TUESDAY, 22 JULY 1997, THE ATACAMA DESERT
(Llano de la Paciencia)

We got our truck back! With the help of several soldiers and their large jeep with a winch, we managed to get our truck out of the valley in which it was stuck. The procedure actually required them to move their truck into a "stuck" position, from which they could gain enough leverage to get our truck out. So after pulling our truck up to the same level both trucks were "stuck," but at an easier location. After digging out from that level, both trucks were free by mid-afternoon.

We are experiencing a water shortage in the desert. "What did you expect?" is what most of you are probably thinking. But the problem is that our source for bottled drinking water has temporarily dried up. We are managing well with limited quantities for the time being, and we do have plenty of non-potable water, but the lack of drinking water has caused some difficulty. We are going through our juice reserves much more quickly than expected, and Deepak has been unable to prepare an Indian meal for us because he needs enough water to make lots of rice. Fortunately we expect to have more water delivered very soon; we already got some in a thermos tonight, and look forward to receiving another large water-cooler bottle tomorrow.

Today we tested some new "farming" code. Part of the autonomy system is a module that allows you to specify a point in the world where you want the robot to end up. When used in the autonomy system, this module sends the robot to that point, all the time avoiding obstacles along the way by taking brief detours. For nighttime operation, however, it is not possible to use the safeguarding system since it relies upon images of the terrain. So we adapted this system for nighttime driving by eliminating the dependence on the range maps, and changed it to move from one point to another, and another, for any number of points. To test it out, we made octagons all night; we hope we don't confuse any astronauts or extraterrestrials with our 40 meter diameter stop sign in the desert. This system ran from 2140 tonight until 0820 Wednesday morning, and accumulated approximately 6.6 kilometers of travel, all of which was commanded via a computer in Pittsburgh.

We got an additional 280 meters of autonomy today, but it was hard-fought. The area in which it was started was quite hilly and obstacle-laden terrain, and Nomad had quite a hard time negotiating it since the obstacle avoidance system is still being tuned; Nomad is good at detecting obstacles, but the automatic avoidance of them is still being worked on. In fact at one point I had to intervene manually to get it out of a cul de sac, but the way the system is set up I cannot see images at the same time that the navigation system is running (because it slows things down too much). So I was driving blind trying to get out of the area, and managed to run into a three foot wall. It was at slow speed and did not appear to damage anything, but makes the point just how necessary it will be to have a robot that can automatically deal with surrounding obstacles on another planet.

Our new team members Steve and April finally got a chance to see Nomad in person during a fueling session. And while they were there we managed to sneak a few pictures from the panospheric and stereo cameras, so now they will have an idea of Nomad's view of them. And today folks got an early start on celebrating Mark Sibenac's birthday tomorrow.

Total distance today: 12.332 kilometers


Last Modified on: Sat Jul 19, 1997