Let's Make Robots!


This will eventually become a robot project, but because my progress is so very slow, I decided to post the bits and pieces I have as a blog. That way you guys can start commenting and I get to show my pretty pictures.

The bot I'm working on will combine some of the things I allready posted as walkthroughs. Mainly it will feature the mintvelt object tracker and inter-picaxe serial communication.

The board is allmost finished. Just need to put in the lead wiring for the logic circuits. It has 3 picaxe 08M chips that each have a very specific function. The brain will be used to make decisions about which way to go, based on input from two bumper switches and from the object tracker.


The board is build on a small prototyping plate I found somewhere. I like this one because it has 7 screw terminals. Left 2 are for power and the four on the right for the motor. Basically the result  is a souped up version of the picaxe 08m motor driver board that drives Fritz! little 8 

The Brain (chip 1) is connected to the motor controller (chip 2) directly through a 330Ohm resistor. I chose to connect the supporting 08Ms on separate lines to the brain because I expect quite a lot of serial traffic between the object tracker and the brain. This way the motor controler wont have to wait each time the other two chips are communicating. The object tracker (chip 3) is on a two way serial line as described in the walkthrough. The line is held high with a 22K resistor and each chip is connected to the line with a 330 resistor.


This a bit of a rough draft. I'll try to add a full schematic of the board as it is laid out later. All the connections are on the back of the board and none of the logic/signal connections cross each other. As I started on the ground and lead wires, larger jumps had to be made so the end product is rather messy.

To save room, the serial programming jack has been replaced by 3-way headers. As the output pin (pin 0) is used for serial, motor or servo communication, all three picaxes have a jumper to disconnect the output temporarily. The output and ground pins on the programming headers can be used later to connect leds when the bot is in action. The 4 signal pins that are used to command the L293D motor driver run past a 5 pin header (ground + 4 pins) that can be used to connect 4 LEDs. No room for resistors for the LEDs on the board so they will have to be soldered somewhere near the LEDs.


I'm working on getting the object tracker smaller. It now runs off a single 08M chip which forced me to make it use only one servo. The servo turns the sensors inwards and outwards and the brain-chip will have to make the whole thing turn.

The servo has mistakenly been mounted in the middle (vertically) so the center of the servo axis is not in the middle. O well. This is just a prototype.

The object tracker will use two SHARP IR sensors (80cm range). To save space, the ears have been cut off using a knife which doesn't leave a clean result. The connector has been removed as well. I hate that thing, because it makes it impossible to mount the sensor inside a case.

I drilled tiny holes in the sides of the sensors (NOT ALL THE WAY THROUGH!) to be able to insert a tiny axel made out of a piece of paperclip. The whole thing is held together using double side tape, but thats just during prototyping and testing.


Well thats my progress so far. I hope to run a few tests in a couple of weeks. Like I said: progress is very very slow!


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Good job on the sensor set up. Looks very clean and neat. Can`t wait to see the rest of the robot come together.

I am so happy these ideas wern't forgotten. I love this seperate brain system. I have adopted it for walter, just with 40's... And the 4-eyes too! -Awesome.

Quick idea with your servo/sharps... They sell these at the model airplane shop and can get wicked tiny!


They are only about a buck or so each and are 100% slop-free...

I'll seek those out. Thanx!

I've been working on a similar project(sadly it's been shelved due to lifes little complications) but I have to say, multiproc setups are fun, and can be a headache if you get into a deadlock...

 I'm really interested to see how your comes out!

have you posted anything on your project? If so, can you send me the URLs? I'm curious to steal your ideas.

This is the board setup that I have...and as you can see, it's been a while since I last posted anything to this.  On a postive side, I have updated it since the last date I posted and had tested code for the intercom of all of the procs.  I have another post were I had tested the motor control with a short vid of the motor ramping up and down in both forward and reverse.

Since this was a part of my voodoobot, I had also done some other mods to that, added a servo and controls arms scrapped from my old rc car parts bin. The last things I was working on though was the code for the servo stearing and a ir beacon setup. The ir beacon stuff I had actually gotten to work well in testing it and was starting to work on a more complete design like a small box and board layout. 

Anyway, again, this is the proc setup. I use jumpers to set which proc I want to program on the board. Its actually kinda funny, looking at those old shots and comparing it now, it has at least 25% more added to it since then to get the pin out for other components.