www.MobileRobot.org/RAS/

 

Portland Chapter

 

IEEE

 OREGON SECTION

 

 

 

Next Meetings

Previous Meetings

Announcements

Mission

Links

Contacts

 

 

Next Meetings

 

 

Tuesday February 24th, 2009 at 7 PM

Integrating the Cognex Insight Smart Camera into Industrial Vision Applications 

 

Jason McDaniel, Bruce Hagerty, and Paul Woodhouse
Olympus Controls
Beaverton, OR

 

  • Introduction to Olympus Controls and the Cognex Insight Smart Camera.

  • Constructing vision solutions with the EasyBuilder application.

  • Example selections of optimal optics, lighting, and vision tools.

  • Interfacing to industrial robots and PLCs.
     

Jason McDaniel has been an automation engineer with Olympus Controls since 1998.  He has a BSME from Washington State University and an MBA from Portland State University.  Bruce Hagerty has been an automation engineer with Olympus Controls since 2004 and has a BSME from Lehigh University.  Paul Woodhouse has been an automation specialist (motion control, vision, and robotics integration) with Olympus Controls since 2002.

At the next meeting we will elect new officers for this chapter, including president.  We invite all interested members to become involved in the leadership of this organization.

Free and open to the public.  Please register HERE.

For further information please contact Nelson Bridwell at (503) 740-5102 or Bridwell@MobileRobot.org

 

OHSU/OGI Wilson Clark Center

20000 NW Walker Road
Beaverton
, Oregon

 

 

Previous Meetings

 

Thursday February 21st, 2008 at 7 PM

Practical Challenges for Machine Vision Deployment in Industrial Automation
 

 

Jerry Entrikin
Applications Engineering Manager
ATS Systems Oregon Inc.

 

 

This presentation will explore some of the diverse machine vision applications encountered at ATS Systems Oregon, a leading custom automation company.  Application of lighting, optics, cameras, software and motion can often be used
to solve machine vision applications that aren’t covered by “out of the box” systems.
 

Outline:

ATS Systems Oregon Inc. Overview
Machine Vision Application and Challenges

Camera Calibrations
Focusing on Transparent Surfaces
Looking at Cylinders
Alignments, Peaks and Metrology
Image Noise Reduction for Defect Detection

Questions
Senior vision engineer position at ATS in Corvallis.

Jerry graduated from Reed College and Columbia University in 1983 with degrees in physics and electrical engineering.  He is a 20 year veteran of the industrial automation industry with interests in high precision machine vision guided automation.

Free and open to the public.  Please register here.

For further information please contact Nelson Bridwell at (503) 740-5102 or Bridwell@MirageRobotics.Com

 

OHSU/OGI Wilson Clark Center

20000 NW Walker Road
Beaverton
, Oregon

 

Thursday January 17th, 2008 at 7 PM

Robotic Planetary Exploration

and Sample Return

 

 

Nelson Bridwell

ATS Systems Oregon

 

A team from the Portland chapter of the IEEE Robotics and Automation Society will be competing in the ICRA Robot Challenge, a simulated autonomous planetary exploration and sample return mission, at the 2008 ICRA International Conference on Robots and Automation in Pasadena. This will be a serious university/professional level event.

At our January meeting we will discuss the design and development of our robot, Lurch, a CoroBot that will be equipped and programmed to utilize stereo vision, structured lighting, and texture-based segmentation of landmarks in order to perform 3D mapping, navigation, exploration, object recognition, and manipulation.

 

 

 

 

Nelson Bridwell is a senior vision engineer with ATS Systems in Corvallis, Oregon.

 

Free and open to the public.  Please register here.

Individuals who would like to participate in this competition are invited to immediately contact Nelson Bridwell. We are particularly interested in engineers who have worked with machine vision and 3D models.

 

For further information please contact Nelson Bridwell at (503) 740-5102 or Bridwell@MirageRobotics.Com

 

OHSU/OGI Wilson Clark Center

20000 NW Walker Road
Hillsboro, Oregon

 

Wednesday December 5th, 2007 at 7 PM

Microsoft Robotics Studio

A Technical Introduction

 

Joseph Fernando

 

Microsoft has created a new software development kit for the robotics community – Microsoft Robotics Studio (MSRS). MSRS provides a software platform that enables the development of a wide variety of applications which can be used across a variety of hardware, applicable to a wide audience of users. This talk will provide an overview of Microsoft Robotics Studio and will outline Microsoft's motivation for robotics in the hobbyist, academic and commercial markets. The content is suited to both novices as well as advanced robotics professionals.

