Select a subgroup to view their Weekly Report.

This week could be called the "most endeavoring and most frustrating" week because it was all about finding and fixing problems for the chassis. After we finished assembling the chassis, the programmers installed their prewritten program. As we sat down awaiting some great performance from the robot, our enthusiasm shattered instantly when the wheels twitched, squeaked, and cracked as if they had a mind of their own. It turns out that the DENSO window motors navigating the turning of the wheels heated too much and shut down and the shafts connecting them were flexing (because they were not compounded by two plates). We found programming to be the major source of error, but we also suspected part of the error to be from our side. For the programming problem, the electrical team posted the problem on Chief Delphi. As for the mechanical problem, we decided the treads of the wheels were too grippy and that they were producing a lot of friction and side load to the motors causing them to overheat. The next day, we switched the treads to the ones we've used for the past few years which have less traction. While preparing the treads, we used a new tool called a tread-puncher to punch holes into the treads. We also added a crescent shaped metal piece onto the wheel so that the treads could be screwed on securely. We took a long time before we finally put the wheels back into the modules. Mr. Ing also switched the left-sided window motor to a right-sided one so that the wheels would be controlled by two right-sided ones thinking that it would give the electrical team an easier time to program. In the meantime, we took apart the kit-bot chassis, which was no longer needed, and cleaned the machine shop. We machined two spare Bosch pieces, found some scrap plates, and with them mounted the 55 lb metal block onto the chassis so its weight would be similar to that of a finished robot. In order to fix the flexing problem, we made two plates with a cap welded onto them to support the window motors. After all modifications were done, the robot was driven again; this time with easier control.

It would have been a good adjustment, except that the wheels still turned with unsynchronized speeds and alternating directions. Mr. Ing decided to gear the window motors down in order to help solve the problem mechanically. To do that, we had to replace the chain system we had earlier with a new compounded (three-layered), heavier, and more complex one. We geared down the motor by chaining it to the bigger side of the double sprocket, while the smaller side drives the rest of the chain system. In addition, we added some idler sprockets, repositioned the window motors and the transmissions boxes. The new chain system was hard to assemble and added weight to our already-heavy chassis, but helped solve the problem of slipping chains. At first we doubted the effectiveness of the new chain system, but it turned out to be brilliant considering the many problems we had.

The chassis was finally moving with smoothness and improved maneuverability with the programming adjustments. Our efforts were fully paid off as the robot whirled around in its crab mode and raced across the ramp in car mode. However, the car mode is unable to reverse. Although watching it was a little life-threatening because the driver was driving as if he was about to run people over. Seeing that it worked, we were now able to move on to the next phase of building - the kicking mechanism.

This week we mostly worked on the crab drive wiring and programming. When the chassis was close to be done, we programmed it with the code that we've been working on from our preseason research. We then calibrated the encoder to the correct angle to the setpoints for long and wide configurations which worked fairly well. However, since our crab drive system is chained on two separate sides for both power and rotating, the biggest problem we encountered is synchronizing both sides to react as fast as it can and reach the same target at about the same time. For two days, we attempted to solve the sychronizing problem both on mechanical and electrical side. Mechanically, a plate was mounted to stabilize the rotating motor. Electrically, we set all the rotating and driving speed controllers on coast instead of break which lessened the overheating window motor problem. These changes had a significant improvement in synchronizing both sides. Then we tweaked the PID settings a bit to fine tune it. We found the turning difference occasionally jumped when the target angle and the desired angle were in opposite directions. We fixed it by sending the minimium value through the PID which fixed the large overshooting problems. The crab modules are now pretty much synchronized after we tune down the proportional value (P) of the PID. While testing, we noticed that when the robot is in crab mode (as opposed to tank mode, arcade mode, etc) the reaction was somewhat different from what we had expected. With Ms. Owens' help, we found that the problem was caused by the crab modules not stay in it their previous position. We fixed it by feeding back the previous encoder value back to the modules. We also added a circular deadband to limit the joystick so that we can reverse without actually rotating the modules.

We also continued to work on the absolute encoder readings. The task was finished, but we had a hardtime calculating the total angle. We weren't able to test some of our other code due to computer problems. We started working on a new task assigned by Ms. Owens: autonomous with crab drive. We taught some of the members on our team the basics of autonomous by showing them last year's code.

