

- Robot cannot be used in production, while it is being programmed.

-------------------------------------------------------------------------------------------------------------------