We’ve been sitting around McMurdo waiting to get on a Herc to Byrd Camp since Saturday. In the meantime, our team has been conducting flight tests around the local area and working on instruments. One particular instrument that’s been acting up is the LiDAR.
LiDAR stands for Light Detection and Ranging. It’s pretty much the cooler version of laser-based altimetry, where you shine a laser at the ground from an aircraft and use the reflection time to tell you how far above the ground you are. LiDAR is fancier because it uses a pattern of laser light which sweeps over a swath of ground below the aircraft and maps what the surface looks like.
Ours has been acting up for unknown reasons recently. It’s temperamental by nature, required to be at a certain temperature and is useless when there are clouds around. But even under good conditions, it hasn’t been performing properly and occasionally shuts down unexpectedly. So on Sunday we took the opportunity to conduct a second ground test.
Let’s recap the first ground test:
1. The LiDAR rack is moved outside and put on a box to get it up off the ground. It’s carefully insulated with a heater inside.
2. It’s wired up to power and a computer, which controls when the laser is on and collects the data coming from the signal receiver. In other words, the computer is where you enter the top secret code that’s required to turn on the laser.
3. The LiDAR gets some glamour shots for the twitter feed.
4. Someone holds a target 100 m away from the laser to give it a surface to reflect on. In its normal orientation, this would just be the surface of the ice.
5. Someone else makes sure the beam is on the target by taking a picture, since everyone involved is wearing protective goggles that completely block out the laser light. (So to see the light you have to look through a camera lens — oh, optics!)
The complicating issue for Sunday’s test is that unlike the last time, the LiDAR is now mounted in the plane. It’s placed in a rack with the laser pointed down through a window in the floor.
But the system isn’t designed to map a surface that’s only a meter away, so it was necessary to deflect the beam out to a further distance. Now, there’s probably more than one way to do this, some safer than others. But in Antarctica you have to work with what you’ve got. So our tactic of choice was to use a spare mirror someone had in their room to redirect a Class 4 laser onto the flight line.
For the sake of justification, our plane was parked at the end of the flight line and nothing was going on in front of us, especially since it was a no-fly day and most people had off work. Other than our team, the only person I saw out on the flight line was someone grooming the snow around the C-130s parked a distance away. And mind you, I kept my eye on him so he wouldn’t go driving through our laser and blind himself.
And so our setup was thus:
The LiDAR I showed before pointed down onto the mirror which was propped up on an ammo box of notebooks. That deflected the beam to a white piece of plywood sitting on the lift on the back of a mat track truck. All we had to do was adjust the mirror to be in line with the target, then drive the truck out to 100m, adjusting the mirror and target as we went, and then start collecting data.
Now, I know we had more or less done this test before. But it was warmer then. The sun was behind the target. And most importantly, I stayed inside in the warmth for most of the time since I wasn’t wearing goggles. This time around was different. Others on the team had been looking for a “frustrating” task for me to do so that I could “learn something”. And with this, I think they found it.
The whole process was incredibly tedious, and not just because the laser would unexpectedly, and sometimes unbeknownst to the operator, shut off in the middle of the test. It was already hard to see with the sun’s reflection on the target and the camera screen, so we weren’t always sure if we had lost track of it or if it had shut off. Add some wind, cold and paranoia about people driving through the laser and you end up with 3 hours to drive the truck only 60 m away. At that point we called it good and began a series of tests. Then followed the data collection which typically involved us setting something up and then sitting outside in the cold for 5 minutes doing nothing and then changing the configuration and doing the same.
Of course about 3.5 hours in, the LiDAR fails again. It’s unhappy about being connected to A/C ground power instead of the in-flight aircraft power. It’s chilled from having the fuselage door open. The GPS it needs just doesn’t feel up to the task today. So, after 3.5 hours, we got 15 minutes of data.
And you know what? That 15 minutes of data showed the LiDAR is working now.