3/01/99 Monday Sorted parts in lab.
4/01/2000 Tuesday We sorted parts; verified that we had all of the parts in our kit. Discussed some potential strategies for the competition. Discussed whether we should actually test the cubes to see what kind they are, or if we should simply assume that the cubes are of some particular type based on where we pick them up. Trying to decide how aggressive we want to be/can be in the competition.
05/01/2000 Wednesday Today we decided on a basic strategy for the contest. We decided that it would take far too much time to actually sort the various blocks, and that we should therefor avoid any strategy that requires sorting. We also decided that we wish to use an aggressive strategy that involves decreasing the opponent's score. This will involve dropping professors into the enemy jail, but we do not yet know exactly what pattern we will be using. We also attended the second lecture, where we learned basic things about the various sensors. However, at this stage we do not know much about how we intend to use the different types of sensors and motors. It is obvious that the Polaroid motors will be used for the drive train and that an arrangement of four CDS cells will be used for orientation, but we don't know anything beyond that.
06/01/2000 Thursday Today we met in the morning to discuss strategies. We decided on the following basic strategy: 1)At the start of the round, drive up the steps to the central platform and, using cages that deploy to either side of the robot, grab the two professor blocks there. 2)Drop these blocks into the enemy jail 3)Come back to our side of the board along one edge of the table, grabbing the professor block as we go. 4)Drop this block onto our campus 5)Grab the two known hackers and place them in jail 6)Attack the other robot
We decided that tank treads would be needed to go up the steps, so we went into the lab to look at Romeo and Juliet and see how tank treads can be built. However, upon seeing how small the steps were, we realized that tank treads are unnecessary and a standard wheeled robot would be able to handle it. In the evening we attended our first recitation, where we learned about gear trains and bracing.
07/01/2000 Friday We attended the optional Java tutorial in the morning and in the evening we attended Lecture 3. Most of today was spent on designing gear trains. It is a far more difficult task that we had expected. We are trying the get as little friction as possible, and this is taking some time. We came up with 2 or three different designs over the course of the day, each of which was eventually rejected.
08/01/2000 Saturday Today we finally decided on a gear train design. Since we want an all-wheel drive robot, 4 trains were built and assembled in a prototype chassis. Like yesterday, most of today was occupied by the process of building gear trains. Tomorrow we will go to lab and test this design.
09/01/2000 Sunday We went into lab today and assembled our motors. It took us some time to find the correct amount of heat shrink to use, but we eventually assembled what the TA thinks are decent motors. We realized today that the spacing between professor blocks on the raised platform(18 inches) will make our original plan difficult to implement. We are considering a modification of our plan that involes bypassing the platform but still grabbing one of the blocks from it with a raised arm. Because of problems with mounting the motors on the chassis as it was built, we could not do our tests today.
10/01/2000 Monday Today we soldered and created our first battery pack. We also tested the all wheel drive and discovered it gave a lot more trouble than we expected turning. We used the 4 thin wheels and there was alot of dragging. The robot did go quickly when moving forward and backward. It was also able to go up the steps. The gear ratio 45:1 . As we expected to do alot of turning we decided to go with two motors on each of the back wheel and have the front wheel act for turning. After speaking with the TA's, we discovered our gear design was not efficient. It left alot of the axle exposed, in addition we used to long axles and the overall design was not compact. We started a redesign the gear train. As mouting the motors had been a problem in testing the first gear train, we decided to take the motors into consideration while redesigning the gear train.
11/01/2000 Tuesday. Continued working on the new gear train. After speaking with the Ta in lab, decided to have the two front wheels controlled by a servo each with two motors each on the back wheel. In the afternoon we started and finished the problem set which was handed out on Friday 07/01/2000 . In the evening we attended recitation. At recitation we learned the importance of gravity and the relevance of walking in a straight line to feedback.
We also briefly covered line following and wall following. After recitation we discussed our strategy with the recitation instructors and out TA. They all suggested that before we try to give our opponents negative points we should first try to score positive points for ourselves. Following this advice we yet again changed our strategy to first get hackers before attempting anything else.
We also continued to work on the gear trains.
12/01/2000 Wednesday. Finally finished the gear trains. They both spin and backdrive well. We have not tested them with motors as yet. We decide to go with one front wheel instead of two ( a bicycle design). This front wheel will be controlled by a servo. We are still deciding on a name for a robot. So far the tentative name is Tracy Lord. This name is a combination of names of girls Shastri has dated, been in loved with, asked to marry or been served restraining orders by.
