A Micromouse is a robot that can get to the center of a maze autonomously. Below is a picture of our current prototype (un-assembled). You can see the diameter of the robot chassis is 5 inches and the size of the various components (microcontroller, digital sensor on left, analog sensor on right, wheels and motors).
REASON
The reason why I chose Micromouse as my senior design project is because I wanted to leave a legacy behind at UC Irvine before I graduate. I believe that having Micromouse as student project here at UC Irvine would really help EECS students gain valuable hands-on experience.
Micromouse is actually an IEEE competition that both UCLA IEEE and UCSD IEEE have as student projects. UCI IEEE does not have any student projects and so I saw the need to create one. It is important to believe that what you do has meaning which is why I chose Micromouse with the intention of doing it again for the 2013-2014 school year as an IEEE student project. I also wanted to get into robotics so I saw this an great way to get started.
TEAM
The UCI Micromouse team consists of 4 members, all seniors EECS students:
Irvin Huang - Project Lead - 4th year Electrical Engineering major
Timothy Chan - Software engineer - 4th year Electrical Engineering / Computer Science Engineering double major
Alvin Perlas - Software engineer - 4th year Computer Science Engineering major
Jainam Shah - Hardware engineer - 4th year Electrical Engineering major with a Biomedical minor
TEAMPICTURE
MINUTES
Dr. Harris told us to take 3 readings, see if they're close together (far off ones ignore)
He told us why documentation is importnat, esp when in a company. Intel sotry of floating point decimal
CYOA!
harris questioned how we will control motors. tim mentioned we will use ramp input.
tim mentioned two modes
1) mapping mode
Ian: write functions for all four directions so that when we write
-slippage is hard experimentally so ignore it for now the modify them to account for slippage
2) running mode
November 9, 2012 - Week 6 Friday
Irvin created lookup table for the Sharp IR sensors.
He also soldered 22pF capacitors onto the +Vcc and GND to reduce power supply noise.
HOW SENSORS WORK
The sensors work by triangulation: sending a pulse of light and measuring the distance based on the angle of the reflected pulse. This light is then sent through a special precision lens onto a CCD array The CCD array determines the angle and passes it to the rangefinder, which assigns an analog value.
Below is his setup.
Here is the lookup table created:
Errors encountered:
Sensors values all changing despite only one sensor connected
Beeping noises occurs when I bump the wires
Reason is because power wire is bumped, disturbing Vcc.
Meterstick is in inches but measurements are in cm
Running tape meter on side may obstruct light beam from sensor, NOT recommended
Mark values on meter
November 12, 2012 - Week 7 Monday
Irvin finished testing analog IR sensors.
January 14, 2012 - Week 2 Monday
Irvin, Alvin and Tim met to discuss functions needed (movement - turn left, turn right (x2 = u turn), ), look up table) and the look up table. Tim explained the the three sections of the look-up table: noisy, good and unreadable. He also said why we are averaging two values to get the actual distance. For example, if the robot gets an analog reading of 120, it will use the look-up table to average the distance values of 130 and 140. Tim started the look up table function which is stored as main.c in the dropbox.
January 14, 2012 - Week 2 Monday
Harris told us to write function to go 16 cm.
And another to get all sensor data.
Don't be calling motor functions all the time.
Instead use your own functions?
Higher level is getting robot to go straight after skewing.
No comments:
Post a Comment