2009 ARVP AUVSI Journal

From ARVP

July 25th

• 7:25 am

We have our snacks, our books and laptops and most importantly our robot Bearacuda. To start off our trip Lewis is at the wheel. We had to leave bright and early today as we need to get to Renton, Washington by nightfall. That is about 14-15 hours away from Edmonton by car. The four of us have agreed that the driving shifts will be about 3 ½ hours maximum. First stop will be Jasper.

• 11:34 am

Alright so we just stopped for a break in Jasper. Ed is now driving with Lewis in the co-pilot seat. Man, the scenery up here is pretty wicked. I got a sweet picture of some bighorn sheep off to the side of the road. Our next pit stop is in Kamloops. Things are going pretty well so far. Everyone is having a rockin’ time! And now there is gridlock on the mountain road. Is it too late to take the I15?

• 3:34 pm

Chris is driving with Brodi as co-pilot. We are at the ‘Coke-a-holic’ highway and we are ascending through some very dry hills. We managed to pass all the camper vans holding up our progress. With Chris’ music playing, we’re really crusin’ now. 8% grade for 25km down. We were actually driving through the clouds. Our fuel economy was actually approaching infinity!

• 6:31 pm

Okay so we just stopped in Chiliwack for a quick dinner. It was raining very hard for the last little bit. The rain was heavy enough that we actually had to slow down. Glad that Brodi is driving now. We should be reaching the border in the next hour or so. Really hope that we don’t have any major problems with the crossing. We shouldn’t, but it would really suck if we had to take Bearacuda out in the rain. Even though it is waterproof, the electronics are not.

July 26th

• 7:40 am

Alright, so our first night went really well. The hotel had an amazing continental breakfast. A waffle and some yogurt is a great way to start the day. Ed is the first to driving and Chris is the Navigator/DJ. But first we have to find a gas station before we leave the Seattle area…

• 12:16 pm

Chris has taken over as driver with Lewis as navigator/DJ. We finished lunch at Panda Express, grabbed some franchise coffee and set back on Highway 5 for the long drive. So far, this part of our journey has lead us past some very familiar names. We have crossed Alberta Street, Terwilligar Blvd and around Vancouver Port. With the exception of the sudden rain we encountered, the weather has been warm, sunny and generally really nice. The atmosphere is one of brotherhood and enthusiasm as we make our way south through the green trees and gently rolling countryside. The music heard has ranged widely from the carefree celtic sounds of Great Big Sea, modern tunes like Metric, to the classic 80’s sounds of Ali-ha, all the way to the Christian gangsta rap of Jesusbeatz.

• 5:29 pm

Okay, so we were too busy enjoying Mt. Shashta and had no time to enjoy Weed. So we are now in California and boy is it hot. While we were driving we saw this huge accident where the vehicle was resting on its windshield and the front hood. No joke. So Brodi is at the wheel now and Ed is navigator/DJ.

• 9:26 pm

Lewis is driving right now. It is very dark and I can’t see what I am writing now. It is late and we have one more shift to do. Hopefully we get to the room before midnight. We drove at the height of fireworks at around 10:04 pm which was pretty cool.

July 26

• 9:07 am

Yeah, we slept in as we did not get to the room till after midnight. We were all pretty tired so we slep in till after 7 and had breakfast after 8 am. Brodi is driving right now and he made this wicked 30-point turn to get us out of the parking lot. We are trying to make our way to some beach. Ed is navigating. And Lewis has to use the bathroom again…

• 10:23 am.

So Ed has really earned his spot on this trip. The Pacific Coast highway is a beautiful highway. The winding roads, combined with the eerie fog rolling in from the sea, lent an almost supernatural feeling to the scenery. The wide range of colours present in the moss clinging to the rocks were the perfect compliment to the muted colours of the surf. The houses are all pretty cool too!

• 11:07 am

Travelling through this swirling fog gave everyone pause in the trip to contemplate the parallels of our long journey. The fog, always changing and obscuring the road ahead, hid the sudden twists and turns that we had to make. For once the Jester was devoid of words. Not only does the fog hide the road ahead, it shrouds the sheer cliff faces that followed our vehicle faithfully mile after winding mile. We four weary travelers continue our trek through the mist.

