ROBOTICS AT USC

WeUPSTATE
focus on developing novel and innovative robotics applications.

First, lets talk about the

RULES!!!!

RULE 1: If we are doing an in-class assignment or activity and you are not
in class that day, you automatically loose 50 points from this
assignment/activity grade and cannot make this up.

RULE 2: Absolutely no food or drinks of any kind on your robot

table/workarea.

RULE 3: You must leave your work station exactly the way you found it your workstation must be properly cleaned up after activities/assignments
are completed.

RULE 4: When debugging or running a program, be prepared to press the

Emergency Stop button at any time.

More RULES!!!!

RULE 5: Check that no-one is within the robot work envelope when
switching arm power on or moving the robot. Always take tool offsets into
account. Putting someone in an un-safe situation results in an
automatic F in the course. No exceptions.

RULE 6: When running a program, the speed on the arm MUST NEVER be
set higher than 10% without explicit approval from the instructor first.
Putting anybody in an un-safe situation results in an automatic F in
the course. No exceptions.

RULE 7: Always be aware of your surroundings. Alert others IMMEDIATELY

if you observe a possibly dangerous situation. Putting anybody in an unsafe situation results in an automatic F in the course. No exceptions.

Course Grading

Machines in the lab

Five Stubli RX60 arms with CS7B controllers

6 dof, Programming environment: V+
Three Adept 550 arms
4 dof, Programming environment: V+
One Staubli RS20
4 dof, Programming environment: VAL3
One Fanuc LR Mate 200iC
6 dof, Programming environment: TPP

Stubli RX60s

Some Notable features:

Cables pass through inside of joint gearing
Stubli first to make this work
All motors enclosed
Reduces particle emissions
The two-click air-pressure release value
Designed by Stubli and was an important contribution to
industrial robotics.
Maintenance is easy
Show in class how to check oil and feel for wear.
Repair Process (typical in industry)
Motors, amps, etc that fail are simply swapped for new
components. No trying to fix the faulty component Why?

Whats going on inside the

machine?
Not the focus of what we do, but here are the

basics.
Inside the arm
Motor for
joint 1
Motor for
joint 3
Motor for
joint 2

Whats going on inside the

machine?
Each joint has
an amplifier that
controls the
electricity to it.

Of course several other

electronics are present, a
hard drive, CPU, etc, etc

Community Partnerships

Stubli

Initial donations of our robotic equipment

Paid internship available
This could be

SEW Eurodrive

Paid internship available

$30,000 cash donation to fund:

YOU

Student Research Assistant here in robotics lab!

Robotics Summer Camp

Materials, food, pay a student assistant

About Stubli

Stubli is a mechatronics solution provider

with three dedicated divisions: Textile
machinery, Connectors and Robotics.
With a workforce of over 3000, the company
generates a yearly turnover surpassing 1
billion Swiss francs.
Originally founded 1892 as a small workshop
in Horgen / Zurich, Stubli today is an
international group with its head office in
Pfffikon, Switzerland.

About SEW Eurodrive

A world leader in drive technology and a pioneer in drivebased automation.

Power transmission and motion control products.
Introduced the gearmotor in 1931.
History of innovations the first variable speed
gearmotor, early development of electronic drives, some
of the first successful efforts to decentralize control and
the first motor with energy-efficient copper rotors.

Curriculum Development

Available Courses:

SCSC 314 Introduction to Robotics

SCSC 399 Independent Study (with van Delden)
SCSC 580 Introduction to Artificial Intelligence
SCSC 585 Introduction to Computer Vision

Automation Focus Area in Computer

Information Systems degree.

A collaboration between Computer Science,

Engineering Technology Management, and
Business.

