Yeah I suppose. But I really want to use DC motors in a real PID controlled loop.
Motors the size of the ones I posted earlier.
Alan KM6VV
Alan,
Here is a six wheel skid steering rover that I was working on before I moved to TX from CA : viewtopic.php?f=20&t=3650
It does not have independent suspension, but it can probably climb over most curbs. Ground clearance is around 1.5". I think that I can seal it up enough so that it can be reasonably water resistant. The wife and I are retiring to FL from TX and I plan to use it during hurricane season for remote surveillance around our town home. I plan to work on it here in TX over my company’s Christmas break of about two weeks.
Regards,
TCIII
Hi TC,
Thanks for the link. I’ve built a 4WD using similar motors and wheels for a client. You should have a lot of fun with it!
You’re right, it would probably transverse a lawn and some very minor sidewalks that I have just fine.
But it’s a lot of fun to speculate (dream?) how one could build a suspension.
I’m still waiting for additional data on running my new motor control board from I2C, so I got sidetracked into verifying my Table Top 'bot jaw (gripper) design. I re-worked it (again), now to use a four-bar linked jaw system. Can’t wait to see how it works!
Alan KM6VV
If the 4 wheel frame is hinged in the middle such that the front and rear can twist independently, much of the need for independent wheel suspension is eliminated. I’ve seen bots and I think actual army vehicles built in such a way.
That’s a possibility. Or a “rocker” on each side. Both sides would be connected together with a differential, which would transfer the traction, and keep wheels on the ground. Kinda like a subset of the Mars rover with its “rocker boogie”. Now for a way to make it out of LM wheels and motors…
Alan KM6VV
kdwyer,
I am presently putting together a Traxxas E-Maxx with parts purchased off of eBay. I am to the point of just needing the motor and the ESC to complete the chassis. The CC motor and ESC run around $250-$300 even on eBay. Anyway, I think that there would be a problem trying to create an AWS chassis because there would be no room for the servos and the the steering servo saver on the rear of the chassis. There are lateral links from the knuckles to the chassis to allow for adjusting the toe in of the rear tires, but mounting the servos and steering components might be more effort than it is worth.
Just a thought.
Regards,
TCIII
zoomkat,
The Tamiya Truck chassis, which was sold by Pololu, had that sort of suspension. The front and rear axels could rotate around the center line of the front and rear differentials.
Regards,
TCIII
km6vv,
I am presently in the process of completing a Traxxas E-Maxx chassis from parts purchased off of eBay. The suspension components are the same used by DPA’s jBot. Building a six wheel rover out of E-Maxx chassis parts would be a little difficult due to the fact that there is no E-Maxx differential with motor couplings at both ends. I suspect that an E-Maxx differential could be modified to that configuration allowing for the use of stock E-Maxx chassis parts and substantially reducing the cost of the rover.
Just a thought.
Regards,
TCIII
TC,
DPA is basically “ganging” the wheels on one side to the same motor, thus no differentials, isn’t he?
Alan KM6VV
KM6VV,
That is correct. That way he uses skid steering to steer the jBot. He used cogged belts and pulleys to drive the axels on either side from their respective motors. Using the E-Maxx differentials with one modified to feed through both ends would greatly simplify the design as the stock bulkheads and spring towers could be used instead of custom hardware as DPA did.
Regards,
TCIII
If you use the E-Maxx differentials, then you need to use Ackerman (or similar) steering, which really changes the whole idea of DPA’s drive.
I’m just not ready at this time to invest the time and $$ necessary to develop my copy of DPA’s 'bot. I did get a low-cost 4WD chassis a little like the E-Maxx at the end of last year, but I haven’t done anything with it.
I am still perusing the navigation aspect of the RoboMagellan project, I have a 9DOF IMU to play with, and a new GPS module. However, I’m now leaning towards implementing them in a helicopter. Might be 'quad, but after seeing a real Hobby-Helicopter fly (and getting one!), I’m more anxious to make it autonomous (after I can fly it!).
Thanks for the comments.
Alan KM6VV
I’ve committed to a Robomagellan 'bot for next April.
Early work on Ebay has netted me the wheels and suspension parts I was after. Traxxas E-Maxx parts.
Lower chassis plate is fiberglass, the next two decks (below and above the fiberglass bulkheads) will also be of 1/8" stock.
Into each of the grey bulkheads will go a machined round aluminum plate to take a ball bearing race. There will be two motors each driving the three wheels on its side via 1/5" pitch timing belts and pulleys. a series of six drill-rod shafts will drive the CV joints to the wheels. The opposite ends of the shafts will be supported by bearing blocks.
A 15 A RoboClaw will drive the motors from a 24 V battery.
More on electronics later.
Alan KM6VV
Ha!
We need to expand the heading from:
“2WD, 4WD, Carpet Rover, and other rover robots.”
to:
“2WD, 6WD, Carpet Rover, and other rover robots.”
Neat idea! I’d love to see it under power.
Why don’t you let me machine a chassis for your suspension?
Sure!
It’s the length that’s stopping me from easily machining it. For this stage of the chassis, it’s a 24" x 6" piece of 1/8" fiberglass. Two more decks go on top, middle about 5" wide (top 7" wide), length to span the three red bulkheads. Bulkheads on 8" centers. So far, I’ve simply drilled the bottom chassis on the drill press. Next I’ve got to mount the bearing blocks for the pulleys, and the belt tensioners. More holes!
A little while before power, I forgot to order two drive belts for the motors. And I still have to finish making the bearing blocks, the tensioning blocks, and the motor mounting bulkheads!
Alan KM6VV
Give me a drawing of what you’re looking for and we’ll see what we can come up with.
Thanks I’ll work on it this weekend.
Alan KM6VV
Good project idea.
Why not 6 motors ?
Look here you may find some parts or project start. Look for rollers
rcgroups.com/cars-trucks-and-parts-fs-w-564/
DiaLFonZo,
By ganging three wheels to each motor you only need two encoders, one for each motor, to be able to determine wheel speed and distance travelled. In other words all three wheels on each side are turning at the same speed and go relatively the same distance. No two motors turn the same speed for a given voltage. The spec is usually +/- 10%. If you were strictly using dead reckoning for navigation, that kind of error would be deadly.
Just a thought.
Regards,
TCIII
Encoders are usually only accurate when there is minimal or no slippage. Using encoders on the wheels of a vehicle like this would probably provide accurate distance measurementonly for limited straight line travel under ideal conditions.
@zoomkat,
Encoder (odometry) correction is a real art. Here is how it is done even when the rover does not move in a straight line from beginning to end: geology.heroy.smu.edu/~dpa-www/r … r/imu_odo/
Regards,
TCIII