Live from RoboCup 2009 Day 4

This week Richard Allen, Physicist, NIST, will be live blogging from RoboCup 2009 covering the MEMS-scale robot league.

July 3, 2009:

Today we have scheduled the Two Millimeter Dash.

The ETH team performed their required six runs, with times in the 350 ms – 500 ms range.  These times include delay from official start time and delivery of stop signal.  On my camera, I count between 4 and 8 frames per event.  At the rate of 30 frames per second, this works out to 133 ms – 267 ms for the robot to travel this distance.  The range of values is partly due to Dominic changing the operating parameters for the later runs.  Here is one of the runs of the ETH robot.  Watch quickly or you’ll miss it.

The USNA team is still not getting their robots to move.  We will allow them time tomorrow to attempt the Two-Millimeter Dash one additional time.

The views presented here are those of the author and do not necessarily represent the views or policies of NIST.

Please be respectful when posting comments. We will post all comments without editing as quickly as possible during business hours as long as the comments are on topic and do not contain profanity, personal attacks, or promote commercial products or services.

Live from RoboCup 2009 Day 3

This week Richard Allen, Physicist, NIST, will be live blogging from RoboCup 2009 covering the MEMS-scale robot league.

July 4, 2009, Entry 4:

Today is scheduled the Two-Millimeter Dash

However, as with all research activities, there are difficulties.  The two teams are not ready:  the Naval Academy is repeatedly having troubles getting their robots to move, perhaps due to the excessive humidity in the venue and the ETH team has set up on the Slalom Drill course, rather than the sprint course.  With all participants in agreement, we choose to invert the schedule and time the obstacle course today and the Two-Millimeter Dash tomorrow.

All events take place on a soccer field provided by NIST.  We produce the playing fields, four on a 10 mm square chip, in the NIST CNST Nanofab.  The remaining steps of packaging and bonding are done in laboratories of the NIST Electronics and Electrical Engineering Laboratory and the NIST Manufacturing Engineering Laboratory.

Image 7.1

The soccer stadium, shown here,

Image 7.2

is a huge improvement over the arena used in 2007.  In 2007, an entire glove box was brought from NIST for the participants to use.  This glove box was large enough to hold a wafer prober.  While this provided the needed tool set, it was very heavy and difficult to move between locations.  So instead of a nitrogen flushed glove box, we have this small box with ports for power (sides) and nitrogen (top and bottom) and a switch to use to select which of the four fields is in use.  In addition, we have built these with safety interlocks to ensure that dangerous voltages are not present when the box is open.

When we were deciding how to participate in non-U.S. RoboCup events, we realized that the large equipment we were using simply wouldn’t suffice.  In addition to reducing the footprint of the glove box to inches, we acquired a small microscope to which three cameras can be connected.  So our entire glove box-probe station setup is reduced to this

Image 7.3

In the Slalom Drill, the robot must traverse the field in the presence of one or more obstacles.  The teams score is based on two parameters:  time and number of obstacles.  In addition to these parameters, the teams are allowed to choose how the robot is controlled—either by computer or by a human operator (teleoperation).  In the event the robot is teleoperated, there will be a penalty assessed to the score.

In discussing teleoperation, perhaps I should digress and mention that in all of the RoboCup leagues, teleoperation is either strictly forbidden.  For the Small Size league, the robots are centrally controlled by a remote computer; in the other leagues, all robots must perform autonomously.  Due to the challenges faced by a team considering participating in the Nanogram Demonstration in simply fabricating functional robots, we have chosen this hybrid approach.

Before I can capture the Slalom Drill, I face more computer woes:  The video driver – or the operating system – on our primary image collection system, is not capable of driving two monitors on which the same image is displayed.  So, trying to get the desired image on the monitor that everyone sees when they enter the hall – and from the keyboard I can only see the back of – was really a challenge.  Then the system hung – probably related to the problem mentioned yesterday – and appeared on reboot to have wiped out my primary account.  So I set up a new account to get running again, only to get a strange error message referring to a missing initialization file for a piece of hardware we aren’t using on this trip.

Meanwhile, we had gotten a video splitter and had put all displays on a single signal, so our audience got to watch me uninstalling the driver for this file and rebooting.  Talk about stress.

