January 3, 2005
We mounted our wheels and the motors to the peg board.
Hello World program done. Go Team Awesome!
Orcpad and the gyro are soldered.
All the pieces are put together.
Hello World App for orc completed.
January 4, 20050
0 1 0 1
We mounted the computer, OrcPad?, and OrcBoard (temporarily) on the pegbot.
Characterized short-range IR sensor, found that the beam wasn't quite symmetric.
Wrote program to make bot stop six inches before an obstacle. We may need to calibrate things further.
January 5, 2005
Mounted the camera on the bot. Checked out its functioning with the PadCam?. Wrote a program to capture an image with the camera. Had lots of trouble getting the captured image to actually show up on BotClient.
Finally got BotClient to work after switching to a different computer. It was kind of exciting to see the camera doing its work.
January 6, 2005
Spent lots of frustrating hours trying to get our barcode detection program working properly. Still stuck, but we're calling it a day.
January 7, 2005
We coded up a preliminary Move class containing several overloaded methods. The gyro seems to be working very well for the rotation code. Breaking up the coding was definitely a good idea and saved us lots of time. Also started on putting together some shaft encoders. These will be a necessity if we wish to drive straight!
Unfortunately, we missed the checkpoint, but I'm confident we can get it working before Monday. Hopefully, we can fix the bugs in the image processing code soon. That seems to be something we have a lot of trouble with, and occasionally BotClient dies.
Things to do this weekend:
Detailed Mechanical Design - what sensors do we want on our robot, and where do we want them?
Some more detailed software design before we move on to bigger and better things.
January 10, 2005
We finished assembling the optical encoders and wrote test code, but they seem to be unresponsive.
Image processing is working. We are able to draw on images published to BotClient.
We set up CVS and committed all the code.
January 11, 2005
Checkpoint1 finally done, now currently working on detailed mechanical design. We are currently preparing for the design review and hope to start machining tomorrow following the design review presentation.
January 12, 2005
After the Design Review, we did a major re-structuring of our present code, and wrote up basic documentation for all our classes. Also spent a significant amount of time brainstorming ideas on how to implement mapping for the exploration round. Our plan is to use the barcodes as guides and establish nodes at certain places which relate to the placement of barcodes.
For tomorrow:
wandering code
build a base
filter the top line
January 13, 2005
Claudia's Birthday! Happy Birthday...
Wander code is coming along, but not completely done. We can now do a sort of zig-zag around the floor. The bottom base and upper base is cut, the wood size was 2 inches smaller than we expected, so I had to do some last minute re-arranging of my design (probably should have checked out the materials before I started making a model...d'uh).
Akua is working on a better find red ball method, it seems to be coming along fine.
January 14, 2005
Wandering code is done, so is the code for the servo motor. We just have to incorporate this together somehow to complete the checkpoint. However, Akua is gone for the weekend. That just leaves me to program... :-o err...
January 15, 2005
I went window shopping for clear plastic pipes. No luck, looks like we're either have to use white ones (grumble) or order them online.
January 16, 2005
More progress! I completed the Checkpoint 2 code. It seems to work under test circumstances, but I'm sure a lot of it will have to be re-calibrated once everything is moved off the pegboard. Contruction of our new body is going slowly, so far nothing has changed since Friday.
For tomorrow, I think most of the work will be mechanical... putting everything together, testing it, and so forth with the code. Mock Contest is Thursday, we should get started on some mapping code. That's all for now.
January 17, 2005
Blue line filtering code is done, we are currently working on putting the robot together. We put everything together to test the filtering code and forgot that we still had more holes to drill. This resulted in a subsequent dis-assembly of our robot again so that the base plates can be clamped together for drilling aligned holes.
Update: Robot has been built! Final calibration of Checkpoint 2 is needed before it can run successfully. We are calling it a day, cause it's 8:30, but we are almost there!
January 18, 2005
A picture of our assembled robot:
Checkpoint 2 finally done! The wandering code still needs to be tweeked a bit. Right now, the wandering takes way too long; the robot can barely find a ball in 3 minutes.
For tomorrow, we should revisit our imaging code to see if we can write up a better find ball method that can distinguish between two individual balls. Also, I plan to get the barcode recognition working as well. Both of these imaging functions were kind of left hanging because of the rgb to hsv issues. Now that it's fixed, it shouldn't be too difficult to finish.
Other stuff to do eventually include naming our robot and painting it (possibly with gold spray paint or something).
January 19, 2005
Lots of coding done today! I finshed the barcode reading function, Akua finshed the goal detection code, and Claudia improved on the wandering algorithm. With these functions done, we should have no problem putting it all together for tomorrow's mock contest. However, one thing that still needs improvement is the ball detection code, Akua is working on that.
January 20, 2005
The mock competition was today. Unfortunately, our robot did not do what we wanted. Basically, it got stuck and couldn't escape. We'll need to work on improving our escape code.
Finally made some progress on a smarter ball detection code. Once that's ready, it shouldn't be too much trouble to adapt it for other objects.
January 24, 2005
Getting ready for the second mock contest. We decided to spend most of our effort on the movement code to make sure that the robot mostly does what we want it to do. We want the bot to explore as much of the playing field as possible, and hopefully capture some balls (yeah, we know we were already supposed to have done that).
Added threading so the bot will stop after 3 minutes, and set things up so we can run our programs from the OrcPad?.
Jin is hard at work on the screw mechanism.
January 25, 2005
Mock Contest 2 coming up...Wish us luck
January 26, 2005
So yesterday, we finally decided to abandon our screw mechanism. After much sadness of what was already accomplished, we decided our new strategy is just going to be a simple bulldozer that gets balls and shoves them into a goal.
Today we decided our code was getting way to confusing. Thanks to Dave's help, we decided to implement a finite state machine instead and code a large switch for this. We're getting close! We tested most of the various individual methods, and they seem to be working. So once the fsm is put together, we can get everything working and test them tomorrow.
January 28, 2005
Competition Day! Today, we completely dis-assembled our robot and painted it copper. As a result, one of the motors was re-connected backwards and for the entire round, our robot just kind of spun around. We got another try later, and the robot went for a ball, but it missed by a little bit. Overall, if nothing else... our threading worked and we managed not to get negative points for running over time. We are pretty satisfied that we got to build this and code everything within one month, considering the fact that none of us have ever had much programming experience. Fun times...