large autonomous robot brain and drive system ideas?

So I have been looking at several different instructables and articles online and I'm really interested in trying to build a large robot of my own, similar to but updated and maybe more refined than the robot butler instructable here. Right now though I'm trying to figure out what I would run it off of (computer, micro controller etc) and how to interface the motors with this control system. Wheelchair motors seem to be a great option for the motors and wheels so what sorts of options do I have to interface that with a computer or how could I interface it with just an arduino or something similar? The robot butler instructable used phidgets and several relays to control everything but I'm wondering if there is a simpler or better way. I want to be able to control the two wheelchair motors and wheels as a base as wells as probably several servo motors for the robot's arms and potentially a webcam and some speakers. Any help anyone can give on ways to do this sort of thing or guidance on how to even begin is greatly appreciated. Thanks! 

sort by: active | newest | oldest
rickharris3 years ago

1. Start small and build skills and knowledge.

2. Plan, Design, Think - ON PAPER it will in the long run save you a lot of time, money and effort.

3. Study, in detail, what other people have done and use their experience.

A reasonable answer to your question would take many words as there are so many variables.

Your skill level in building and programming and electronics.

Your expectations as to what this robot might do or look like.

Your budget - Wheel chair motors will set you back up to £150+ each.

A wheel chair motor may draw as much as 70 amps under load, Your going to need a substantial controller for that kind of load - this is why people use relays.

For complex actions, look at the Picaxe system, Arduino, Using a laptop with an interface card, The raspberry pi, the beagle bone.

If you expect this machine to be able interact with people then your going to need some complex sensing, perhaps vision.

Start SMALL build up to where you want to be.

NOTE: There are commercial robots that do very complicated things - they have cost $1,000,000's to develop.

dstech (author)  rickharris3 years ago

Thanks for the advice. I know I really do need to build up my skills and plan everything out. I guess right now I'm just still trying to gather information and figure everything out, but hopefully I will be getting some things down on paper soon.

All good stuff - but you will have to stop repeating yourself, I say but you will have to stop repeating yourself....

kenyer3 years ago

What Rick said (only I believe more in the make-mistakes-to-learn-from method and less in the plan-everything-in-advance method). This might be a helpful place for some inspiration: http://letsmakerobots.com. It might also be an idea to see if there is a mini-sumo-robot competition in your neighborhood. That is a perfect way to cheaply start programming and building autonomous robots. Most members love to share their experiences.

dstech (author)  kenyer3 years ago

Thanks for the tips, looks like a good website.

rickharris3 years ago

1. Start small and build skills and knowledge.

2. Plan, Design, Think - ON PAPER it will in the long run save you a lot of time, money and effort.

3. Study, in detail, what other people have done and use their experience.

A reasonable answer to your question would take many words as there are so many variables.

Your skill level in building and programming and electronics.

Your expectations as to what this robot might do or look like.

Your budget - Wheel chair motors will set you back up to £150+ each.

A wheel chair motor may draw as much as 70 amps under load, Your going to need a substantial controller for that kind of load - this is why people use relays.

For complex actions, look at the Picaxe system, Arduino, Using a laptop with an interface card, The raspberry pi, the beagle bone.

If you expect this machine to be able interact with people then your going to need some complex sensing, perhaps vision.

Start SMALL build up to where you want to be.

NOTE: There are commercial robots that do very complicated things - they have cost $1,000,000's to develop.