This was our second to the last week to complete the animation. Our internal deadline is the same day as Valentine’s Day as well as Chinese New Year (the real deadline is the day after). It encourages us to work harder because as soon as we submit our work, we will be showered with chocolate kisses and red envelopes. Michelle started compiling the scene; two different materials were flipped to the inside of a sphere to simulate the sky and ocean. She imported the island with the palm trees, chair, table, and the umbrella. Michelle had to reapply materials because the objects lose their properties once they are being exported. Even though Sean applied materials to the train on his computer, Michelle has to reapply them again once it gets to her computer. Yueshi has been wondering if there is any way to preserve the materials when they are exported and is going to do research on that soon.

Yueshi learned how to use part of the particle system to create a blown away effect of the word “Credits”, and it took around three hours to render. In contrast, Michelle experimented with the fairy dust effect using the particle system. The product started with the word “Animators” and then each of our names rolled by individually.

One of our concerns was that our laptops will not be able to handle when everything is imported. With just the sky/ocean, the resort center and the props listed above, the laptop began slowing down a lot and froze once. Therefore, we decided to split the scene in two when animating. When we show the track/train, the whole resort-center-island will not be part of the scene. In fact, it will be deleted. That way, the computer will have an easier time animating. The island will only be shown in the end when it is needed. We thought that by doing this it would be more efficient and time saving as a result. As a result, we think it will be more efficient and time saving. We intend to start on the animation soon to assure that we will have enough time to render the whole 30 second animation.

This week has been mostly about...the VIDEO.

Writing the script. Testing out ideas. Watching ideas fail. Coming up with new ideas. Watching those fail, too. Asking for help. Asking for permission. Receiving input. Coming up with even more ideas. Getting hopeful. Testing the ideas again. Etc.

Has it been fun? Oh yes. Has it been torture? An even bigger yes. But slowly, we can see everything falling into place. All that remains is to try again. Armed with a hopeful solution to the huge subtitles issue, we feel like this may work out.

What subtitle issue? Uh, let's just say that the original subtitles were not effective. Not only were they noisy, but they were flimsy, uncontrollable, and very painful to watch. They did not unroll in sync with the words. They flapped and crumpled and sometimes even went off the viewing screen. At some point, they got stuck which was when we yelled, "CUT!" and gave up on them.

Now, though, we've come up with a new idea, one that somewhat resembles a pulley system or rollers? We're going to make use of all those abandoned antennas created by Mr. Lau's students! There will be four. Two in the front will have holes drilled in the top and bottom and have poles going into them. Then there will be two in the back with a pole in the middle. The banner paper will pass through the front poles and then around the back pole. Someone will sit behind all the antennas and roll the back pole, thus scrolling up the subtitle banner. That someone should also hopefully be able to read the words from the back, since marker on banner paper is very see-through and so the subtitles should be in sync!

And the cast has hopefully memorized their lines by now. The props are all done. The location has been set. We're almost good to go.

We have not forgotten about the essays, though. We have now finished the Executive Summary for the Chairman's essay (which was oddly harder to write than the essay itself. Huh. Go figure.). The Chairman's essay itself still needs some statistics which shall be worked on as soon as possible before the PR leader starts getting on us. As for the Woodie Flowers essay, it may be rewritten entirely at the request of the PR leader. That remains to be seen.

The website work is coming along nicely. We're just trying to fill in the information holes. There are still a couple of people who need to give us their quotes and/or profile information, but at least all the mentor biographies have been written and we have almost all of their pictures.

The scouting sheets were also made this week. Hooray! And we have also been able to brainstorm ideas for the handouts since we have a PR alternate from the mechanical team. When she's not working on mechanical stuff, she has been helping us out!

And, for the sheer fun of it, we have been compiling everyone's daily/weekly posts from what has been posted the past few weeks on Google Wave.

As for our little Picasa photo battle with the electrical team, we are losing quite terribly. Last time we checked, they had over 1,200 pictures while we only have a little over 800. Sadness.

That's it for week three. We are sleep-deprived, sick, behind on homework, and ready for week four! WHOO!

Another week has passed and we have updated all the mistakes that were on the Week One pages and all the VEX pages. The problem was that we cropped some of the pictures in order to fit it in the frame. We later found out that the frame itself can expand to the original picture size and so many of them needed to be redone. Also the Members page is almost done - all we need is a few more testimonials. Another thing that we have done was get the mentor pictures and profile icon for them for our site. Next, we made thumbnails and captions for all of the Week Two gallery pictures. We were able to post the Week Two progress by the end of this week. Our plan for next week is to finish the FAQ (frequently asked questions) page and finish Week Three gallery.