Non-FRC Bot
 

So I have started to build a non-frc bot on my own time with my own money (ouch).

Figured I would design something that actually could be used in real world applications. To keep things simple I will stick with a land based robot. So first question is where could robots be used?

Military:
Non-combat and combat supportive roles. (over 1500 bots are used in middle east)

Private Sector:
Multitude of purposes including surveying, lawn care, inspection, window cleaning. . .

So with a few exceptions a robust drivetrain would be a must. So what would be the harshest environment a robot could be used?

Sand:

Good abrasive and good at getting stuck in (hey this could be next years game!)

Areas with sand typically have mounds and peaks that could stop a bot in its tracks. . .


Currently I have most of my electrical portion of my robot already together

Main controler would be a netbook.
Sub controller would be an arduino for IO.
2 Jaguars with a 2CAN
up to 6 victors
up to 6 spikes
An IFI controller from '07 if I need it
2 CIM motors
2 GPS
wireless N radio with posability of a 3g modem as well


Anywho first non-rhetorical question.

What do you think would be the best drivetrain for sand navigation?

I would think that smaller size 2 motor drive would be advantageous being its what I already have.

Keep in mind that keeping sand out of the mechanics are a good thing.

Re: Non-FRC Bot
 

"balloon" tires would probably work best for sand. Think about the wheels on a garden cart. I think you can buy them without the cart :rolleyes:



and planetary transmissions direct driving two wheels. You'd probably want it to move slowly, so lots of reduction would be nice.

You kind of need to figure out the rest of the requirements before you get too far. although it looks like this might be another one of those backwards design projects--start with the materials, then figure out a use for them. Not quite the normal way engineering works!

btw if you have water nearby, you should consider making an underwater ROV, they're also fun, and you get a whole new set of challenges to overcome. We ended up building a pretty spiffy rov last year

Re: Non-FRC Bot
 

In a place that there are no roads... take your own along...

Take a look at treads...
On flat even ground ... totally unnecessary..
but for varied terrain...all types..

Treads...they take the worry out of getting there...

They do take more maintenance and are less efficient than wheeled vehicles...








"My other car is an Abrams...."

Re: Non-FRC Bot
 

I think treads would be good at traction but not much more.

Sand would wear them down quick unless they were made of steel which would weigh a ton.

Balloon tires could work. May use wider lawnmower tires though.

would the AM planetary be seal enough for this task?


Yes I understand it is backwards but that's how research goes. I want to build a base for software development that is also capable of rougher terrain.


So let me start making a list of requirements

Mechanical:
1) 4+ wheel base with 2+ wheel drive (would prefer all wheels driven)
2) Capable of semi-rough terrain including sand.
3) Power transmission sealed from sand to prevent damage
4) Large enough to mount different actuators for testing purposes
5) Ability to turn is still nice so Width to Length must be minded
6) Enough power to move itself plus 50 lbs of other "equipment"

-----------------------------------------------------------------

As a side note:
I have a gokart sitting at my shop without use.

Rear has 25 inch tall by 12 inch wide wheels which should be more then enough "flotation"

I could modify the steering to be more "non-human user" friendly.
(it has a crappy rack and pinion. swing radius from steering to linkages is about 2 inches)

it has a 5hp briggs which could still be used or converted to be all electric. regardless the current power transmission is an open 440 roller chain with an unknown tooth ratio but size of the driving sprocket is about 1.5 in and driven is about 14 inches

the suspension on it would also have to be upgraded before it could be used harshly though. (solid rear axle mounted solid to frame)

or all of it could be scrapped for parts for the robit

Re: Non-FRC Bot
 

I like the idea of using parts you already have. But sometimes the parts are just not right enough for the job.

A concern for a robot like this is control. I'd try to err on the side of too slow of operating speed, rather than too fast. That means you might have to get creative with transmissions, to get enough reduction. The sand environment is tough on chains and toothed belts. I dont know if the AM planetary gearbox is sealed well enough, but I expect you could add a dust seal over the output shaft if needed.

We have used V belts for reduction and making a "forgiving" drive system for mechanisms on our FRC robots, you might look into them. Around here it's easy to get large and small pulleys, as they fit the very common evaporative coolers we use in the summer.

btw if you look at the garden cart....it could have the wheelbase and track width modified, and you might be able to use the steering system. They turn very sharp.

Re: Non-FRC Bot
 

Agreed just something to get up and running. Modifications could always come at a later time.


