Today a new team member arrived in the desert. Carlos Guestrin, a student visitor at Carnegie Mellon's Robotics Institute, will be with us for more than a week to demonstrate the Skyline Position Estimation system. The system works by taking a panoramic image of the surrouding skyline, and comparing that skyline to a store7 JULY 1997, THE ATACAMA DESERT (Llano de la Paciencia)d digital elevation map. It finds the best fit between the current view and all similar views in its database, and can then pinpoint your location to within about 300 meters. In its first rough test today, the system managed to locate our operations truck location to within 180 meters (two pixels), after searching a 1600 square kilometer database.
We took some time today to make a few repairs to Nomad. The front right wheel motor, which had been turned off for more than a week while we got new parts made, is now back in action! A very convincing demonstration of this was its climb over an 18 inch rock in Nomad's path; the wheel motor successfully demonstrated its ability to handle such "obstacles". We also added some relays to allow us to remotely power cycle the onboard panospheric camera and navigation computers. These will enable us to have more "hands off" operation while powering up Nomad. In the past, we have sometimes had to power cycle these computers individually (i.e., turn them off and on again).
Today was our first successful test of a second satellite link to the Universidad Catolica de Chile in Santiago, the capital of Chile, about 1000 kilometers south of the Atacama Desert. They were able to view images from the panospheric camera on Nomad in real time. Unfortunately, the test was run at night, so the images were rather dark; the phrase "jet black" comes to mind. But it bodes well for the establishment of another observation/control area for Nomad.
Just so you don't think we have things too easy here, we managed to lose a good hour of operations this afternoon due to power problems. Our generators and Uninterrupted Power Sources here at our command station are becoming quite burdened with computer monitors, satellite equipment, backup satellite equipment, overnight battery chargings, and a coffee maker. It can take quite a while to bring everything back up, when circuit breakers keep blowing, and we have to keep switching power supplies every few minutes. Eventually we do get things working again, but new problems like these keep us on our toes.
Finally, today was quite a productive run for Nomad. We managed to accumulate 8.557 kilometers of remote or autonomous operation today! And that in spite of two hours spent on repairs, and a mishap during an autonomous run, when Nomad drove out of line of sight of the hilltop station and up to a steep slope, before sliding back down and turning away from it. The extra distance today was due, once again, largely to the patterned search driving mode (affectionately known as "farming"), which accounted for 4384 meters of travel today (autonomy contributed 852 meters).