Moses Lake Field Notes


Scarab Field Report LRI

Date: June 1, 2008

Location: Moses Lake, WA

Ground Condition: Dry


  1. Arrived at field site.
  2. Camp was set up on the edge of the base camp perimeter.

  3. Scarab electrically checked out successfully.
  4. The wheels appeared to be loose so all bolts/screws were tightened in the wheels. Also the black Lexan in the wheels was secured for additional dust protection.
  5. Omnistar correction came up successfully.
  6. Site survey and gathered potential GPS way points.
  7. We pulled three cars out from sand traps.


Dave's Blurb (K. not W)

It's great to finally be here in Moses Lake, WA. There is a lot of sand (I know these are sand dunes). Scarab's base camp is on the edge of tent city it consists of an Enterprise Rent-a-Truck and a canopy that's on the side of the truck. After filling sand bags so that our tent doesn't blow away and completing the checkout on Scarab we got to drive around in our pickup truck over some of the dunes. Our truck was not able to get over a lot of the slopes; however we did find some nice slopes and long side hills that Scarab will love.

Some other quick observations:

Cars and sand do not get along.

Chariot and Athlete are BIG



Scarab Field Report LRI

Date: June 2, 2008

Location: Moses Lake, WA

Ground Condition: Dry


  1. Position estimator came up properly.
  2. Collected TRIDAR data using the triangulation laser.
  3. Collected data sets for calibration the velocity camera at various poses in daylight.
  4. Collected calibration data for the new wheel base sensor.
  5. Installed 2 LED light strips.
  6. Collected data sets for calibration the velocity camera at various poses at night with LEDs at 2 different intensities.
  7. Wheels stopped responding correctly.
    1. The motor controller reported a position tracking error and the CANBus failed.
    2. After restarting the Maestro we got an "error initializing" error
    3. This problem was solved with multiple power cycles of the Maestro and of the 48v power
    4. We believe that this was caused behaving low power conditions since Scarab was running on battery for the night time position estimator tests(#6).


Dave's Blurb (K. not W)

Today we get to actually start working out here (you can see the field report for that stuff). The interesting thing today was getting to talk to a lot of the locals. I got to talk to farmers about the "old air force base" where they now train Japanese pilots, as well as bikers about different techniques for getting in and out of the difficult dunes.

Some other quick observations:

The sand gets very cold at night

Lighting a camera is difficult (especially when you can't see the image)



Scarab Field Report LRI

Date: June 3, 2008

Location: Moses Lake, WA

Ground Condition: Wet


  1. Since it was raining the team separated to do miscellaneous things. Some of these include:
    1. Purchasing and tracking down supplies.
    2. Calibrating and preparing the new wheel base sensor.
    3. Preparing posters/media/web site.
    4. Installed temperature sensors for the robot using the DLP to read in the temperatures.
    5. Analyzed the motor controller fault observed the previous day. When the Maestro reported a position error and the CANBus failed all the motor axis turned off (mo=0) and did not reset.
  2. Sealed the TRIDAR with RTV along the seams.
  3. After the rain stopped Scarab went to Budweiser Hill (steep slope Southeast of the camp) to test navigation software.
    1. The autoleveling behavior failed and resulted in the linear actuators extending completely. Autoleveling was disabled so that we could proceed with testing.
    2. Scarab drove ~100m to the first waypoint.
    3. The second waypoint required that Scarab descend a steep slope, instead Scarab hugged the top edge of the hill and did not descend.
    4. These tests were on battery power.
  4. After ~two hours the motor controller failed due to a position error (similar to yesterday), the controller also reported an undervoltage error which was not observed yesterday: Scarab was able to recover by restarting scarabController on the PC104 (yesterday the maestro needed to be restarted several times).
  5. After getting the Maestro back up and running the joystick died and refused to be killed so the PC104 needed to be rebooted, after which the joystick worked: this might have been due to low power (weird things happen when Scarab is low on power).


Dave's Blurb (K. not W)

Even though it was raining today was still fun. For the first time since we've been here we were able to take scarab out to the treacherous parts of the dunes. We also got to see Chariot driving around; it's pretty interesting how it drives diagonally.... For dinner we joined a group from Chariot and went to a Mexican restaurant.

Some other quick observations were:

Go karts need more power

I still did not get my jalapeño from El Rodeo



ScarabField Report LRI

Date: June 4, 2008

Location: Moses Lake, WA

Ground Condition: Dry


  1. Ascended slopes and cross slopes at Budweiser Hill (id#1).
    1. Position error kept occurring during ascent.
      1. Scarab was only able to ascend the bottom part of the slope due to slippage.
      2. ScarabController can now handle and reset the controller after position tracking faults.
  2. Scarab went to a second area (Red Ant Pit) north of the basecamp
    1. North Slope (id#2a) Diagonal ascent, unleveled and leveled tests and descent tests. Observed strong soil with small rocks, hard packed.
    2. West Slope (id#2b) diagonal ascent both leaning into the slope and not leaning into the slope. Observed dense loose fine sand.
    3. South Slope (id#2c)(Tripod Hill) ascent and descent tests: Observed compacted sand with fine and large particles.
  3. Autoleveling is now working again after errors yesterday
  4. The generator was overloading when all three battery chargers were in use.
  5. Scarab conducted inchworming tests and errors in wheel coordination and large roll was detected.