Curriculum Development
cont

Peer-Reviewed / Published
Several papers published recently:
Research

Sebastian van Delden. 2010. Getting Your Robotic Arms around Computing Curricula. In the Journal of
Computers in Education. Published by the American Society for Engineering Education. vol 1(4). Pp 91-101.
Oct-Dec Volume.
Sebastian van Delden. 2010. Computer Science Meets Industrial Robotics: A Visual Servoing Project for a
Computer Vision Course. In the Journal of Computing Sciences in Colleges. Volume 25, Number 6. Pages
85-92. Select papers from the 15th Annual Northeast Meeting of the Consortium for Computing Sciences in
Colleges. Hartford University.
Sebastian van Delden. 2010. Industrial Robotic Game Playing: An AI Course. In the Journal of Computing
Sciences in Colleges. Volume 25, Number 3. Pages 134-142. Select papers from the 25th Annual Eastern
Meeting of the Consortium for Computing Sciences in Colleges. Villanova University, Pennsylvania. January
2010.
Sebastian van Delden and Nicole Tobias. 2010. A Novel Approach to 3D Contour Recovery using Structured Light
Mounted to a Robotic Manipulator. In the Proceedings of the 15th IASTED International Conference on
Robotics and Applications, Cambridge, MA, Pages 167-173, Nov 1-3, 2010.
Sebastian van Delden and Frank Hardy. 2009. Robotic Eye-in-hand Calibration in an Uncalibrated Environment. In
the Journal on Systemics, Cybernetics and Informatics. Volume 6, Number 6. Pages 67-72.

****Student coauthor

Peer-Reviewed / Published
Several papers published recently:
Research

Sebastian van Delden and Wei Zhong. 2008. Effective Integration of Autonomous Robots into an
Introductory Computer Science Course: A Case Study. Journal of Computing Sciences in
Colleges. Select papers from the Sixth Meeting of the Consortium for Computing Science in
Colleges, vol 23(4), pp. 10-19, April 2008.
Sebastian van Delden and Benjamin Overcash. 2008. Towards Voice-Guided Robotic Manipulator
Jogging. In Proceedings of the 12th World Multiconference on Systemics, Cybernetics and
Informatics. Volume 3. Pages 138-144. Orlando, FL. July 2008.
Sebastian van Delden, Ricky Farr, and Seth Hensley. 2007. An Automated Camera Orientation
Recovery Algorithm for an Eye-in-Hand Robotic Manipulator. In Proceedings of the 5th IEEE
International Workshop on Robotic and Sensors Environments. Pages 1-6. Ottawa, Canada.
October 12-13, 2007.
Sebastian van Delden. 2006. Constructing a Simple Visually-Guided Robotic Part-Grasping System
with Off-the-Shelf Components. In Proceedings of the 18th IEEE International Conference on
Tools with Artificial Intelligence. Pages 211-216. Washington, DC. November 13-15th, 2006.
****Student coauthors

Other Research
Contributions

Alex Umrysh and Sebastian van Delden. 2011. Object recognition in a Robot Workarea
using Hand Gestures. Poster at the 7 th Annual SC Upstate Research Symposium.
April 15, 2011.
Nicole Hodge and Sebastian van Delden. 2010. Contour Recovery using Structured
Light mounted to a Robotic Manipulator. Poster at Discovery Day 2010, USC
Columbia. April 23rd, 2010. ****Best Poster Award for Computer

Science and Engineering

Nicole Hodge, Robert Mahmoudishad, Mark Parrish, and Sebastian van Delden.
2009. A Novel Robotic Approach to Contour Recovery using Structured Light.
Presented at the Fifth Annual USC Upstate Research Symposium. March 27th,
2009
Derrick Thompson, Jose Reyes, and Sebastian van Delden. Spring 2007. Vision-Based
Robots Playing Pong. Presented at the Third Annual USC Upstate Research
Symposium. April 6th, 2007.
William Bittle, Sayed Shahabi, Ashley Bryant and Sebastian van Delden.
WOODBURNER: An Automated Robotic Character Etching System. Poster at the
Second Annual USC Upstate Research Symposium. April 21st, 2006.
*** STUDENT COAUTHORS

Ongoing projects. Need

Help!
Rapid Robotic Application Development using visual and voice

cues

Android/Robot/Bluetooth:

http://www.youtube.com/svandelden#p/u/8/dsdh2WlWwys

Summer Camps:

http://www.youtube.com/uscupstateresearch#p/u/5/UIVM8SByVDs
http://www.youtube.com/watch?v=dxo1NbBblE4

http://www.youtube.com/svandelden#p/u/6/3mItUbcqWms

New ideas?

