1) Introduction: -

Man has been known through the ages as a tool-making creature. One of the jobs his tools couldn’t do was thinking. In his eternal quest for the concept and realization of a thinking machine spawned various disciplines of modern thought more recently artificial intelligence and robotics. This fascination for more intelligent machines has spurred research into flexible manufacturing systems (FMS), that has been at the cutting edge of advances in manufacturing processes over the last decade. Our interest in this field of study prompted us to choose this as our final year project

2) Project Overview: -

The project involves the design and fabrication of a general-purpose programmable robotic arm anthropomorphically similar to a human arm. The robots ability to be programmed makes it ideal to perform a series of mechanical motions that are industrially suited to a variety of production jobs like material handling, painting, welding etc.

The size and shape of the proposed robotic arm is basically a scaled down jointed arm configuration as seen in the Unimation PUMA 500 or the Cincinnati Milacron T3 Robots.

There are basically six degrees of freedom, which provide any robot with the capability to move the end effector through the required sequence of motions intended to emulate the versatility of movement possessed by the human arm. These degrees of freedom (of which 3 are arm and 3 wrist motions) are: -

• Vertical traverse: Up and down motions of the arm caused by pivoting of the entire arm about a horizontal axis.
• Elbow Traverse: Caused by pivotal of the elbow.
• Rotational traverse: Rotation about the vertical axis (left or right Swivel)
• Wrist swivel: Rotation of the wrist
• Wrist Pitch: Up and down rotational movement of the wrist.
• Wrist Yaw: Right and left swivel of the wrist.

Our project will be using five i.e. omitting the wrist yaw motion.

The configuration of the robot in terms of commercially available robots would be similar to that of the Eurobtec IR52c that is available in Germany.

3) Project Objectives: -

Our primary objective would be to design and fabricate a multi axis robotic arm that can be used as a teaching aid for engineering sciences. The arm would be fabricated locally with indigenously developed designs and easily available components thus providing an import substitute for the costlier and less available foreign robots. The design would demonstrate the use of robots in industrial applications and give the student an insight into its programming,

Our secondary objective would be to study and analyze the various optimization methods/algorithms available and if possible attempt to design new method and develop new algorithms for simpler and more efficient algorithms for required motions.

4) Project Details: -

The organization of the project is basically of a modular type where the entire structure is broken down into 3 modules namely:

Module 1: Mechanical robotic linkages

This module has been further sub divided into 5 divisions: -

• Base: Using a 10kg-cm torque motor for basic swivel of entire arm.
• Shoulder joint: Using a 28kg-cm torque motor for shoulder pitching motion.
• Elbow Joint: Uses a 10Kg-cm and permits planar pitch.
• Wrist Joint: As seen in the human arm with the exception of Yaw.
• End Effector: It uses a lightweight aluminium body and plastic gears. Belts/ropes or chains have not been used; thus friction has been kept to a minimum.

Module 2: Electronics

The electronics of this project can be subdivided as:

Signal Processing

The robotic arm will be using stepper motors (6) for realizing the arm movements. A stepper motor too will power the end effector. The calculations and processing will be done on a standard PC/XT/AT 486 or Pentium. The signal will be output through the LPT port line (Line Printer Port). The eight available data lines will be used as follows: -

One line will be to enable the circuitry/laser. Three lines for selecting the six motors (with provision for using 2 extra motors) – using a 3:8 decoder and the remaining 4 lines to supply the step sequence for the stepper motor.

Signal Amplification

The step sequence from module 1 in the form of pulse trains is too weak to drive a motor hence need to be amplified. The Torque of the motors will range from 10Kg-cm to 0.3 Kg-cm with current requirements of 0.6 amps to 4.5 Amps. Since operating frequencies are relatively low, Darlington pair amplifiers have been designed to supply the necessary current, to the motors..

Regulation 

  The current requirements for the motor vary and the required voltages and currents are supplied through 220V AC main power supply. This voltage has to be stepped down to the required amplifier voltages and rectified to DC due to the use of DC Stepper motors.

Module 3: Programming and interface development

We have opted for an interface-kernel type of structure for our program. The kernel will be in QuickBasic and will be a stand-alone Executable. The Interface will be built in the Windows 95^® GUI using Visual Basic. It will be executable in 4 modes namely:

1. Graphical Mode: For movements and operations the user will be required to graphically indicate the start and stop positions and the program will decide the shortest route using a selected algorithm and perform the operation.
2. Programming Mode: With online and offline control with a programming language similar to the Eurobtec IR52c.
3. Teach Mode: This allows the user to manually control the robot so as to teach it a particular movement or series of movements which would be recorded and then replicated on request.
4. Voice Mode: This mode which is the salient feature of our software has an option for Voice activation/control built in order to enable the user to control the robot vocally as a demonstration of "Voice controlled NC methods".

5) Difficulties/Challenges

• The fabrication and design of the signal processing circuitry was a challenge due to its complexity and the number of lines/wires (due to the use of two independent buses), debugging is very difficult and the chances of loose contact are high.

• The power transistors in the amplifier module tend to heat up and hence are to be mounted on a heatsink. Due to the lack of funds we had to construct the Darlington pairs ourselves instead of costlier single package alternatives (MOTOROLA) hence the complexity of the circuitry increases.

• Since the logic of our electronic design allows only one motor to be stepped at a given instant, for PTP (point to point) motions the step data has to multiplexed very efficiently so as to achieve the required motion smoothly. Due to this our program will first have to optimize the route and then multiplex the step data.

• Due to the use of Basic as the primary language, calling of interrupts for real-time delays (in ms) may not be feasible. The effect of this would be that our robot would run faster on a 486 than on a 586 etc. The proposed workaround for this would involve incorporation of a customizable variable, which is configured at installation. Due to this difficulty the kernels for the voice and teach mode are built in Turbo C, which allows real time delay.

• Tolerancing and fabrication of mating gears will pose a challenge due to the extensive use of worm and worm wheel pairs which are extremely sensitive to various parameters such as center distance and alignment.

• Due to the use of a multitasking environment such as Win95, we have experienced difficulty in the use of the port due to sharing with third party programs and simultaneous use of our own kernels. The workaround for this has been the use of only the data pins hence leaving the printer ready line etc. Blind that prevents the use of the printer for printing programs.

6) Costing: -

Module 1: - Mechanical Fabrication

Stepper Motors ranging from 10kg-cm to 0.3kg-cm

Aluminium, Gears, fabrication costs etc.

Module 2: - Electronics - Signal Processing, Signal Amplification

I.C’s (74ls138, 373, 04 etc), LED’s, RPacks, Wires, Development/bread/general purpose boards etc.

Transistors (BC147, 2N3055, SL100, diodes, capacitors

Module 4: - Programming and interface

All the programming and interface testing has been done in-house hence no expenditure is expected excepting the burning of CDs and other methods of data storage. 

Documentation/Reports/Manuals/etc

Miscellaneous costs/accidental wastage/etc including accessories such as FM wireless transmitters and headphones.

 Back to the HomePage