Microsoft Robotics Studio delivers in three areas of software and all will be described in this technical introduction:

-A scalable, extensible concurrency and distributed runtime architecture that can span a wide variety of hardware and devices. The programming interface can be used to create applications to drive robots using 8-bit or 16-bit processors (from a connected PC) as well as 32-bit systems with multi-core processors; and devices from simple touch sensors to laser distance finding devices.

-A set of tools that make programming and debugging robot applications scenarios easier. These include a high resolution visual simulation environment that integrates 3D software physics supplied by the Ageia Technologies PhysX engine. While Microsoft Robotics Studio can be used with programming languages such as those included in Microsoft Visual Studio and Microsoft Visual Studio Express, also included is a new visual programming language that enables the creation of applications using a simple drag-and-drop interface.

-A set of useful technology libraries (services) to help developers get started with writing robot applications, and tutorials which illustrate the basics of how to get started in a variety of programming languages.

 

Joseph Fernando is a Principal Program Manager and Architect at Microsoft. Joseph has over 17 years experience in the computer industry in engineering and software product development. Previously, Joseph has driven initiatives to enhance software architecture, design, and development practices at Microsoft. Further, he has architected productivity tools, has served as a Development Manager in the Digital Media Division, and has led multiple teams in Visual Studio/Development Division. He has helped ship over 16 products while at Microsoft Corporation

 

Free and open to the public.  Please register HERE

 

For further information please contact Nelson Bridwell at (503) 740-5102 or Bridwell@MirageRobotics.Com

 

OHSU/OGI Wilson Clark Center

20000 NW Walker Road
Hillsboro, Oregon

 

Thursday November 15th, 2007 at 7 PM

DARPA Urban Challenge 2007

Videos and Interviews

Nelson Bridwell

 

Nelson will present video highlights and interviews from the DARPA Urban Challenge that will take place at the Southern California Logistics Airport (formerly George Air Force base) in Victorville, California on Saturday November 3rd.  Twenty robots will launch starting at 7:00 AM, attempting to complete a challenging 60-mile course in under 6 hours. $3.5M in prizes will be awarded to the fastest and best vehicles that are able to finish while obeying traffic laws and performing maneuvers such as merging into traffic, navigating traffic circles, negotiating busy intersections, and avoiding obstacles.

 

(For those who are not able to attend the Urban Challenge or this meeting, beginning at 6:30 AM on Nov 3rd the Urban Challenge website will host a live webcast of the event.)

At this meeting we will also discuss the possible formation of a team to compete in the May 2008 ICRA Space Robotics Challenge in Pasadena, and robotics career opportunities in the Portland vicinity.

 

Nelson Bridwell is a senior vision engineer with ATS Systems in Corvallis, Oregon.

 

Free and open to the public.  Please register here.

 

For further information please contact Nelson Bridwell at (503) 740-5102 or Bridwell@MirageRobotics.Com

 

OHSU/OGI Wilson Clark Center - Room 407

20000 NW Walker Road
Hillsboro, Oregon

Tuesday March 13th, 2007 7-9 PM

 

FPOAs Meet the Challenges of Image Processing for Machine Vision

 

Ron Bell

Chief Technology Officer

Today's system requirements compel designers to develop faster ICs in less time. As a result, designers are increasingly turning towards programmable logic devices rather than custom silicon or ASICs. While FPGAs have served this role, several factors have prevented them from addressing many high-performance applications.

  • Limited clock rates - The fastest FPGAs have maximum clock rates in the range of 500 MHz, too slow to meet operating frequency and throughput requirements in high-performance applications

  • Post-layout timing issues - After place and route is completed, the maximum frequency (Fmax) for designs may drop to under 200 MHz. Considerable design expertise and effort is required to raise Fmax above this level.

  • Lengthy design iteration - The process of functional simulation, RTL synthesis and place and route extend the time required for design optimization.

