Houston--Concepts for a young machine

I've been planning, shopping, ordering parts, and waiting for what seems like forever. I want to show you guys what I've been working on, but it's still just a pile of assorted parts while I wait for certain important odds and ends to get it all together. But a couple of weeks ago I put together a sketchup model to get an idea of some sizing and fit and finish. Since I couldn't upload the model, instead I offer a few screenshots of what I've been working on.

A few explanations are in order. I want to build something that looks at least something like my concept of a robot. In other words, one of those mechanical pseudo-men that I grew up reading about and watching on TV and movies. I've been somewhat involved with robots longer than a lot of you have been alive. My first robot was a cardboard box I taped to the body of a large red tethered control convertible toy car. (We didn't have radio control vehicles back then. Besides, I was only 5.)

The next one was cut out of posterboard and sat on top of a toy car that backed away from table edges and obstacles it hit. It's eyes, (flashlight bulbs attatched to the cars headlights.) Lit up as it backed away.

Then there were a series of books by the likes of David L Hieserman, Gordon McComb, Martin Bradley Wienstien, and the like, that inspired projects that just never went anywhere. Oh, there are boxes in my basement that are full of assorted motorized platforms, light tracking mechanisms and things I managed to attach to various computers I've owned--(Anyone want a Vic 20 motor speed controller?) But nothing I considered truly a robot.

There are a few more advanced projects begun, but languishing. Seems there was always something else that my time and money needed to go to. And besides, it wasn't possible to build a REAL robot.

I've got nothing but respect for you guys that have attatched motors and servos to microcontrollers and put them under your control, or gotten them to behave intelligently. And I'm absolutely awed by the ones that have practical applications.

Finally, I have time and sources (thank God for the internet and ebay especially.) for all the things I need. There is a huge selection of microcontrollers, software, motors, materials and tools, and other people working on things that I consider "robots."

My dreams have bowed (somewhat) to practicality. But I believe now, that I can build a robot that will be able to recognize me, greet me by name, cogitate and act upon my spoken word, and find its way around my house with minimal input from me. And so, I offer "Houston" (House-Ton. The way they say it in Manhattan.)

Right now, Houston is a Rover 5 with a 4 channel motor controller and an Arduino Mega 2560 hanging, (almost literally) off it. It's never traveled farther than the 10 foot usb cord that tethers the Arduino to my desktop PC.


Under my desk is a mini ITX SBC, a pile of servos, and assorted hardware that will go into the finished project. On it's way is a battery pack, some sheet aluminum, voltage converters, and corrugated plastic.

On the "someday" list is a Spider controller, an Xtion Pro Live, and maybe some meccanum wheels

The pictures show what the robot will look like--more or less. I already don't like the arms and have decided (after some experimentation) that I can bend and cut some thin aluminum into the shapes I need to hold some servos in the positions I need them to be in--without spending a fortune on Lynxmotion parts.


The next stage, will be to get to the "robot base" picture. Attached to the top of the Rover 5 is a deck that supports the motor controller, voltage converters, (Not pictured-- a second one I needed JUST for the SBC.) the batteries, Arduino or Spider controller, and other electronic assemblies, which are represented by scale boxes with pictures of the devices. Above that deck is a second deck which will support the SBC and (not pictured) hard drive.



When all of that is together and (more or less) working, I'll add the supports for the arms, and cover the whole thing in a shell made of coroplast. I've been "playing around" with ROS, (which is why I have the SBC in there.) And ultimately, I hope to turn this into a ROS system.


I feel that weight wise, I may be asking a lot from a Rover 5. Ability wise, I'm stretching the limits of an Arduino, and, my battery and power supply might be a bit undersized. I'll post this as a robot project as soon as I have it all bolted together and moving untethered. I'll post the sketchup file if someone can tell me what I missed.

I agree that weight may be a

I agree that weight may be a problem. You may also have balance issues with it being so tall and having such a narrow base. 

What are you planning to make the body from? Have you considered something really light weight like corregated plastic panels? Look at some of altapowerdog’s work as an example. You can make something strong, but light weight.