• 12:32 pm

After emerging from the flowing and rolling curves, twists and shapes of the lush mountain trail we have driven under, over and through clouds. This stretch was reminiscent of Mountain roads through the Rockies, but with a tropical feeling of colourful plants and a sheer cliff with the Pacific ocean below. It was time for a shift change so we stopped to take photos with elephant seals as they warmed themselves in the mid-day sun. We moved on a little further to have lunch at the foot of Hearst Castle, once home to William Randolf Hearst of Citizen X fame. The sky moves quickly overhead, breaking from a bright blue like the ocean to a dense and looming fog, choking out the horizon. Never threatening, yet covering the land like a warm wool blanket. We continue south on Hwy 1, with Chris behind the wheel.

• 3:08 pm

The four of us are making our way through the savannah-like scrublands of northern California. We have just received word from the flight team that they have arrived safely from their trip and are on their way to Kona Kai where we will set up for the rest of the competition. Kelsey McCloud [sic] has just contacted us to confirm that our radio interview will be aired on CBC tomorrow morning, sometime around 6 or 7 am. We will provide the audio file later for those who’ve missed it.

• 4:00 pm

Lewis is now back in the drivers seat after a quick bathroom break near Santa Barbara. We are about 2 hours outside of LA. We should be just in time to hit rush hour traffic. The land around us is quite barren, except for low lyi8ng deser shrubs. This is strange for me because of the proximity to the Pacific ocean. One would think that being this close to a large body of water would create a large amount of precipitation on the surrounding area. That is nature for you.

• 6:25 pm

So we travelled near Beverly Hills and Santa Monica. Traffic was not very bad coming into LA. Got the party started with some Jesusbeatz. It was epic. Now Ed is driving the final shift to our long awaited destination of San Diego. Communications with our flight team have been non-existant these past few hours. We shall see them tonight.

July 28th

• 2:11 pm

Okay, so we got into our hotel about 8:30 pm. We unloaded our gear and tried to go to dinner, but sadly the hotel restaurant closed at 9 pm and we had to order pizza. Our rooms are cozy and they are right next to each other with a door in between them so we can move between the rooms quite easily. The internet sucks though. Veselin has cobbled together a system by sadly my iTouch can’t connect to it. So, this morning we slept in tell 8:00 am and we headed to breakfast close to 9:30 am. We went to Denny’s. They have very little fruit and yogurt type meals and it bugged me. The omelet I had was good though. So we then went to pick up some groceries as it is doubtful that we will be able to go out for breakfast every morning. At 2:00 pm we have the orientation presentation, which I am at right now. More to follow after the presentation.

• 8:19 pm

So we went for dinner at TGIF downtown. Rory did not come with us as he was meeting up with one of his San Diego friends. We are currently trying to get Bearacuda ready for some late night pool testing. Probably will not be able to do much vision as it is now dark out, but we can check to see if the hydrophones are receiving a signal, if the camera is receiving an image etc. More of a system check to see if the hardware is working correctly. Rory cut off some of the plasti-dip around the flange on Bearacuda and it looks pretty sharp. Alright, Brodi has said it is time to move out on to testing. Later!

July 29th

Okay, so this entry is a recap of the whole day. We got up around 7:00 am and had a quick breakfast in the room. We packed up the van and sent out the first strike team of Chris, Rory and Veselin to unload the gear and see if they could get the AUV up and running for some water tests. Things didn’t go as planned in the morning as we had some trouble connecting with the wireless and just some other little nagging problems. The rest of the team showed up later and we managed to get Bearacuda into one of the small testing pools where we recalibrated the IMU, and tested the PID to make sure that we could travel straight so we could qualify. Sadly we were unable to connect to the camera, which was not in itself surprising as we have forgotten to plug it in before. This meant that we had to take it out of the water and open it up so we missed our first spot in the pool. We then pushed it up to after 11 am and we were trying to get things to work, but the camera just wasn’t working. We opened it up and saw that the power connector for the camera was severed on one of the lines. The wires seem to have been fraying and it just happened to break that one time we put it inside the hull. So we pulled the electronics rack out and we got ready to replace the plug – a relatively easy operation. The only problem was that we forgot what the polarity was on the plug. We had it written down somewhere, but we needed to close it and seal it up ASAP as our next time slot was coming up. So we gambled on which side was ground and which was power (and for those of you out there going ‘Just check it with a DMM or something, we couldn’t. The way the camera is set up does not allow us to do a conductivity test to see what ground is and what is power without first powering up the camera.) Oh, we lost on our gamble. Big time. We plugged it in and suddenly a faint smoke began to emerge from the back of the hole. Thankfully we killed the power very quickly but the damage was done. At first, we thought that we had destroyed our video capture card that connects to our embedded computer. The wire that connected to the camera signal line had melted and that was the source of the smoke. Oh, and just for the record, we did not make it to the pool.