Have a neat idea you want to implement?

INTRODUCTION TO
INDUSTRIAL ROBOTICS
Sebastian van Delden
USC Upstate
svandelden@uscupstate.edu

Industrial Robotics

Primarily assist in the automaton of the

manufacturing process

78% welding and material handling

10% assembly

Over a billion dollars of new industrial robots

shipped to North America each year on
average.
However, America lags behind several
European and Asian countries.

Trends in Industrial Robots

Industrial Robots currently in

use

Why Automate??

Robot prices continue to drop while labor wages

grow
Faster, more accurate/flexible
Operate in dangerous environments

Example Robot Makers

Maker

Programming Language*

ABB
FANUC
HP
IBM
Motoman (Yaskawa)
Staubli

AMPL
Karel, Proficy
MDS
AML
JRC
V+, VAL3

* Are constantly evolving, with the new trend in off-line

programming environments

What do you study in

Depends
Robotics?

Engineering programs would focus on the actual

physical design of the machine, motors, gearing etc.
Computing programs would focus more on the
application side of robotics: programming,
incorporating input sensors (cameras, etc),
networking/communication, etc..

Programming the machine to do something interesting.

Our focus here at USC Upstate

Tech programs would focus on basic electronics,

maintaining/servicing the machines, etc

BASIC INDUSTRIAL
ROBOTICS CONCEPTS
Which will be covered in this course

Representing Objects in 3D

Position and Orientation

Space

A frame is rigidly attached to each object to define

position and orientation
A frame is just a 3D coordinate system.

Transformations

Moving one frame with respect to another

one is called a transformation.

Matrices and
Transformations

We will learn how to use matrices to

internally represent and manipulate
transformations, and position and orientation
information
Applicable to several fields:

Robotics
3D modeling and programming
Computer Vision

Some Manipulators Concepts

Consists of rigid links and joints which allow

relative motion of neighboring links
The orientation of a rotary joint is measured with
a joint angle.
Orientation of a prismatic (sliding) joint is
measured with a joint offset.
The number of degrees of freedom (dof) is the
number of joints of a manipulators.
Singularity Point Manipulator try to move its
joints at infinite speed to continue fluid motions.
(gunner example)

Some Manipulators Concepts

An end-effector is what is attached at the free

cont
end of the robot

Gripper, pointer, welder, camera, etc..

Represented with a tool frame

A base frame is attached to the non-moving

base. Frames are also assigned to each joint.
The robots work envelope (or work space) is
everything that the robot can reach.
Manipulator jogging is the process of manually
moving the arm around.

Typical frames in a robotic

environment

Typical work envelope

schematics

Input/Output

A robot can gathering data from its environment

with input sensors

Buttons, switches, keyboard, microphone, camera

A robot uses its actuators (motors) and endeffectors (grippers, etc) to effect its
surroundings.
Well learn how to make our robots gather
images from a camera so that it can interact
with the environment based on this visual data.

Kinematics

Treats motion without regard to the forces which

cause it.
Forward Kinematics of Manipulators:

compute position and orientation of the end effector given a

set of joint angles.
EASY(er)

Inverse Kinematics of Manipulators:

given the position and orientation of the end effector,

calculate all possible joint angles that could achieve this
position and orientation.
HARD
Staubli RX60s As many as 8 possible ways to achieve end
effector pose

Typical robot schematic/dimensions needed to calculate

Kinematics

Finally, by the end of this

course, well know how to

Program and operate our Stubli V+ manipulators.

Do simple Stubli VAL3 and Fanuc TPP
programming.
Calculate end-effectors transformations and other
gripper concepts.
Calculate the Forward Kinematics of a manipulator.
Understand basic Inverse Kinematics and trajectory
generation.