We started the bracing of the front and back wheel.
13/01/2000 Thursday. Decided how to mount the servo on the front wheel. We worked on combining the back wheels and started to build a housing for the motors. As each of the back wheel had two motors, this took a long time as we had to get both motors aligned at right angles (horizontal or vertical) to the gears and we had to decide how to house the the motors. Finally at the end of the day we were able to mount the four motors. We still have to test them with the battery packs.
On a side note, today was the first time it snowned in over 300 plus days at MIT. The temperature is expected to drop to single digits tonight and with the wind it will feel like -25 degrees. I will leave lab at 12 midnight. If you do not recive any more journal entries from me .... you know why ....
14/01/2000 Friday Based upon our reciatation yesterday, we once again decide to redo our strategy. We are considering not going for the professors first but instead a hacker. This is to guarantee that we get into the second round.
The gear trains work with otut he motors, they back drive really well, but when the motors are put on, there is alot of skipping. Worked on eliminating the skipping.
15/01/2000 Saturday We took apart and rebuilt the gear train about 30 times today. There was still no success in eliminating the skiiping sound. We were very frustrated.
16/01/2000 Sunday After the frustrations of fixing the gear train, we decided to take a break from the gear trains today. We focussed on the other hardware aspects of the robot and began to think about what software we would need.
Bought second servo today for mounting the forward cage in which we plan to possess blocks. Tried to learn Java.
17/01/2000 Monday. Americans cried when Martin Luter King was shot. We cried because we had to work on 6.270 . Built a chassis to connect the two reear wheels and the front servo-held wheel. Also began construction on the cage to capture the blocks.
Unfortunately, the sqeaking sound still persists. Will try to eliminate it tomorrow.
1/18/2000 Tuesday Today we continued to work on the gear trains. We have spent a long time working on trying to deal with a squeaking noise that we hear in the train, and we now thing we have found the problem. Some of the 8-tooth gears on our motors are loose.
1/19/2000 Wednesday We replaced some of the 8-tooth gears on the gear trains, mounted the servo for our cage, and built a housing for the battery pack.
1/20/2000 Thursday Today we worked on wiring up our sensors. We discovered that the reflectance sensor packages we were given do not work, so we tried to find a way to make a reflectance sensor package out of a phototransistor/LED combination. After many hours of experimentation, we found that the if we mount the two on either side of an alligator clip, they work well.
1/21/200 Friday Today we continued to work on our sensors. Wiring these things up takes a lot longer than we had expected. We decided where we wished to mount them on our robot. We also finished building the cage today.
1/22/2000 Saturday Work has begun on the programming of the robot. We spent the day writing simple test routines to test things like turning, driving straight, deploying the cage, etc. We will be working on this for many days, I suspect, because it will take a while to find the proper parameters.
1/23/2000 Sunday Today is Anant's 19th birthday. He spent half the day in lab working on the same stuff he was doing Yesterday. The rest of the team also wrote code for performing rooutines such as line following. We also began soldering the sensors.
1/24/2000 Monday We worked on coding and testing our robot. We tried to drive straight using the shaft encoder API. Unfortunately, despite trying various speeds, the robot failed to drive (or reverse ) in a straight line. We finally gave up, adn decioded to write our own API for driving. We soldered the IR beacon, battery charger and other sensors. However, given the nature of the hardware and the software, we doubt we will use much of the sensors.
1/25/Tuesday Today was the qualifying round. We failed to qualify. We had some code ready, so we decided to load it. Unfortunately the code for Robot.java (this created our robot) falied to compile. We discovered there was a port conflict which we were unable to correct as ew had to take the robot to Round I.
1/26/Wednesday We spent all day working on qualifying and finishing our code. We qualified. After set backs with the shaft encoder not working and the API's for turning, moving and locking the servo being improperly written by the organizers, we had to re-write the API ourselves. This took longer than expected.
1/27/Thursday Contest day. We had a double win in Round II, so we qualified for the finals that night. Unfortunately, the volatage on the battery was not at the correct level and we were not allowed to change it. Consequently, the robot did not turn the correct degree resulting in our elimination.
The organizers were kind enough to let us take part in a demo round with the correct battery pack. We won that round.