I think I need to do some research as to how the robots in the middle east fare to the sand and how they do it. Anyone know what iRobot uses for these situations?


Actually v-belts like the ones you describe are also used in HVAC systems and less sophisticated riding lawnmowers.


I actually have one of those carts although its been hrm . . . repurposed as a herbicide sprayer.

Re: Non-FRC Bot
 

Try micro-drives.com they ahve a selection of planetary gearmotors. i use them in my oun projects. you have to order over the phone but at about $30 a gearmotor it's cost effective especially if you go with a Motor-in-wheel design.

Re: Non-FRC Bot
 

Incorrect, actually. Treads are often used when the surface that needs to be crossed is marshy, somewhat swampy, unstable, dirty, or sandy, or some combination of those.

The objective is to spread out the weight so that the pressure is less so that you don't sink down into whatever it is you're trying to cross. Think snowshoes--if you're not wearing them, you break through any crust the snow has on it and go knee-, shin-, or ankle-deep until you get to the point at which the snow under your foot will support you. If you're wearing snowshoes, you go much less deep, because there is more snow to take your weight. Tank tracks take the entire weight of the vehicle and spread it out over the length of the vehicle by some width that, for a large military tank, could easily be a yard or two (taking both treads together).

For sand, it may not be as much of an issue, but it's still worth looking at. There's a reason a lot of heavy construction equipment and tanks use treads. An awful lot of small, light military/police-type robots use treads.

Large wheels are also an option, especially if you want car-type steering. Just make sure that they're big enough to let the robot scoot over the top of the sand as much as possible, or that the robot is heavy enough and has enough ground clearance to pack the sand behind it.

Re: Non-FRC Bot
 

True sir but the issue of endurance remains. Being a medium to small bot any ort of tread that will withstand sand over a long period of time would have to be most likely steel which would be rather heavy compared to the overall objective of the robot.

Re: Non-FRC Bot
 

Plastic or rubber treads would probably work well and last a long time. I expect the treads would cost more than pneumatic tires and wheels, and the mechanism to support them in proper alignment would be harder to build, and more complicated and expensive, and difficult to make sand resistant.

Although the steel treads on my 1940s vintage International Harvester crawler do still work after all this time....

Re: Non-FRC Bot
 

Nasa has been dealing with sand for a long time. They prefer the rocker-bogie suspension.

Another function that tracks can allow is climbing of square-edged obstacles like steps and stairs.

Re: Non-FRC Bot
 

We did a modified version of that for the breakaway game except we had 8 sets of wheels each half had their own articulating point rather then 1 articulating 1 solid

Re: Non-FRC Bot
 

Yes, but they have a robot on mars that's stuck in the sand and hopefully hibernating right now.

Re: Non-FRC Bot
 

But keep in mind, it did last much longer than it was designed to.

Re: Non-FRC Bot
 

@All

I like the enthusiasm that everyone has. A couple of points that I need to make first before I continue work.

I am limited in my fabrication skill. I have experience repairing and modifying things but very little in building something completely from scratch.

I am able to weld so having a ridged frame is not an issue.

I believe the best way to go about this would be to modify an existing platform or to build a simple one. Then go on to make an advance chassis or modify the existing one to increase capabilities.

So here are the options in my eyes.

Option 1) Build a square 4 wheel chassis
Option 2) Mount a system to the Go-kart chassis utilizing the current drive train
Option 3) Mount a system to the Go-kart chasis converting it to all electric

Advantages to each
1) Can build to any need. Can spin rather then turn.
2) Simplest to build (add a servo for throttle and a motor for steering). Potentially could go further then 1 or 3
3) More controllable then 2 yet simpler then 1

Disadvantages
1) More time from build start to run
2) Another area of failure with the engine. Unable to spin.
3) More power drain then 2 and less capable then 1

Re: Non-FRC Bot
 

Would the range of the wireless control system affect your decision? Will it operate autonomously some of the time?

I think you'd probably be happiest with a skid steer robot....much easier to get out of sticky spots. And since you are not sitting in the vehicle when you're driving it, you will probably get into more sticky spots than you expect

Re: Non-FRC Bot
 

Ideally it would be autonomous with manual backup.

Re: Non-FRC Bot
 

I've got an idea. If the go-kart chassis is anything like I think it is, tear the current drivetrain off of it entirely, and put a 4WD skid steer on instead. Also weld on a belly pan to help keep sand out of the internals, if there isn't a pan already.

The go-kart tires might work OK for use on sand, too, if the robot is fairly light.