These factors have led to a new category of very-high performance, programmable logic ICs such as the FPOA from MathStar, Inc. The FPOA™ (Field Programmable Object Array™) is an array of building blocks called “objects,” which are programmed at a higher abstraction level than with finer-grained FPGAs. MathStar’s Arrix FPOA provides an array of 400 objects operating at rates up to 1-GHz, along with with on-board memory, external memory controllers and high-speed I/O. The FPOA’s synchronous architecture and deterministic timing ensure 1 GHz operation, with no reduction in maximum frequency due to timing problems.  The result is higher throughput than what is available in other reprogrammable architectures.


MathStar’s FPOA architecture supports very high speed image processing such as two-dimensional filtering, Sobel edge detection, flat field error correction and JPEG 2000 encoding.  When operating at 1-GHz, the FPOA can achieve superior performance levels:

·         Color space conversion               1 Giga pixel per second

·         2D convolution filter                     1 Giga pixel per second

·         Sobel edge detection                   500 Mega pixels per second

·         Flat field error correction              500 Mega pixels per second

·         JPEG2000 Encoding                   150 Mega pixels per second

 

Because the FPOA is a reprogrammable device it is well suited to machine vision applications, where both flexibility and performance are critical..

 

Machine Vision Related Press Releases by MathStar:

    MathStar in Machine Vision
   
    http://www.mathstar.com/Pdfs/Mathstar_050806_Press_release_v1.1.pdf

 

 

Since the onset of the semiconductor industry, Mr. Bell has been deeply involved with chip development and design, beginning with early research and development of cell libraries and simulations, and later (while associated with CalTech) becoming a key early developer of silicon compilation tools that are the basis of standard current day EDA tools.

 

Mr. Bell's early career was launched at Sperry/Univac where he helped develop the first integrated circuit based mainframe.  At Unisys, Mr. Bell was awarded the position of Unisys Fellow. As Vice President of the Unisys UNIX Division, he was responsible for creating a complete Unix hardware and software product line based on the Intel Processor Architecture and the Unix open standard, and later served as Chief Technical Officer and Vice President for Unisys Computer Systems Group. At LSI Logic he led the Semiconductor Product Development Group (first ASICs to support Fibre Channel) and the Consumer Markets Group (set top boxes, PlayStation I.)  Mr. Bell also served as CEO for a small company leading the development of highly parallel multi-processing solutions to address digital multi-media applications. Recently Mr. Bell served as Vice President of Engineering and Chief Technical Officer for Micro Linear Technology, developing highly integrated wireless solutions for networking and digital voice applications.  Mr. Bell joined MathStar as Chief Technology Officer in May 2003.

 

Mr. Bell served IEEE as Technical Chairman, Computer Elements Group, and is a Chairman and a perennial member of the IEEE Vail Computer Elements Workshop. Through his industry positions, he actively funded university research programs at Carnegie Mellon, Penn State and the University of Utah. He holds a B.S. in Electrical Engineering and M.S. in Computer Science from the University of Utah.

 

 

Free and open to the public.  Please register at: http://www.cpd.ogi.edu/coursespecific.asp?pam=2206

 

For further information please contact Nelson Bridwell at (503) 740-5102 or Bridwell@MirageRobotics.Com

 

Oregon Graduate Institute Wilson Clark Center

Hillsboro, Oregon

 

Monday May 1st, 2006,  7-9 PM

 

Mobile Robotics Software Competition

Team Strategy Meeting

 

 Nelson Bridwell

 

In order to allow more time for serious design, development, and testing, the date for the first Mobile Robotics Software Challenge has been rescheduled to Saturday August 12th.

At this meeting we will weight several options and finalize key decisions for the robot competition in August.

For the Visual Odometry Event we will decide which platform will be used to test the entries, how it will be driven and controlled, and how we will determine the actual trajectory of this platform in order to evaluate the accuracy of estimates from each team.  Based upon these decisions, we will immediately publish size and weight limits for each team, as well as the mechanical specifications for mounting their systems on the mobile test platform.

Nelson will bring along and demonstrate the autonomous capability of some platforms.

For the Obstacle Course event we will determine the method by which the distance traveled by each entry will be determined.

For the first competition our Portland team will focus exclusively on the Visual Odometry task and not enter the obstacle course event. Nelson will discuss sensor placement, optical flow, and strategies for making the most accurate motion determination.

