The date is: Fri Jan 29 08:07:21 MST 1999

This week in Junior Design group five made some very basic 
decisions.  We decided to use modular sensor boards that 
would be connected to a main I/O board which would be 
interfaced with the HC11.  We hacked the IR sensors provided 
by the department. We tested the sensor to ge a rough idea 
of what voltages to expect from it.  We designed the board 
to include the amplification, the detector, the emitter and 
the 555 driver for the emitter.  Next, we found out that 
there is a component that does exactly what we had made the 
sensor to do, but for cheaper than we could build it for.  
We are thinking of just using plexiglass stubs for the 
balancing wheels on the base, but we will have to discuss 
that further.  

Next week, we plan to build one of the IR sensor boards just 
to get a feel for working with protel and the board etching 
process.  Of course we will be building it on a proto board 
first.  We will also be designing an altera chip to do the 
PWM given some input from the HC11. 


The date is: Sun Feb  7 22:44:59 MST 1999

Sorry this report is late.
This week we began building and testing some infrared 
distance sensors.  We will probably be using the gp2d12, but 
I figured it would be a valuable learning experience to 
design and build the distance sensors ourselves.  We also 
began designing an altera program to do the pwm with a 8 
possibly 9 bit input from the hc11.  We connected the motors 
to a mosfet to simulate the pulse width modulation.  This 
was done to determine the proper frequency at which to run 
the h-bridges.  
Next week we will order the gp2d12 sensors and begin 
designing the control algorithm for wall following.  We will 
be designing and building the floor and possibly the fire 
sensor.  We should also be able to connect the wheel 
assembly that we have chosen to the robot base.  We will 
attach the battery and try to get a main power switch.  We 
have ordered some sample dc to dc converters to supply the 
main boards with 5V.


The date is: Fri Feb 12 08:14:41 MST 1999

This week we began learning how to use the microsim pc-board 
design program.  We designed an amplifier for the floor 
sensor(which is just a photo transistor and a resistor).  
The amplifier is just a comparator with a adjustable voltage 
suply to the positive input and the sensor output to the 
negative.  This will give us a digital output to feed to the 
hc11.   We have a working design for the pwm in altera, it 
is adjustable for four different ranges(100%, 50%, 25%, and 
12.5%) each of which is broken up into 64 pieces which will 
be controlled by a six bit output from the hc11.  We briefly 
tested the gp2d12 sensors and decided that we will have to 
use some amplification, the gain of which will still have to 
be calculated.  I came up with a block diagram of the 
entire system and organized it onto 3 separate boards.  The 
overall design is done and we are currently working on the 
presentation.
Next week we will probably just be focussing on getting the 
presentation perfected.  We will be using the library room 
212 to practice the presentation on tuesday.  I am hoping to 
also start getting the amplification board designed in 
microsim, but somehow I don't see that happening next week. 


The date is: Fri Feb 19 08:19:25 MST 1999

This week was mostly focussed on getting the presentations 
done, but we did get a little work done other than that.  We 
built tested some of the sensor-amplifier pairs to test for 
power consumption.  We tested the motors with 6V DC input to 
simulate a 50% duty cycle, and we perfected all of our 
designs up to this point.  We were also able to get the 
third wheel on the chassis, but after some breif testing we 
found that the coat hanger was not strong enough to support 
the robot once it had been crimped.  We will be trying a 
really robust paper clip this weekend.
Next week, (actually starting this weekend) I will be 
attempting to design the PC board for the amplifiers.  It 
will have chip mounts for two frequency to voltage 
converters, five op-amps, and all of the related wires.  
Assuming that the setup for the frequency to voltage 
converters is not very in depth ie it doesn't need alot of 
tuning with resistors and capacitors, I should be able to 
start etching the board next week.  I think we will also be 
to the point of programming the altera chip to do the PWM 
and possibly even be able to design the PWM and power 
conversion board.   


The date is: Sat Mar  6 14:38:58 MST 1999

This week was extremely productive.  first thing we did was 
test the pwm program on an actual altera chip using the hc11 
clock and an o-scope to check the output.  This looked ok, 
so we built the board that connects the altera device to the 
h-bridges.  Dave and Ryan came up with the external 
circuitry for the frequency to voltage converters and 
tabulated the output voltage versus frequency.  We then 
plotted the data and found it to be extremely linear.   It 
now goes from 0 to 5V for 500Hz to 33kHz which is the 
maximum encoder frequency(for motor voltage of 12V).  This 
should enable us to directly compare desired and actual 
speed since the value being sent to the pwm board is an 8 
bit number corresponding to a 0 to 100% duty cycle.  We 
tested the PWM board and it works perfectly.

Next week we will focus on getting the fire sensor working 
properly.  Also we will be tabulating the amplified signal 
from the distance sensors versus voltage and hc11 values 
from the hc11.  Also I am hoping to get the sensor board 
built and tested.  With this done we should be able to begin 
getting the wall following algorithm working.