At first we were more concerned that we had damaged the embedded computer, which is basically the brain of all of Bearacuda. Thankfully the embedded computer was fine. We had brought another video card with us so we set about trying to make a new cable. The first problem was that we did not have any connectors that were small enough to fit onto the video card. Then we discovered that we did not bring out second video card. Yeah things looked pretty bad for vision. But, our team is adaptable and actually works quite well under extreme pressure. That is where the magic happens and magic did visit our tiny little tent. We used some of our testing wire to create some individual connectors for the two pins we needed for the video. Then, we carefully and lovingly cleaned the video card. By the evening, we were able to successfully re-establish connection with the camera. Bearacuda could see again!! The downside was that we did not get to do any real testing in the actual real pool. But things are working now, and we are going to go and find some dinner now. The plan is to have an early night and get up early to get to the site early to hopefully get Bearacuda in the water and under the gate to qualify. Going to be tough getting up at 5 am tomorrow though… oh well, we are used to late nights and early mornings.

July 30th

• 3:40 pm

Okay, so we managed to leave the hotel at 6 am and were the third team at the site in the morning. We forgot to bring the power cord for the new router that we bought yesterday and the external power supply for the embedded computer so testing of the embedded computer was limited. We removed the fan from the electronics rack to free up some space for the internal cables, namely the thruster cables as they are numerous and take up quite a lot of space. Once the other team arrived (with IHOP for breakfast), testing was able t o begin in earnest. Cables were tightened on the roboteqs (another problem that has plagued the platform since day one), and they were also organised and tied together where we decided it was best to secure them. And then when we did the out of hull test, the camera still was not working. We re-remade the camera cable and then it was working properly (again). At around 10 am, we assembled Bearacuda and did an in-hull systems test. The test passed. We were happy. Bearacuda was taken to the test side of the pool and we put it in the water at 10:30. The good news is that we managed to qualify as Bearacuda flew like an arrow straight through the gate. The other good news is that we were able to use the remote control to move around to find the buoy and the proper depth that it was at. The bad news (and it was quite bad but funny later) was that our depth PID had a bit of a freak out and started to oscillate. Long story short, it pulled out boat (HMCS Prince David) underwater. The boat holds our wireless USB stick and it was submerged along with the booster for our USB cable. Yeah, that was the end of our run for the day. We brought it back and let everything dry. As I write this, the USB key and the cable work so we can communicate with the embedded computer inside of Bearacuda. The bad news is some other little problems have emerged (kill-switch is acting up). We are also trying to implement a rule in the software where Bearacuda will never go beyond a certain depth. More to follow as things develop.

July 31st

So today was the first full day of the competition. The previous few days have just been practice days for the teams to get their AUVs tested and working. Static judging also took place today. Two teams of judges will visit each time throughout the day, with out warning, and each team has to give a ten minute presentation about their current AUV. Both groups of judges combine their points and each team will be ranked at the end of the day based on the static judging as well as the technical report, website and team uniforms. Last year, the first group of judges caught as a bit unaware and as a result, we did not place as high up as we would have liked. This year was much better and combined with our excellent journal paper (if I may say so myself), our sweet uniforms and a decent website, we placed 11th overall in the judging for the day. We moved up about four or five spots from last year so are team was definitely pleased. This was very important as it allowed us to choose a better time slot for the second day of the competition tomorrow. This was important because today Bearacuda died in the water. Let me explain. As you know, we were having some problems with our kill-switch and micro-controller from yesterday. We are still having troubles with the micro-controller, but we managed to correct the kill-switch problem. The problem we have found is that the cables seem to be quite touchy when we place the electronics rack into the hull of Bearacuda. To make matters worse, the micro-controller reset button that we built in does not seem to reset the micro-controller, so every time we need to reset it and it is in the hull, we have to pull the electronics out. So before our qualifying run in the afternoon we discovered that the embedded computer could talk to the micro-controller if we turned on the power when the electronics were outside the hull and then kept them powered while they were placed inside the hull. Not a good solution in any way, but you sometimes you have to work with what you got. So we tested the electronics out of the hull, gave them the okay and placed them carefully inside the hull. We then hooked up the thrusters (no camera as the video card was officially toast and beyond repair or miracles) and they worked. Software then added a very simple program to Bearacuda to have it simply lock onto a heading, submerge to a certain depth and move forwards. At this stage, we just needed to get through the gate and we could rest a little bit easier knowing that we at least got through the gate.

