I'll start with the setup.
I work at a huge arcade place called Dave & Buster's. Not sure exactly how big it is in actual square feet, but, it's easily 3/4 of a football field. Yes, it's video games, but, there's also quite a bit of mechanical engineering involved. Note the LIFE SIZE Rockem Sockem robots for instance
As you can see, it's not just computers. So all the techs have to haul around all their tools, or else they end up having to walk all the way back to the tech office to grab what they need, which ends up wasting a lot of time, so, I decided that what I needed was a toolbox that followed me around. I'm mostly following what Professor Bill is doing, but, I do need this toolbox ASAP, so what I'm doing is building the frame and box first (I haven't added the walls yet obviously)...
Now, to the modification. Bill is using a coupler to connect the motor shaft to the wheel, but, as I said, I need this up soon, so, even if I have to wheel it around myself, it'll still be an improvement to carrying bags and toolbelt. So here's my mod...
The 3/8 inch spacer has 8 equidistant holes in it. 4 tapped, and 4 not, So what I did was attach 4 screws thru the hub to the spacer on the wheel side, then 4 1 inch screws to the hub on the motor side. This way when I'm ready to install the motor, I'll just slide it in and lock it down.
I've tested it and it fits perfectly. So all I need to do now is make the walls for it, then I can use a rope to drag it around until I get the motors and circuitry finished. I kinda think that Bill has given us enough so far that I can (for now anyway, maybe, I think) use a blue tooth adapter on the Arduino to drive it. But, that's not til after I get the walls finished, and, who knows, maybe Bill will have progressed further by then and my plans will change (as most engineering things do once you actually start building it)
Oh, I also increased the drive wheel size from 4 to 6 inches to get more speed, which I doubt I'll ever need, but, just in case, so I mounted the rear wheels higher than the front wheels to level the thing out
That is freaking awesome. So often the end value of robotics projects is iffy at best. This is a genuine problem that robotics can solve.
Please keep us posted with plenty of pictures of your progress.
Thank you !
Latest progress is the skin of what I am, for lack of a better term, going to call, the "body"
I was cutting acrylic for 2 days, thanks to the wonders of cordless tools. Luckily, there were no curves involved, and I didn't wanna even start cutting until I had found a proper framing material, which I did, in vinyl siding edge parts, which I'm going to also have to modify
I probably should have worn a mask. I'm going to be smelling burnt acrylic for days
I've been thinking about your project. The environment seems ideal for setting up fiducial markers The arcade is already filled with an overwhelming amount of shapes and colors, so it would be easy/possible to convince management that you should mount some sort of fiducials around the room.
Fiducials are cool because they can be processed by a raspberry pi and simple camera. They allow you to easily gather a lot of information about where the robot is situated in the environment.
1. Each fiducial is unique and easy for the robot to process.
2. Each fiducial is the same size. By looking at the apparent size of the fiducial, the robot can determine the distance to the marker.
3. Each fiducial is the same shape. By looking at the perspective of the fiducial, the robot can determine its bearing to the maker.
As the robot moves around the environment it can determine its position to a high degree of accuracy.
Extra credit is available if you can figure out how to send the robot back to the tech office and ask whoever is around to give it a tool or part you forgot!
I wanna start off by saying that the fiducial markers are a fantastic idea !
My original idea was to have some kind of marker on the back of my belt (green triangle or something) because the idea was to have it follow ME around with the idea being that it would never be more than a few feet away from me
Now, my boss at work is aware of what I'm doing and is totally in favor of it, which helps because I'm building it AT WORK (not DURING work hours, but, still at work) however, the topic of having the thing roam freely around the place hasn't yet been broached, and I'm not sure how well that idea would be received by the management who isn't involved in the tech aspect of it. The skin is acrylic opaque white, and I plan to have it well lit up. My boss even suggested a rotating yellow light on the "head", which I think would be cool
But, wandering about on its own ? That might be a tough conversation to have (unless, of course, I can prove the viability of it)
With all that being said, since we do have to make maps of the place, and change the map every time we move a machine or get a new one, the fiduciary markers would be a boon to that task even if we don't allow complete autonomy of the bot, so Thank You for that ! And I will bring it up to my boss !
Yes, having it go back to the tech office is extra credit 🙂
Wandering around in public adds a whole bunch of safety considerations 🙁
Maybe the method used by these golf caddys?
Probably the easiest way to get 'ToolCart1' to follow you around is to wear a radio beacon. That way the robot always knows where you are. The Beacon can also be serve as an emergency shut off.
From there you could use either 1D or 2D lidar to have the robot avoid obstacles and stay a preset distance from you.
It seems similar to training my dog to walk politely at my side. As long as I had a pocket full of pieces of hotdog she would stick close to my side.
From there is was up to her to look around and avoid obstacles. She has three modes:)
1. If I am standing still she sits at my side. When I say 'OK' she is free to wander around quietly sniffing whatever seems interesting.
