Derik Thomann
Josiah Rosmarin
Nestor Hernandez
Jason Furtado
Competition went well and we were able to show off our arm which was extremely satisfying. We almost scored but got caught on the goal post... We still feel we had a good showing and enjoyed the experience. Before the competition we had quite a scare when our arm servo died luckily we were able to find one before the actual competition. It has been a crazy month but i have really enjoyed the experience.
With impounding only hours away I thought I would comment on last nights enjoyable time in 6.001
lab. Our beautiful arm broke as it attempted to attack me viciously. As we speak two team members are machining a new arm which will hopefully arrive soon for mounting. Our code has progressed quite a bit and it looks like we will make a decent showing tomorrow at least :) even after the hell of the last 24 hrs.
We have successfully taken our robot apart and put it back together and it is still in working condition! All code is written and we will be making everything work all night. Ball seeking and Goal seeking need some help and hopefully we will arrive there by morning and test until impound time.
Mock Competition 2 was less than stellar. We still had the problem with the arm shorting the board but as of now we have fixed the problem. We aquired a seperate power source for our arm servo and hopefully that will relieve that problem. Now we must deal with a new issue. When we initialize our code from the orc pad it does not initiate the camera... I sense a long night ahead.
The gyro is now working well and we have finished with the ball lining up code. Added code to wandering state to make it more robust. Obviously much work to come. The real issue right now has to do with the servo arm. Each time it runs it will cause the computer to restart because it asks for too much power. Tomorrow we will look to find a seperate power supply for the arm.
Added a gyro to turn a certain number of degrees. Made camera faster. Tomorrow we have to get connect wondering and ball searching in the FSM.
The FSM is now working and we can drive around without hitting stuff. The arm algorithm is working well but still have to connect them directly to the battery. Tomorrow's goal is wander until we see a ball, align up to it and pick it up. We'll see how we end up...
Well last night was an all nighter for Nester and I over at Deriks. Coding all day and night does a body good. We were creating a FSM infrustraction for our machine and also trying to get our arm to work. Turns out our Servos would be well served to be connected directed to the battery for power... Mock Competition today had us driving around badly using our IR sensors which were not well calibrated. Well we now have a strong base to build off of. Tonight and tomorrow will be debugging and then building off this FSM structure. As for the robot, Josiah has some ideas for improvement, so he may be having some long days AGAIN trying to make a new arm and a better ball slide. We'll im ready to clean up lab in a lil bit... Looking forward to sleep right now... Although sleeping over at Deriks, I ate better than I have in weeks. THANKS amy!!!
The arm is a beautiful thing... now to make it work is a different story. First thing is first though. Our coding infrustructure was based on potentiometers which broke today... Tonight we will attempt to come up with a control scheme to try to do something at the mock competition tomorrow. We'll see how we come of this tomorrow. One thing I can say is the robot looks pretty.
Derik could finish the arm in lab today and we are all excited to see it. Worked more on camera processing, odometry fns, calibration. We are really hoping our pots are functional soon but it looks like all the odometry information that we have coded for may be all for naught. Our potentiometers which we were using to measure how far we have traveled may not be able to work. Therefore we may have to hack something together soon.
Arm to be laserjetted tomorrow. Worked on odometry fns, calibration, PD controller and more camera processing. We are trying to write fns to use our pots to measure distance and control our robot using it.
Enjoyed the weekend.. arm is still under construction and organizing software infrastructure. We should leave planning behind soon and go into coding time soooon.
Arm work continues... hopefully it will be complete by tomorrow. Mounted some sensors on the robot. Finally know how to use CVS from a windows machine which will make coding much quicker.
Chassis Rebuilt to remove orc board from back of robot. It is now securely on top of the robot in a much safer location. Laying groundwork for future code. Writiing general Utility Functions for use in higher level control code.
Got the arm assembled, will probably need to refine. It could grab the ball better. Worked on getting saved images to be useful... will code own rgb_hsv function. More design and interface to come. Next step, basic controller with some behaviors so we can pass checkpoint 2. Actions... hmmm.
Working on design some more and arm work. Need to be able to pick up a ball by Friday to score a point.
Passed Checkpoint I and working on design of robot and how we will build off current ideas.
Finished Checkpoint I code and should be ready for checkoff tomorrow. Code is not going to be usable for the actual implementation. It is basically a hack which will have to be fixed.
Spent 4 hours starting to plan out software architecture and how we will control the robot. Drew some diagrams of how our robot control structure will be organized.
More designing of our robotic arm, planning sensor uses and locations. Beyond that we have been working on our image processing algorithms. We can now find multiple bonds as long as they don't overlap. We just need to drive toward the ball based on its location on the screen to pass Checkpoint I.
Quote of the day: "We don't have quotes of the day anymore, it's so sad." -- Amy mourning the loss of the quote of the day.
We started over and are building up classes we will use for image processing in the future. Unfortunately we have not yet been unable to reach a point where we read in pictures and find the red spot. Hopefully some work tonight will catch us up.
The chasis is coming along nicely and we are now on a solid base to drive on... when we get to that point.
"Day Three: Feature Recognition"
Mechanical team has a solid works model of the robot and will being machining soon. Software side, we had some issues getting our camera to publish images onto botclient. Turned out we just needed to reboot the bot to get the update that had been send out earlier. Therefore we had some people that had to leave so we will write the code for the bar code tonight to get checked off tomorrow.
"Day Two: Sensor Characterization"
Useful commands:
add java:
java -jar maslab.jar ip-address
current ip address: 18.62.12.186
update java on athena:
add -f java 1.5.0
scush JAVAHOME
We've decided to use an ultrasonic range finder (SRF04) as our sensor because it allows for general object detection within a wide horizontal field. We'll probably use short-range IR for more precise front-object detection. We were able to get our robot to move and stop when it came within 6 inches of a wall.
Quote of the day: "Hopefully it's not one person emitting that smell" -- Amy in reference to the sweaty sandwich smell
"Day One: Hello World with the Orc Pad"
We soldered, created a pegboard robot, wrote code to display "Hello World" on the lcd of our working display. We ate dinner.
Oh yeah... there was some covert opening of the box and syncronized flippage -- well, not so syncronized.
Hello world wasn't as easy as it should have been -- damn baud rate.
Quote of the day: "Bastard. Stop making sense." -- Derik to Jason