We've made steady progress over the past few weeks.  But not today.  After improving in our second run, we took a step backwards in run 3 today. 

The morning practice session didn't go well.  We made a change in the navigation software and it refused to work at all.  So we reverted to an earlier version and started working on fixing the fix.

Before we could get a handle on things our qualification run came up (earlier than expected).  DARPA has started running with two vehicles on the course at the same time to speed the process.  We got called at about 1 pm, and the launch went quickly.  The first gate was a sign of things to come.  IRV clipped the cone and knocked down the metal panel that marked the left side of the gate.  Over the next few minutes IRV hit the left side marker for at least half of the gates.  No points for contact. 

The launch team was watching from the top of the launch trailer with a good view of the early part of the course.  There was concern of what would happen at the tunnel with what looked like a left bias. But it went ok through the hay bales leading to the tunnel and in the tunnel itself.  There looked like a left bias in position, but not enough to cause trouble.

Coming out of the tunnel (were we lose GPS) we did well getting to the next gate (tire stack at the exit of the gate), but we clipped the left cone on the gate (same as run two).

Next came the "rental car" obstacle in the squarely in the middle of the track.  We hit it squarely on run 2.  Better news this time.  Maneuvered around it without contact.  But after that, in a clear section of the track, IRV began to maneuver as iff there were obstacles in the path.  Likely that we were seeing more phantom obstacles, but that will need to wait for the data analysis.

Things got worse when we got to the next section.  The rough road was a challenge.  We ended up stuck in the hay bales on the left.  The launch team was getting ready to call for a "pause" or "disable" when IRV broke free and started moving again.  But a short time later we were stuck on left side hay bales again and this time it was serious.  IRV had climbed up on the bales and was clearly stuck.  The launch team requested a "pause" and headed for the recovery vehicle.  At that time the DARPA chase vehicle reported a fire, and there was lots of smoke clearly visible as we approached.  The hay bales had worked their way under the vehicle so far that the catalytic converter ignited them.  DARPA fire extinguishers and eventually a track fire truck were required to get it out.  We returned to the garage "on a stretcher" (the local name for the recovery truck). 

On inspection we found that there was little damage.  But we also noticed power steering fluid leaking, and found that the plastic power steering reservoir had melted (again).  The fire may have aggravated it, but we now think it's coming from steering workload.  The fan we had mounted on the hood wasn't enough to prevent excessive heat buildup.

There are now 16 teams that have completed at least one full run.  It's becoming clear that there will be 20 teams with a completed run.  That would preclude a finalist position based on points alone.  We'll need to get to the bottom of our problems soon if we're going to get into the final 20.

CLICK HERE to view more photos from Day 5.