Dave's Blurb (K. not W)

Today we started "serious" mobility tests. One of the things we learned was that our initial slope is one of the most difficult slopes here at the dunes. It consists of fine particles on a loosely packed angle of repose slope. The next site we moved to was a bit easier for Scarab to manipulate on. Now that we are handling the position tracking errors Scarab has been behaving beautifully.

Some other quick observations were:

If we don't go to a restaurant by 9pm, we'll end up at Denny's

There are a lot of small flying bugs that I can't name



Scarab Field Report LRI

Date: June 5, 2008

Location: Moses Lake, WA

Ground Condition: Wet


  1. Scarab now logs the position error in an attempt to figure out why the position error keeps occurring.
  2. Scarab started a 1km traverse and completed ~1/3 of it.
    1. TRIDAR died and did not respond to pings. It recovered when power cycled.
    2. When approaching a hill Scarab tried following the base instead of ascending.
    3. Test called off due to rain.
  3. Started data analysis of mobility data
  4. Installed new LED configuration (4 strips).
  5. Tested velocity camera in the dark with the new configuration.


Dave's Blurb (K. not W)

This morning I got to see prototype astronaut suits designed for walking in(the current suits are designed for floating). They have two different types, a soft suit and a hard suit. The soft suit is soft fabric with bearings at each joint. The hard suit has molded plastic all over, the astronauts have to climb up in a ladder and drop down into the suit. The astronauts actually prefer the hard suit since it remains more flexible when pressurized (The engineers like the hard suit since it allows them to mount heavier and more components).

Today we attempted our first long traverse. A little less than half way into the est we started getting hi winds mixed with rain - in order to protect the rover we were all holding a tarp over the rover while fighting the high winds. We also had pizza that we were able to eat under our tarp with Scarab. At night after dinner we went back to he field site to test the position estimator and we were able to see rain in the distance pouring down, as the night went on we could see the rain shifting to the North and missing us.

Some other quick observations were:

Buy tarps with the right aspect ratio

If the National Weather Service says there is 20% chance of rain and its raining outside, then it's really not raining.



Scarab Field Report LRI

Date: June 6, 2008

Location: Moses Lake, WA

Ground Condition: Dry (wet for short period in the afternoon)


  1. Scarab demonstrated mobility on multiple slopes; Straight Ascent, Descent, and Diagonal Ascent.
  2. Tested inchworming (now seems to be working)
  3. Tested the Position Estimator using new calibration numbers for vehicle height.
  4. First 1km night traverse
    1. TRIDAR failed several times
    2. SPAN failed due to a log file reaching its maximize size
    3. Success
    4. Total traverse=1.2km


Dave's Blurb (K. not W)

This morning was spent driving around various slopes and testing different ways to get up a slope. Inchworming (In technical terms: Peristaltic Motion) came together today so we now have another technique to ascent steep slopes.

This evening Scarab drove 1.2 km autonomously in the dark. Scarab started driving around 8pm and finished around 6:30am. For most of the traverse it was cold and very windy (it is good that Scarab is aerodynamic)


Some other quick observations were:

There is an art to putting a tarp over Scarab in high winds.

The sand becomes cold very quickly when the sun goes down.



Scarab Field Report LRI

Date: June 7, 2008

Location: Moses Lake, WA

Ground Condition:Dry


  1. Showed/Demonstrated Scarab for many of the local people.
  2. Mobility tests in Pamps Pit.
    1. Scarab performed various mobility tasks on 4 slopes. Each slope had a different terrain and slope.
    2. Inchworming was tested successfully.
  3. The yaw controller was also tested on slopes. The desired heading was maintained on a ~15 degree slope.


Dave's Blurb (K. not W)

After last night long night traverse (it actually finished around 6:30 this morning) we decided to sleep in a bit. Afterwards we were back on the dunes showing off Scarab to the public who seemed very interested in what we were doing. I got to talk to a person who welded components for the Apollo astronauts, farmers, wildlife specialists, fishers, the guy who ran the power lines over Moses Lake, and many others. After wards we took Scarab out to Pamps Pit where Scarab was able to diagonally ascend a 20-26 degree slope on small particles (it was beautiful) with no visible slip. After wards we headed out so that we could get our first full night of sleep since arriving at Moses Lake (and also so that we could be well rested for tomorrows 1km night traverse).


Some other quick observations were:

Dave W making a crater (remote control bulldozer...)

Chariotmade a big sand pile with its plow

A lot of bikes/ATVs

We got a (just one) Tweel!

Who is Pamp???



