Chillout Zone
Let's Make Robots!
Home

Tank 1.0

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
By Kero
@ Wed, 2009-06-03 01:36
Kero's picture

Great robot, too much capabilities, avoid objects, avoid edges, light seeking, very good idea for the wheels.

 

By Edgee
@ Sun, 2009-03-29 22:14
Edgee's picture
Is awesome, it has a real sense for whats around it. 
By Trag
@ Sun, 2009-03-29 03:17
Trag's picture
  Very smooth movement, nice and controlled. Congrats on all the hand crafted components too, nice work!
By Chris the Carpenter
@ Sun, 2009-03-29 00:35
Chris the Carpenter's picture
Can we get a little more information on the Handcrafted 5x LDR array sensor?
By noise0
@ Sun, 2009-03-29 02:10
noise0's picture

there is not much to say

it consist of 5 LDR which cover 90 degrees

the structure is built with 1 mm plastic sheet

each LDR is inserted in a pipe ( drinking straws)

each single LDR is pulled down by a resistor

the back of each LDR pipe is sealed with black paint

 

By ezekiel181
@ Sun, 2009-03-29 02:50
ezekiel181's picture
Did you try to match the LDRs first? I found sometimes LDRs of the same type and even from the same batch have slightly different resistances.
By noise0
@ Sun, 2009-03-29 11:46
noise0's picture
i took a bunch of 10 ldr, and i chosed the more similar 5
By Gareth
@ Sun, 2009-03-29 11:49
Gareth's picture

Once you place them into tubes ..... you can forget matching them - the length of the tubes they are in and the geometrics of the holder array has a big influence.

Therefore its better to equalise them by software........ (by my experimentation you are only talking of a 20-40 decimal difference anyway between LDRs)

The Tubes have to be really-really DARK (including backend) (this goes also for IR LEDs) if you plan to use as directional sensor.

By noise0
@ Sun, 2009-03-29 12:09
noise0's picture

yep, software tuning is necessary to obtain perfect results

 

check this file

http://noise0.altervista.org/5photo.zip

there are results of raw test on the sensor...

can you suggest me a may to obtain a smoth equalization?

 

I have tyed this :

 

if LDR1_reading is between (0.100) then

    LDR1_reading*=0.99

else if  LDR1_reading is between (100,200) then

   LDR1_reading*=1.20

else if  LDR1_reading is between (200,300) then

   LDR1_reading*=1.11

 

and so on

 

but this way produce unwanted steps of  comutations  wich make the behaviour not smoth

 

i think th right way would be somethk like:

y=  a*x^3 + b*x^2 +c*x +d   and so on

 

but have no idea how to calculate a,b,c,d,e, 

 

 

By rik
@ Sun, 2009-03-29 12:17
rik's picture
Did you mean http://noise0.altervista.org/tank/tank1.4.zip ?
1234next ›last »