The date is: Fri Mar 12 16:04:15 MST 1999

This week was very productive, but we also suffered some 
major setbacks.  We got the sensor board built and we are 
just waiting for the connectors before it will be fully 
funcional.  The fire sensor is now working and ready to be 
optimized.  The switch and fuse were mounted in a piece of 
angle iron and will be mounted on the board soon.  The major 
setback was that when we connected the h-bridge board to the  
motors it blew out our one 64 cell altera chip.  We will be 
ordering another some time next week.  We are doing 
extensive testing on the h-bridges now to make sure that we 
don't kill another altera chip.  
Next week is spring break and we will be here for a few 
days.  We are going to attempt to get the h-bridge board 
working properly.  Hopefully if we can get the h-bridges 
working we will also be able to attempt wall following.  We 
will also be working on our code for doing the wall 
following and possibly the candle finding algorithm.


The date is: Fri Apr  2 22:05:25 MST 1999

Sorry I got this in late again, I really hope you get it.  
This week we got the chassis completed and we got everything 
running off of battery power.  I tested the sensor board 
with all of the new parts and with the battery and 
everything worked perfectly.  Unfortunately something is 
going wrong with the pulse width modulation.  There seems to 
be something wrong with the chip or the board.  We are 
attempting to run the chip separately to see if the problem 
is in the board.  This might be a major setback since we 
can't start adjusting the constants in the code until we can 
effectively run the wheels.
Next week we are planning to have all of the bugs worked out 
of the pulse width modulation and be able to test the 
program.  We are planning to first get the wheels going the 
same speed and then implement and test each of the sensor 
subsystems individually.  The robot is expected to be wall 
following efficiently and be able to navigate the maze by 
the end of next week.


The date is: Thu Apr  8 18:55:36 MDT 1999

This week we made the decision to scrap the pwm in altera 
and just have the hc12 do it for us.  We wrote a program 
into the flash eeprom on the hc12 to do the pulse width 
modulation.  The hc11 talks to the hc12 just like we would 
have talked to the altera chip if it had worked properly.  
We have achieved wall following capability, but we still 
have to do some adjusting of the constants.  We found out 
that the long distance fire sensor is able to determine 
wether or not the candle is in the room if it just looks in 
the room.  This should simplify our search algorithm 
considerably. 
Next week we plan to get the wall following perfected and 
start working on the code to put out the candle.  Dave and I 
will be going to connecticut for the compitition and we are 
really hoping to have a robot that will be able to compete.


The date is: Fri Apr 23 11:02:10 MDT 1999

This week was pretty exciting.  We went to the contest in 
Connecticut as you well know.  We found out that dealing 
with the variables introduced by changing the location ( and 
thus lighting conditions) had a very unfavorable effect on 
the functioning of our robot.  We made every attempt to make 
it work including coming up with a completely new candle 
search algorithm, and making some modifications to the robot 
itself.  We had a hell of a time getting back to new mexico 
including getting kicked off of the plane in providence and 
having to find another plane home.  Other than that the trip 
was extremely fun and educational.  I am definitely going to 
attempt to go next year if I can find the money.
Next week we are planning to touch up our algorithm and 
maybe start on getting the robot to go home.  We will 
probably be spending most of our time working on the 
presentation for Saturday.


The date is: Fri Apr 30 11:46:03 MDT 1999

This week we tried to get the robot to go home, but there 
seems to be some problem with the go-home algorithm because 
it rarely gets all the way home before stopping.  We also 
spent alot of time working on the presentation.  We are 
planning to practice one more time today, and after that I 
think we will be ready.  As far as making the robot go home 
we are planning to work on that tonight, but I am not really 
sure that my wife is going to let me stay very late.  I have 
been incredibly sick lately and I don't think that staying 
up late for another night is going to do me any good.

Next week, we are planning to work on the paper, and we will 
probably not do anything else with the robot until the end 
of the semester.  Everything is working fairly well and I am 
looking forward to the contest on Saturday, even though Kahn 
and Ben are going to whoop our buttocks. 


The date is: Thu May  6 08:17:35 MDT 1999

This week we really didn't do anything on the robot, mostly 
because it is working beautifully.  We have worked on 
getting the paper designed and later today I will start 
writing my part.  I am planning to write the paper using a 
format that is very similar to the way that we did the 
presentation.  We will start with the overall design that we 
were planning to implement, then explain how things worked 
and didn't work and then explain all of the components in 
detail and also explain the control from the hc11 and how it 
worked.  We will also cover how we plan to change the robot 
for next years competition.  
Next week is finals week and I will do nothing on the robot 
but turn in the paper and clean out our cabinet.  Also, 
this weekend we will be getting together to organize all of 
our parts of the paper, make any necesary modifications, and 
get the table of contents and figures made.