Scarab Field Report LRI

Date: June 8, 2008

Location: Moses Lake, WA

Ground Condition: Dry


  1. Draw bar pull tests were conducted in two locations. The max drawbar pull was 225kg.

  2. Successfully conducted an inchworming test on a steep fine particle slope. The routine was: straight ascent until refusal, and then inchworm up the hill until no forward progress was possible.
  3. We started our second 1km night traverse


Dave's Blurb (K. not W)

This morning was spent doing drawbar pull tests at two locations. Scarab dragged a sled with various weights behind it to apply the back force. We put 12 sand bags, Spencer, and Scott in the sled and Scarab was still able to pull the sled. Afterwards we showed that inchworming works in practice (and not just in theory). Scarab is about to start a 1km night traverse now so I can report on it tomorrow.


Some other quick observations were:

If you drive Scarab onto the back of the trailer it will tip...

If you drive an ATV too fast with kids on the back the Sheriff will chase you down.



Scarab Field Report LRI

Date: June 9, 2008

Location: Moses Lake, WA

Ground Condition: Dry


  1. Last night (and early this morning) Scarab completed a 1km night traverse using the TWM navagator and the SICK laser scanner.
    1. Scarab successfully reached the first 4 way points before deciding to wander into the bushes. At this point Scarab was stopped, the position estimater was reinitialized and batteries were swapped.
    2. Scarab's final location was near the chariot tent instead of near out truck, this was due to drift in the position estimator.
  2. A Lunar Tweel (front right wheel) was put on Scarab for testing:
    1. Dust got lodged in the shear band.
    2. Scarab was able to climb up the crater (in basecamp) with the Tweel
  3. Scarab performed straight and diagonal ascents in Pamps pit. The tests were performed at maximum weight.


Dave's Blurb (K. not W)

Today was an interesting day I got to talk to visitors fer several hours about Scarab (and hand out postcards). Since we just finished the night traverse and we have media day tomorrow we are all going to hit the bunks early to get some sleep. At night it became a bit of a war zone; people vs weather. We had 30+mph winds throwing sand at us. On the chariot tent they had two large buckets filled with sand holding down each corner; because of the winds the buckets were flying around in the air.


Some otherquick observations were:

If you cant see 2 meters in front of you, you might be in a sandstorm.

30+mph Winds + Rain + Tents= Fun (You need math in a field report)



Scarab Field Report LRI

Date: June 10, 2008

Location: Moses Lake, WA

Ground Condition: Dry


  1. Media Day: Scarab spent most of the day driving in and out of the basecamp crater to demonstrate mobility and to show its various kinematic attributes.

  2. Autonomy tests
    1. Scarab successfully navigated through a gap of bushes to find a safe path.
    2. Scarab was able to find a safe path and successfully descend into Pamps pit.
    3. Scarab tried to exit the pit but was rejected by the loose sand. The slope that rejected Scarab was in the safe range but had very loose material that Scarab could not ascend.
  3. Used the TRIDAR to take a full 360 degree scan inside Pamps pit. We believe that the problems with the TRIDAR halting have been solved by removing the error/status transmission line.


Dave's Blurb (K. not W)

Most of today revolved around explaining Scarab to the media. Dave W was interviewed for the NASA edge TV show with Scarab. Afterwards Scarab did a great job navigating into Pamps pit. Scarab approached the rim, followed it to a safe location and then descended the slope while changing its heading to follow the safest part of the slope.


Some other quick observations were:

There are a lot of people in Moses Lake who have worked (or their parents worked) with NASA.

We need a way for Scarab to detect loose vs compact terrain (and the full spectrum between loose and compact).



Scarab Field Report LRI

Date: June 11, 2008

Location: Moses Lake, WA

Ground Condition: Dry


  1. The morning consisted of a photoshoot with all the robots and all the people participating in the field test.
  2. Tested different scan patterns with the TRIDAR. By using different scan patterns (and not raster scans) we can drive continuously without the need to stop for each scan.

Dave's Blurb (K. not W)

Today we figured out how to operate continuously with out needing to stop for each scan. This will let us complete the 1km traverses in less time. We also went live with this website (that is probably where you are reading this from).


Some other quick observations were:

Setting up for pictures can take a long time with slow robots.

Everybody wants to be a rock when testing lasers (Spencer won).



Scarab Field Report LRI

Date: June 12, 2008

Location: Moses Lake, WA

Ground Condition: Dry


  1. VIP day
  2. Pack up day

Dave's Blurb (K. not W)

Today we got to hear what all the other groups are working on and about the lunar architecture. One interesting part was that they want to design an athlete that can split in half to allow for picking up payloads. We also spent the afternoon in Pamps Pit demonstrating mobility. The later part of the afternoon was spent packing up Scarab and loading the truck up. We also met Miss Moses Lake (autographs anyone?).

While this field test is coming to a close, we have enough data collected to keep us busy for a while.


Some other quick observations were:

If it's windy and your posters wont stand build a rock pile around the tripod legs.

If it still wont stand make the rock piles bigger.

When it still falls try picking the sign up every 2.3 seconds.