So the mechanical team bolted on the backplate and we wheeled Bearacuda over to the dock. The battery for the electronics was still plugged in and running and before we put it in the water, we did another quick test to make sure everything was working and it was. Bearacuda was placed into the water and then orientated towards the gate. The diver placed the kill-switch in and Bearacuda submerged itself and set off towards the gate, straight as an arrow. Then, halfway to the gate, it rose and breached the surface (and if the AUV breaches the surface, the run is over). We plugged in our cable to check on the embedded computer and there was no response at all. Bearacuda had just died in the water for no obvious reason (batteries had been fully charged before hand and there was no leak). We ended the run and pulled Bearacuda out, all disappointed that it had not even made it through the gate. When we pulled out the electronics rack, the embedded computer had turned off. That was not good. We are going to work on it tonight in the room and hopefully get something running for tomorrow. We have one more chance to get through the gate tomorrow and they take the better of the two runs so as long as Bearacuda gets through the gate tomorrow, it would be like today never happened. Wouldn’t that be nice? At least with our decent ranking in the judging we got to choose the best possible time slot of 6:30 pm, which is the last one of the day. We lucked out big time. Could there be a change in the winds for Bearacuda and its frustrated crew? Or is that change about to become a tornado of woe? Wow, I need to get to bed. Later.

Aug. 1st

Okay, so we think we found the problem. The embedded computer was overheating, which is weird as it has a huge heat sink and built-in fan. We believe it is the cables fault… again. When the cables are jostled around, we think it cuts a bit of power to the embedded computer (don’t ask me how) and the embedded computer shuts down the least valuable system aka the fan to save power. We solved this problem however by powering the fan from the mainboard as we had installed a 12 V line for the large fan that was installed in the electronics rack, but which we took out as we needed to make some space to properly secure the cables (yeah, we saw the irony don’t worry). So we made a new cable to connect the fan to the mainboard and now the fan will always be on if the mainboard is on. We are still having some problems with the micro-controller and the reset button cable. We did learn that when the cable is plugged in, and just the cable, the reset button on the micro-controller will not work at all. Electronics can be a real pain sometimes. When we removed the cable, the micro-controller could be reset via the button. 6:30 pm came far too soon for our liking, but we did manage to get Bearacuda working in one of the small side pools and statically balanced it. We needed to plug in a new battery for the electronics so we opened up the back plate, pulled out the electronics and manually hit the resent switch. With the new battery in, and the thrusters working, we were ready to make our final attempt for the gate. But not just the gate, oh no. The team was feeling confident that Bearacuda would go through the gate (as it was tested extensively in the before mentioned small pool and it was working quite well) and so we added a little program to make it do a serpentine search pattern in the hopes of hitting the buoy. So in the end there was good news and bad news. Good news is that we made it through the gate on the first try and the bad news is that Bearacuda did not hit the buoy. We are so happy that it went through the gate. Bearacuda had one foot in Davey Jones’ Locker and we managed to bring it back and get it through the gate. What a team we had this year. Yey for us!

Aug 2nd

• 9:12 am

Okay, so today is finals and we did not make it. No worries though, as we are still happy we got through the gate. Our team is currently just hanging out at the tent, though a few of us are going to do some shopping. Not much to do until later in the evening when there is the dinner.

• 9:26 pm

So Cornell won. They completed the whole course and had a time bonus. UVic came in second which was great. We have no idea how we placed and we did not win any minor award. Now we just have to pack up and the driving team heads out tomorrow morning. Man, it is over for another year. I wonder how we placed?