But everything is now working – in the meantime, the ETH team has performed their runs on the Slalom Drill and everyone is preparing to start closing the competition down for the day.

July 2, 2009, Entry 3:

The other team participating in the Nanogram Demonstration at RoboCup 2009 is from Swiss Federal Institute of Technology (ETH).  The ETH Team is comprised of researchers from the Institute of Robotics and Intelligent Systems (IRIS).  This group, led by Professor Brad Nelson, is at the forefront of MEMS robotics.  The team at RoboCup 2009 is led by Dominic Frutiger, seen here driving one of the robots.

Blog 6.1

In addition to Dominic, the other RoboCup 2009 Nanogram attendees from ETH are Bradley Kratochvil, Christos Bergeles, Franziska Ullrich, Christophe Chautems, Raphael Bitzi, Javier Alonso, Markus Wittmer, Amirehsan Sarabadani.  Here is Brad during setup earlier in the week:

Blog 6.2

As with previous RoboCup events, the spectators are allowed to come right up to the team areas.  Here, some future MEMS researchers are checking out the ETH soccer arena.

Blog 6.3

The following are links, with videos, describing the function of the ETH robots and from their nanogram soccer competitions.

July 2, 2009, Entry 2:

Given that between us and the two teams, we brought fewer than 15 computers, we have had enough computer problems!  I thought the robots were supposed to be the hard part.

Biggest problem:  The laptop that Professor Firebaugh brought from the U.S. will not boot up.  It will start to boot, but gives an error message and after the splash screen, resets.

Smaller problem #1:  No one can log onto one of the laptops that we brought from NIST.

Smaller problem #2:  The power supply on the desktop we brought from NIST had a switch to indicate whether the source is 115 V or 240 V.  I’m not sure if the new power supply has such a switch, but I don’t plan to worry about that until we return to the U.S.

Trivial problem:  This is a problem that shows up when the computer boots and definitely looks worse than it is:

Blog5.1.

July 1, 2009, Entry 1:

I missed getting an entry completed yesterday, so I am going to combine my thoughts from yesterday with those today.  As of yesterday, all of the team boxes were unpacked and moved to storage.  Since we are right at the entrance to one of the large convention halls, I think it is important that if the audience sees clutter, they see technical clutter, not just boxes.  We got new equipment for running this year’s event: No longer do we need a large probe station; we got a small microscope on which we can mount probes.  Here is a picture Mike Newman took of Jason Gorman (left), Craig McGray (front), and myself assembling the microscope:

Blog4.1

(Photo by Michael E. Newman, NIST)

…and I swear this wasn’t posed!

I am typing this over lunch with my co-worker Craig McGray and the team from the U.S. Naval Academy.  The USNA team consists of Bryan Watson, his new wife (and honorary team member), Ashley, and the team co-advisor Associate Professor Samara Firebaugh.  Prof. Firebaugh was fortunate that Bryan’s first military assignment past graduation is delayed long enough for him to be here.  The two other students who worked on RoboCup 2009 with Prof. Firebaugh, Eric Eastman and Ashley Skahan, have already started their military service and were not able to be here in Graz.

And by the way, Bryan and Ashley (who graduated last month from the University of Alabama) are spending part of their honeymoon here at RoboCup.  Talk about “taking one for the team.”   This picture shows  Ashley, Bryan, and Craig.   One of the larger soccer fields is in the background.

Blog4.2

We are in a pizza parlor across the street from the convention hall.  We all are commiserating on the issues associated with working so far from our home laboratories and with European power.  Part of the equipment the USNA team brought was 115 V/60 HZ only and did not properly work with the voltage converters that they were able to get in the U.S.  This morning I traveled with Prof. Firebaugh to a local electronics/hobby superstore where she purchased a better power transformer and several other items.

I will introduce our other team, from the ETH Zurich, in a post tomorrow.

Much progress has been made, but much needs to occur before our first event tomorrow:  The Two Millimeter Dash.

The views presented here are those of the author and do not necessarily represent the views or policies of NIST.

Please be respectful when posting comments. We will post all comments without editing as quickly as possible during business hours as long as the comments are on topic and do not contain profanity, personal attacks, or promote commercial products or services.

