Day Two EVA to Pooh's Corner

Before the field test I had made a detailed schedule of the two test weeks. After our autonomous robot EVA to Pooh's corner, the astronauts were going back to Pooh's corner with the robot, and do a geology EVA. This was supposed to happen on Wednesday. If we would not make it on Wednesday, we had a "reserve" day on Thursday. As it turned out we needed Thursday, because severe winds on Wednesday kept us from going outside. Thus, Thursday morning was a busy time. This was the first day the complete Mobile Agents system was put to the test. Hadn't we tested this before? Undoubtedly, many times we have tested the system in the complete configuration. We even had a week long operations readiness test (ORT) at NASA Ames, a month before our field test. Nevertheless, everybody was nervous and as we would soon find out (again), whatever works one day might not work the next. Surprised? No, no one really was surprised about that. Who has not experienced Murphy's Law at du moment suprème? In Mobile Agents we are not surprised by problems showing up at the last minute. Today, it was Boudreaux's turn. When they turned on Boudreaux its motors were not working. "It's in the base", Jeff Graham (the ERA Team Lead) told me, "remember we were having problems with 'the picks' on Tuesday", he said. "Yes", I answered. "Well, today they stopped working. There are four in series, and we don't know if the first one is broken, or all of them, or something else", Jeff said. "We need at least half an hour to determine the problem, and then another half to fix it". "Ok, I said", what could I say? "So, let's delay the start of the EVA to 12:00". Then I thought by myself, what if they don't have it fixed by noon, what should we do? Should we delay or should we go without Boudreaux? Our schedule said we would start at 11:00, so an hour delay was not that drastic. Twelve came and went. By twelve fifteen, Jeff came upstairs in the hab. "We are almost ready to go, give us 10 more minutes". "Ok", I said, but if you're not ready by twelve thirty we will have to go without Boudreaux, because we have a whole bunch of RST stuff to do after the EVA, and we are running late". Jeff understood. At 12:30pm, I yelled down stairs, "Alright astronauts, get suited up we're going. If Boudreaux is not ready after you're suited up, we'll go without him". And then … rain!!! No way, I thought, this is unbelievable, but on Mars the unbelievable will happen, although it would not be a rain delay.

At 1pm the rain had stopped and I decided that if we would not go then, we would not go that day. So, we started the EVA. As it turns out, suiting up the new InfoPack with differential GPS, and two Dell computers, as well as getting two microphones fitted in the astronaut's helmet (one for the speech dialogue system and one for the crew voiceloop) and starting the Mobile Agents system, is not fast. Finally, after an hour Ron yelled from behind his HabCom (Habitat Communicator) computer, "Ok guys, the plan is loaded in the models, we are a GO!!" (see Figure 1). At that moment Jeff walked in and said, "Ok, Boudreaux is ready, we can go!" It was 2:00pm when AstroOne (Brent) said to his personal agent, "Start Egress activity". Boudreaux was ready and the astronauts could take him along on their EVA. What an achievement of the ERA team. They resolved their mechanical problem with engineering that we are so used to at NASA. Failure is not an option and NASA engineers will do everything in their power to accomplish their task. The ERA team from NASA Johnson Space Center is no exception.

Figure 1. Day two EVA plan for Pooh's Corner
(Click for Detail)

All systems go, the AstroOne said to his personal agent, "Start walk to WayPoint12 activity". After a couple of seconds I heard the Brahms agent say to the astronaut, in a new female voice, "Walk to WayPoint12 activity started". Next, AstroOne said, "Boudreaux, follow me". Boudreaux's personal agent received the command a couple of seconds later and sent it on to the robot control software. Boudreaux locked in on AstroOne's GPS from his InfoPack and started following the astronaut at a safe distance. The EVA had begun, as far as I was concerned. Everything was working, and off they went to WayPoint12 at Pooh's Corner. It had taken Boudreaux almost four hours to get there on Tuesday during its autonomous traverse. However, today it was tracking the astronaut making it a lot speedier, because it does not have to do obstacle avoidance when it is tracking a human, as long as the human knows how not to get it in harms way. It took around thirty minutes for the astronauts to reach WayPoint12, with Boudreaux still following. I believe this 700+ meter traverse was a record distance for Boudreaux tracking an astronaut. The afternoon had started promising.