At this meeting we will also decide upon the best time(s) and location(s) for weekly working meetings in order to maximize our progress in the remaining 15 weeks.

Free and open to the public. 

 

For further information please contact Nelson Bridwell at (503) 740-5102 or Bridwell@MobileRobot.Org

 

Oregon Graduate Institute Wilson Clark Center

Hillsboro, Oregon

Wednesday February 22nd, 2006,  7-9 PM

Advanced Workshop for the

Mobile Robotics Software Competition

 

Using the OpenCV Computer Vision Library

 

Sam Siciliano

 

 

Sam Siciliano will discuss and demonstrate Interactive Space, a real-time computer vision application for automatically tracking moving humans with spotlights that he developed for OMSI.  It makes use of the Intel OpenCV computer vision library to obtain high-contrast trackable points and implement optical flow.  Video capture and image display are handled with Microsoft Direct Show and Direct X.  Sam will also discuss calibration schemes to correct for fish-eye camera lens distortion and how the overall system works from a systems perspective.

 

 

Sam received his Bachelors Degree in Computer Engineering from Portland State University in 2004 and currently works as an Embedded Systems Engineer for Logic Product Development.  Before joining Logic Product Development, Sam was an exhibit software developer for the Oregon Museum of Science and Industry.

Free and open to the public. 

 

For further information please contact Nelson Bridwell at (503) 740-5102 or Bridwell@MobileRobot.Org

 

Oregon Graduate Institute Wilson Clark Center

Hillsboro, Oregon

 

Thursday February 2nd, 2006,  7-9 PM

The "Autonomous" Helicopter Project;

Nonlinear Control, Estimation, and Active Vision.

 

Eric A. Wan, Associate Professor

Adaptive Systems Lab

Department of Computer Science and Electrical Engineering
OGI School of Science and Engineering, OHSU

 

 

This talk looks back at the conclusion of OGI's 5 year Software Enabled Control (SEC) DARPA project. I will discuss what we set out to do and what was accomplished in the course of technical hurdles (and shifting project goals). In addition to the high-level overview, technical and mathematical details will be provided on core algorithms developed, including Model Predictive Neural Control, State-Dependent Riccati Equation Control, Sigma-Point Kalman filtering, and Active Vision based control. Future
directions and the impact of these developments on other application areas are discussed.

 


Dr. Wan received his BS (1987), MS (1988), and Ph.D. (1994) in Electrical Engineering from Stanford University. His research is in algorithms and architectures for machine learning and adaptive signal processing. He has led a diverse number of research projects over the years ranging from autonomous unmanned aerial vehicles, estimation and probabilistic inference, integrated navigation systems, time series prediction and modeling, and speech enhancement. Recent research has focused on developing adaptive mathematical models to characterize such biomedical processes as heart and lung sounds and fluid resuscitation for sepsis control. He holds several patents in adaptive signal processing and has authored over 60 technical papers in the field.

 

For further information please contact Nelson Bridwell at (503) 740-5102 or Bridwell@MobileRobot.Org

 

Free and open to the public.  Please register: http://www.cpd.ogi.edu/courseSpecific.asp?pam=1996

Oregon Graduate Institute Wilson Clark Center

Hillsboro, Oregon

 

Tuesday January 10th, 2006,  7-9 PM

Mobile Robotics Software Challenge

Kickoff Workshop

 

The lack of robotics algorithms that are effective in a wide variety of situations and environments continues to be the single most significant barrier to the development of successful mobile robots.  One of the most effective ways to break this software barrier is to stage mobile robot competitions that are specifically targeted at pushing the software envelope.

 

On Saturday June 24th, 2006, our chapter of the IEEE Robotics and Automation Society will host the very first Mobile Robotics Software Challenge.

 

Competition problems will favor robust, versatile solutions, rather than problem-specific throw-away code that can only work in highly-constrained environments.  Entries will not be rewarded for purely mechanical characteristics such as physical size, maximum velocity, or traction systems that are more effective over irregular terrain.  Specialized mechanical components will not be required.

 

To encourage participation, our competition will focus on problems that can be solved with inexpensive mobile robots and sensors.  Because the primary function of this competition is to advance the state of the art, this event will be open to experienced engineers as well as university-level student teams. 