2. If I am walking she stays within 5-10 feet. If I say 'leave it' she avoids a interesting obstacle such as a hamburger someone might have dropped on the ground.
3. If I say 'go play' she is free to run as long as she keeps me in her sight.
The vinyl siding was only an idea. It didn't pan out until I actually found something that fit...
Then I had to think about it for a few days before I came up with the idea of rivets. See, rivets have almost no head on them, so it will make sliding the acrylic in and out easier (I didn't want to attach the acrylic directly to the frame)
So, that'll work... but, I ran into a snag. Remember, I'm building this at work, so, things get prioritized. We moved the game that used to be next to the skeeball, and now there's nothing to stop the balls from getting underneath the game itself (whoever came up with the idea of a game where 5 year old kids and drunk people THROW things couldn't have been thinking how that idea would absolutely not end well)
Yea, I'm not crawling under there
So, the story... There are times when pulling wire thru a ceiling that you might find an easier way to do it... and I did. This little guy is called a "Cable Caster" (I didn't make it, I bought it)
You just open a ceiling panel at one end of the room, and another panel at the other end, fire the dart which has a fishing line attached to it, tie on a string, tie the string to the wire, and pull the wire thru... easy peezy
But, there are also time when there's things in the way, and you don't have a clear shot, so I bought this little guy
There's a camera on the front, and an eyepiece on the remote. You just stick it in the ceiling with a string attached to it, and drive it from one end to the other... easy peezy
To me, a hacker isn't a person who breaks into computers. That's called a cracker. A hacker, to me, is somebody who, in the words of George Carlin, "Takes 2 things that have never been nailed together before, and nails them together". So, I ripped off an idea from a robot arm project...
And redesigned the fingers so that they were longer
Now all I have to do is attach the fingers and the arduino nano (and some batteries, this thing's gonna look wierd) and I'll be able to just drive the car under the game and grab the balls
It was at this point that I started thinking about the "bot in a bot" idea, but, I didn't design the toolbot in such a way that I would have the extra space at the bottom to store the (I don't know what to call it) the "fingerbot" ? "ballbot" ? I was planning to use the space at the bottom for batteries and circuitry. Maybe I'll have to rethink where I put the "brains"... Maybe if I were to add a "head", it might look like a dog
Oddly enough, the frame almost looks like that, doesn't it ?
You'd almost think I planned it that way ?
So I started skinning the toolbot (or ToolCart1)
It's looking ok, for a start. It's gonna need some detail work once it gets closer to being enclosed. I'll probably install the floor before I get too much farther so that I can figure out the placement of the batteries and electronics
Speaking of the electronics, getting the thing to move on its own has taken a step closer with my ballgrabber project. I was originally going to use an android app that I wrote using the MIT app inventor along with an arduino nano and an HC-05 or HC-06, but either of those options would have required TWO extra pieces, and I would have considered using either ES32 or an ESP8266, but, I don't currently have either one of those (although I did order a couple just a few minutes ago), so, I went with what I actually have in stock, which was a Raspberry Pi 0W. It's small, only one piece, and it gave me some practice using python to make something move. I wanted to keep it simple, so, there's only an "open" button, and a "close" button which will have to be adjusted once I find out exactly how big the balls are in relation to where the fingers touch the balls
But, hey, it works 🙂
I installed the door and hinges on the right side today...
I'm going to install a chain to stop the door from opening any further than it is right now
I know I've said it before, but, it's time to install the first floor while I still have the left side unfinished. I say first floor because the electronics are going to go at the bottom, and I want to install a floor above that to isolate the tools from the electronics, and this would be a good time to figure out the access to the electronics
The ideas are still forming... Access cubby, sliding shelf... not sure yet
Got a little distracted. I'd show up early to work only to have them ask me to work... Go figure
So here I am...
After I re-engineered the basement floor, it's in. Now, I've got a couple of 12v10A batteries. They're sealed lead acid, they're the same size as motorcycle batteries. I'm about to mount the tool floor, or, try to anyway
Nice chassis! Kind of reminds me of the mail delivery bots that were in a building in NYC I worked in.
What motors are you running it with?
I haven't put the motors or computers or anything electronic in it yet. I'm still building it
I had an idea, and a plan, and every day that I build a little more, that plan changes. The basement was supposed to be one piece, turns out that wasn't possible. The computers were supposed to be in the basement, turns out they'd be better in the back. I just added the tool floor, and we'll see what changes next
Only added one more piece of plexi. Just this one piece of plexi took 2 hours !
It's getting heavy enough that I think I'm gonna order the 4 inch wheels for the back to give it more torque
It's gonna take some... minor re-engineering, but, I don't want this thing to be laboring under stress. Not sure if I want to move the front wheels inside, or move the back wheels outside. If I move the back wheels outside, that'll put the motors outside too. Not sure I like that idea