A potential problem at the beginning of the EVA now made itself known to AstroOne. As I stayed back at the hab to observe HabCom's interaction with the Mobile Agents system and with the astronauts and ERA team, I had learned early on that AstroOne's GPS agent in the Brahms model was not receiving any GPS data from the GPS system. Not explaining the intricacies of why this was happening (debugging went on by HabCom behind the EVA scene), HabCom already knew that Brent's personal agent would probably not be able to download the voice notes he would create. Indeed, as Brent finished his first voice note, Ron van Hoof (Brahms' software architect playing the role of HabCom) told me that the 'speech act' from Brent's personal agent did not come back to the HabCom's personal agent running on the HabCom computer. That meant that none of AstroOne's voice notes would be stored inside ScienceOrganzer or Compendium. However, AstroOne's dialogue system was recording his voice notes and storing them on the InfoPack computer, and we could therefore retrieve them after the EVA. All was not lost!

By now I was on my way to Pooh's Corner to observe the astronauts in executing their EVA plan and doing some real science. "Download all images", I heard over the voiceloop radio I was carrying on my way to Pooh's Corner. Silence. Then again, "Download my images" in a familiar British accent. That must be Abby (AstroTwo) downloading a collection of images from her digital camera. Then, "HabCom here. Abby's camera has suddenly been associated with 'drive E' instead of the normal 'drive D', and thus her agent cannot find the image directory. Hold on, I will try to change the drive letter here from the HabCom machine". How is that possible? Never, ever do the digital cameras associate themselves with anything but 'drive D'. What in the world is going on? Why now? HabCom, "Ok, I don't understand it, but I was able to associate it back to 'drive D', so Abby you can try it again!" Again I heard, "Download all images", and then less than a minute later, "Ok, I see the images coming in, and being downloaded to ScienceOrganizer". Great, I thought, it is working again. What a relief. And then, right after that, "We are getting an error, Tomcat died on us again and thus the images will not be stored in ScienceOrganizer!!" No, not that! What about the RST team? Now they won't get any of the science data back on Earth. They will be waiting for data and nothing will be coming. I immediately replied to Ron, "Are you on Instant Messenger?" I asked him. "No, why?" he replied. "Well, if Dan Berrios, back at Ames is on then you can ask the ScienceOrganizer team to see if they know what's going on", I said. "Good point!" I heard Ron say, "hold on … Yes, Dan is on and they are looking at the issue right now". Of course, I thought, no one of the ScienceOrganizer team is with us in the field and thus this software has to suddenly have an issue. How can Tomcat just keep crashing? This is not software that normally would crash, but today seemed to be a day that nothing normal would happen. I don't know exactly what the ScienceOrganizer team did, but somehow the problem was resolved half way through the EVA, and data did enter into the ScienceOrganizer database, and the RST was receiving e-mail (see Figure 2).

Figure 2. Science data for Day two of Pooh's Corner EVA
(Click for Detail)

After almost three hours, the batteries of the InfoPack computers were almost empty. We knew that we had to switch batteries during the EVA. The astronaut's personal agents monitor the battery level, and warn the astronauts and HabCom of low battery levels. We "hot swapped" the batteries and were good for another two to three hours. Around 5:00pm we decided to start moving back to the habitat. The two astronauts had been out for three hours. They had completed three of the four sampling activities on the plan and asked Boudreaux to take a panorama and a number of images. AstroOne walked over to Boudreaux, stood in front of his cameras and spoke again, "Boudreaux, follow me". As a trusted mule it pulled the science trailer and followed the astronauts back to the habitat. The ingress activity was started at around 5:15pm. Work was not yet done for the astronauts. After taking their suits off, and our afternoon briefing with the team, they started their sample analysis in the bottom half of the habitat. The RST had provided them with a sample methodology process in which they had to take photographs of the samples inside the hab and provide an explanation. The Mobile Agents system was shut down on their space suit, but the HabCom computer stayed operational to allow the astronauts to create more voice notes and to download pictures from their cameras, so that they can be stored and associated with other data from the EVA.

Unfortunately, when the HabCom computer was brought down from the upper deck, the system was unable to recover from a broken state. The speech dialogue system became unresponsive, and the crew could add no more science data. After dinner, we decided to use Compendium instead. Both Brent and Abby used their stored and downloaded images and voice notes to provide an analysis of the EVA (see Figure 3 and Figure 4). Finally, by midnight I was able to export the crew's EVA analysis from Compendium and upload it into ScienceOrganizer, just in time for the RST to download it and have their 5am SOWG meeting to provide feedback to the crew.

Figure 3. AstroOne's EVA analysis in Compendium
(Click for Detail)

Figure 4. AstroTwo's EVA analysis in Compendium
(Click for Detail)