Live from RoboCup 2009 Day 2

This week Richard Allen, Physicist, NIST, will be live blogging from RoboCup 2009 covering the MEMS-scale robot league.

June 29, 2009:

Today the teams arrived.  Remember that picture from yesterday.  Here is what the hall looks like today:

BlogEntry3.1

The Nanogram demonstration was assigned what is nearly ideal space.  We are immediately at the entrance of one of the two halls.  The first thing that everyone who enters this hall will see is our display.  Until we have activity, this is what they will see:

BlogEntry3.2

(Photo by Michael E. Newman, NIST)

We have had many of the participants from other teams and leagues stop by and look at our sign and at our “playing fields.”

I would like to finish today with a note that I don’t expect would surprise any frequent reader of the MIG blog:  Producing a MEMS device almost certainly requires overcoming a technical challenge.  Many, if not all of these challenges are faced by the researcher attempting to produce a MEMS-scale robot.  We had this fact brought home to us over the past week, as several of our teams decided not to come to Graz, because they had not overcome these challenges in a timely fashion and thus do not have functioning robots.

Perhaps we can collectively brainstorm ideas that might help more universities overcome the challenges and achieve working MEMS-scale robots.  I look forward to seeing your comments.

The views presented here are those of the author and do not necessarily represent the views or policies of NIST.

Please be respectful when posting comments. We will post all comments without editing as quickly as possible during business hours as long as the comments are on topic and do not contain profanity, personal attacks, or promote commercial products or services.

Live from RoboCup 2009

This week Richard Allen, Physicist, NIST, will be live blogging from RoboCup 2009 covering the MEMS-scale robot league.

June 28, 2009

Maybe I was a bit premature with the end of my last blog entry:

Let the games begin!

RoboCup 2009 is being held in the Stadthalle Graz, the convention center in Graz Austria.  The competitions will be held in two large halls, with stadium seating for many of the events.  There are hundreds of tables throughout the halls for the participants to use.  But the teams don’t show up until tomorrow.  For now, the center is set up and waiting.

Stadthalle Graz in preparation for RoboCup 2009

The lonely person in the foreground is Craig McGray, my co-worker at NIST; notice the person speeding through on a bicycle in the background.  I don’t think this will be possible tomorrow.

The views presented here are those of the author and do not necessarily represent the views or policies of NIST.

Please be respectful when posting comments. We will post all comments without editing as quickly as possible during business hours as long as the comments are on topic and do not contain profanity, personal attacks, or promote commercial products or services.

Live from RoboCup 2009

This week Richard Allen, Physicist, NIST, will be live blogging from RoboCup 2009 covering the MEMS-scale robot league.

June 27, 2009

First, I would like to introduce myself to those who don’t know me.  I am Richard Allen and I am a physicist in the Semiconductor Electronics Division of the National Institute of Standards and Technology in Gaithersburg, Maryland U.S.A.  I have been at NIST since 1990.  At NIST my research has been in metrology tools for MEMS (recently) and semiconductor devices.

As I write this, I am on a train, traveling from Vienna, Austria to Graz, Austria.  For the upcoming week, I will be blogging from RoboCup 2009 for MIG.  RoboCup is an international competition devoted to the goal of achieving a team of robotic soccer players who can compete with, and defeat, the human World Cup champions.  Although this goal seems like quite a reach, the target date is 2050.  Soccer was chosen as a means for advancing robotics and artificial intelligence as it is an exciting area, with well-known and well-defined rules and accomplishing this goal will meet many, if not all, of the outstanding technical challenges faced in robotics.

Now the question might arise as to why I am here this week and why MIG asked me to blog this competition:  About three years ago, Michael Gaitan, my co-worker from NIST, proposed to the RoboCup organizers that a MEMS-scale league be developed.  The key parameter defining these MEMS-scale robots are that the largest dimension must be no larger than 300 micrometers.  I will discuss the technical issues associated with these devices over the next few days and I suspect the reader will see many areas where the technical issues associated with making these robots work map to those challenges faced by MEMS device manufactures.

Let the games begin!

The views presented here are those of the author and do not necessarily represent the views or policies of NIST.

Please be respectful when posting comments. We will post all comments without editing as quickly as possible during business hours as long as the comments are on topic and do not contain profanity, personal attacks, or promote commercial products or services.