Help Proximity Bug

[XjrBuddhaX]

** EDIT: New Question Below 8th post **

Hi Guys and Gals. I've just recently started building circuits. I'm new but very willing to learn. Now i'm kind of stuck on my first. I've been building a little handheld car, which kind of acts like a bug. when you go to try and pick it up the motor triggers depending on which side you come from and jets off for a given amount of time.

I guess my question is whats the best approach for the sensors? I've tried hooking up Light Dependent Resistors and running through an OR gate(2 sensors in the front and 2 in the rear), I knew it wasn't going to work but what the heck. No luck.

This would be the input signal for a PIC. I'd rather not have 4 inputs for the PIC if you know what I mean. Any help would be great.

Thanks in advance.

Buddha

 

[kchriste]

This would be the input signal for a PIC.

You could use 3 hooded LDRs connected to the ADC inputs of a PIC. You may want to put a AGC circuit between the PIC and LDRs but if you use a PIC with a 10 or 12bit ADC you shouldn't need to. Then write some code so that the PIC only responds to quick changes in light levels and drives away from the source of motion. Turning the room lights on and off would be interesting.

 

[Sceadwian]

This is a basic problem with any kind of AI, and why most people don't realize how superior even a 'stupid' human is compared to a very fast computer and the processing power it takes to do even the simplest of things because of the incredible amount of sensory information we have available to us.

If you want to understand what I mean all you have to do is put in ear plugs and play loud music in the room to essentially eliminate hearing, and then blindfold yourself. Now sit in a chair and throw an object into the room you're in randomly, and then without getting up from the chair figure out where it is. You now know how a typical robot feels =)

You have to understand that as simple as something like that might be with any one of the senses you take for granted might be, a robot has none of them, and you have to place your own mindset into the frame of the robots in order to understand how it perceives it's environment to figure out how to do anything with it.

 

[audioguru]

I am a human, not a robot. It takes my sensors many months for me to realize that it is winter or summer outside.
A robot would know immediately.

 

[Sceadwian]

How would it know Audioguru? Most micro controllers or embeded systems or sensors don't typically have an 'iswinter' or 'issummer' refrence value to place on a reliable 'timeofyear' sensor. The real world is a whole lot more complicated than that.

 

[jpanhalt]

How about a "capaciflector" or other capasitance-based proximity detector. They can also be oriented in three axises to give a 3D sense of the area. You can search on that term. A lot of the work was sponsored by NASA. Also see: USPTO 5,539,292

My first choice, however, would be IR. You might consider a broad dispersion emitter with receivers for reflected light front, rear, sides, etc. to determine the direction from which the risk or object comes.

John

 

[XjrBuddhaX]

Thanks for all your input. I'm going to push through and see what happens with everything that people have mentioned. I'll keep you all updated on my progress. Once complete I will post schematic and code. Thanks again.

Buddha

 

[XjrBuddhaX]

Heh.. So i've come to a new obstacle. My circuit is to complex to run everything off of 3 AAA or a 9v through a 5v regulator. One side of my circuit will be fine. However when the motor is supposed to kick in, it doesn't casue it eats every bit of power running through the circuit.

So I guess my question is, how can I have a PIC trigger a different power source from the rest of the circuit, just for the motors (1.5v to 3v motors)? If so can you lead me in the right direction?

It's probably something I've passed over but whats the harm in asking.

Thanks again.

Justin

 

[blueroomelectronics]

How are you driving the motors? Transistors? PowerFET? H-Bridge?
Hope you didn't connect them to the PIC directy...

 

[audioguru]

My electric model airplane uses AAA rechargable Ni-MH cells that deliver 5A to 8A without any problem for takeoff. Alkaline cells will just curl up and die.

 

[Sceadwian]

Really? I was always under the impression that Alkaline cells were better at sourcing current than Nimh (though I have no reason for this belief)

 

[XjrBuddhaX]

Driving the motor from the PIC to a transistor, which is hooked to the motor from a different power source.

Buddha

 

[audioguru]

Really? I was always under the impression that Alkaline cells were better at sourcing current than Nimh (though I have no reason for this belief)

I vertically stretched the AAA Alkaline cell's curves so the voltages fit on the graph of Energizer's AAA Ni-MH cell. The Ni-MH performance is MUCH better than the alkaline at 600mA. Each charge lasts as long as an alkaline battery cell and the voltage stays much higher.

 

[Nigel Goodwin]

Really? I was always under the impression that Alkaline cells were better at sourcing current than Nimh (though I have no reason for this belief)

No, NiMh and NiCd can provide MUCH higher currents than alkaline or conventional batteries - to the extent that you need to be careful with them!.

 

[Sceadwian]

I think I just assumed because the cell capacity was higher the current source ability was. The voltage curve during discharge is pretty ugly too.

 

[audioguru]

I think I just assumed because the cell capacity was higher the current source ability was.

I remember old Ni-Cad AA cells with 500mAh capacity. They are still available inside new garden solar lights. My new Ni-MH AA cells have 2500mAh capacity, about the same as a new AA alkaline cell.