 

At this kickoff workshop, we will discuss:

·        the objective and organization of our contest,

·        details for our challenge problems:

·        Sensor Odometry

·        Homing Problem

·        Scavenger Hunt

·        Find the Bomb

·        Obstacle Course

·        options for drive platforms and sensors,

·        useful software resources, such as the Open Source Computer Vision Library, and

·        future tutorials and reading groups on useful algorithms and techniques.

 

This meeting will be an excellent opportunity for all interested participants to introduce themselves and begin organizing into teams.

 

All interested parties are welcome to attend and participate in this seminar, including students.  IEEE-RAS membership is not required.

For further information please contact Nelson Bridwell at (503) 740-5102 or Bridwell@MobileRobot.Org

 

Free and open to the public.   Registration requested.
 

Oregon Graduate Institute Wilson Clark Center upstairs classroom #403

Hillsboro, Oregon

 

 

Saturday December 3rd, 2005 at 1 PM

WaveRunner: Autonomous Vehicle Design

Oregon State University entry - DARPA Grand Challenge

Vehicle will be on display and will be demonstrated.

Matt Knudson and Matthew MacClary

 

 

OregonWAVE "Runner" was the Oregon State University autonomous entry in the 2005 DARPA Grand Challenge, a $2,000,000 driverless vehicle race across 132 miles of desert trails that included tunnels, bridges, switchbacks, and cattle guards.

The base vehicle is a Mini-Baja car that has been upgraded with a more powerful engine and larger fuel tank. This 500 lb platform has been augmented with an integrated GPS/INS, a forward-looking LIDAR scanner, and internal sensors/controls for steering, throttle, gear selection, choke, and LIDAR elevation. All processing is performed on microcontrollers; 4 AVRs and a MicroBlaze.
 


To learn more about the Oregon WAVE Runner, see oregonstate.edu/grandchallenge.

Matt Knudson and Matthew MacClary, engineering students at OSU, will discuss the design, construction, and testing of both hardware and software for their vehicle, which will be on display. They will also perform a live demonstration of autonomous operation of this vehicle.

 

 

 Oregon Graduate Institute, Wilson Clark Center, Hillsboro, OR

Admission: Free    -  Register Here

 

 

Wednesday November 16th, 2005 at 6 PM

Unmanned Aerial Vehicle (UAV) Technology

 

Insitu Group’s SeaScan (Boeing ScanEagle)

Aircraft will be on display.

Jim Newell

 

 

Unmanned Aerial Vehicle (UAV) technology progressed at a slow but steady pace from the 1940s through the 1980's. Since the 1990's, advances in supporting technologies and upheavals in world affairs have vaulted the UAV to prominence in military and emerging civilian endeavors.

In this presentation Jim Newell, Director of Software Engineering for The Insitu Group, Inc., will briefly trace the history of the modern UAV and describe how one in particular, The Insitu Group's SeaScan (Boeing ScanEagle), has accelerated the pace of UAV technology to now be saving lives on a daily basis, revolutionizing the practical UAV application space, and Insitu being recognized as a Deloitte & Touche Fast 50 corporation in 2005.

 

Prior to joining Insitu, Jim has worked on strategic space-based C4ISR systems for the NRO, CIA, NSA, and NIMA, DOD Future Combat Systems, the Air Force Sensor Box Program, Strategic Defense Initiative Battle Management/C3, and at the Los Alamos National Laboratory was a member of the High Performance Computing Systems and Weapon Design Computational Physics divisions.

 

 

 Oregon Graduate Institute, Wilson Clark Center, Hillsboro, OR

Admission: Free    -  Register Here

 

Thursday October 20th, 2005 at 7 PM

DARPA Grand Challenge 2005

Videos and Interviews

Nelson Bridwell

 

At the startup meeting for the new Oregon chapter of the IEEE Robotics and Automation Society Nelson Bridwell will present video clips and interviews from the DARPA Grand Challenge $2,000,000 autonomous vehicle race across 150+ miles of desert that is scheduled to commence at 6:30 AM PST on Saturday October 8th.

At this initial chapter meeting we will also brainstorm exciting topics and speakers for future meetings.  We will also discuss the organization of a Mobile Robot Software Competition for this summer. 

 

 Oregon Graduate Institute, Wilson Clark Center, Hillsboro, OR