Threaded View
Subject
Padu posted on October 21, 2005, 4:55 pm
It's Friday, nobody is really working right? So let's warm up this ng a bit.
I just got the news that my budget for a new mobile platform was approved
(yeeeey), so I need to start worrying about chosing one very soon. The
project is to create agile rovers that can accomplish fully autonomous
missions on a 20 miles range at speeds up to 20mph. It should be smaller
than an ATV but bigger (and more rugged) than an R/C car. It must idle well
(in case of gas engine). It must have built in reversal and automatic
transmission (in case there is more than one speed forward). Low cost (less
than $3000). Finally, it must be able to overcome rough terrain (dirt roads,
small rocks, grass, grades, whatever it is possible to do with an ATV).
I already have a few ideas in my mind, please critique mine or suggest
novel:
- Custom build gas platform: using a small 4-stroke engine (1 to 9HP). Pros:
we can build exactly what we need. Cons: takes too long to build, may have
lots of flaws as I'm no mechanical engineer
- Small ATV. Pros: it's ready. Cons: Too big and cumbersome to carry around,
barely fits the budget of $3000
- This little babe: http://www.theultimategolfcart.com/ . Pros: perfect size,
looks nice (humnn, shouldn't be a reason for robotics right ;-) ), electric
allows us to demonstrate indoors. Cons: too heavy (220 pounds!), I'm not
sure a 1000W motor is strong enough for this app.
I'm leaning towards the last one, but any suggestion is welcome.
Cheers
Padu
I just got the news that my budget for a new mobile platform was approved
(yeeeey), so I need to start worrying about chosing one very soon. The
project is to create agile rovers that can accomplish fully autonomous
missions on a 20 miles range at speeds up to 20mph. It should be smaller
than an ATV but bigger (and more rugged) than an R/C car. It must idle well
(in case of gas engine). It must have built in reversal and automatic
transmission (in case there is more than one speed forward). Low cost (less
than $3000). Finally, it must be able to overcome rough terrain (dirt roads,
small rocks, grass, grades, whatever it is possible to do with an ATV).
I already have a few ideas in my mind, please critique mine or suggest
novel:
- Custom build gas platform: using a small 4-stroke engine (1 to 9HP). Pros:
we can build exactly what we need. Cons: takes too long to build, may have
lots of flaws as I'm no mechanical engineer
- Small ATV. Pros: it's ready. Cons: Too big and cumbersome to carry around,
barely fits the budget of $3000
- This little babe: http://www.theultimategolfcart.com/ . Pros: perfect size,
looks nice (humnn, shouldn't be a reason for robotics right ;-) ), electric
allows us to demonstrate indoors. Cons: too heavy (220 pounds!), I'm not
sure a 1000W motor is strong enough for this app.
I'm leaning towards the last one, but any suggestion is welcome.
Cheers
Padu
blueeyedpop posted on
Re: Ideas for mobile platform
there. It is 3 wheeled, and therefore less stable. It is not intended for
slopes beyond 20 degrees, on grass.
Building a platform capable of 20 mph is going to cost a lot more than 3K.
How about a youth ATV? like a Raptor 50 from Yamaha?
Dimensions
L x W x H 60.5" x 32.5" x 36"
Seat Height 24.3"
Wheelbase 40.6"
Ground Clearance 3.0"
Fuel Capacity 1.8 Gallons
Dry Weight 237 Lbs.
Put an electric motor on there with a clutch for reverse.
(less
Pros:
Re: Ideas for mobile platform
I've been thinking about that golf cart and day by day it seems a weaker
option... if they only had a reseller here in San Diego that I could see and
test it before I buy...
On the kid's ATV side, I had put some thoughts too. The 50 or 80cc seem
perfect, but they don't have reverse, I'm not sure what you mean by putting
an electric motor for reverse. The Grizzly 125 is the only I could find with
built in reverse. I haven't looked at other offstream brands, so my answer
may be there...
Cheers
Padu
Re: Ideas for mobile platform
could tsndem a motor to.
Mike
Si Ballenger posted on
Re: Ideas for mobile platform
I think I've mentioned this before in another post, but the rear
engined snapper lawn mowers might be a good platform to try. If
your project is a "not for profit" venture, you might be able to
get a local lawn mower dealer to sponser an old used one for free
(maybe for a little "NASCAR" type of advertising for the dealer
on it). Otherwise a used one in decent shape might be had in the
$500 range in the want ads. I've got one I've had for ~15 years
and it is *rugged* and dead simple in operation. It hauls me (210
lbs) in and out of the ditch next to the main road without
problems. The belt system that drives the blade would be very
simple to use to drive an alternator and/or hydraulic pump to
support onboard actuators. Foward-neutral-reverse and go-clutch
in-stop is all simple push/pull actuation. Snappers are popular
in the south, but not sure about the rest of the country. If you
have a dealer near you, go buy and take a test ride around the
shop to see how they work.
http://www.snapper.com/rear_engine_riders.html
Pat Farrell posted on
Re: Ideas for mobile platform
[snip]
They are rugged, mostly because they are dead simple.
Do they really have automatic transmissions? The ones I've seen
have a sliding speed/"gear" control with notches, but they
really don't have gears in the motion drive train, they are more
continuously variable using a friction disk. I would bet
that you could remove the teeth on the notches and use a servo
to switch from low to hi ratio. Or just pick a ratio and bolt it
down.
--
Pat
Re: Ideas for mobile platform
Heavy duty gearhead motors modified to be servos like in the
below link from another thread could operate the go-neutral-stop
pedel, foward-neutral-reverse lever, and turn the short throw
stearing wheel. The MSRP for a new bottom of the line Snapper is
~$1,250 ($1,700 for electric start capability), well with in the
$3,000 budget. If care were taken with the modifications, the
moble platform could be returned to a lawnmower when the project
is completed. Anything much lighter might do cartwheels when it
hit a big rock or bump in the road at the higher speeds.
http://www.cpg1.freeserve.co.uk/servos/servos.htm
John Nagle posted on
Re: Ideas for mobile platform
Having just finished with a DARPA Grand Challenge entry, I think
you're underestimating the size of the job.
See
http://www.ghostriderrobot.com/
the self-balancing autonomous off-road motorcycle that ran in the
DARPA Grand Challenge. That meets all your stated requirements.
You just have to be really good at controls.
John Nagle
Team Overbot
Re: Ideas for mobile platform
I agree with you, John, but the Grand Challenge was a race, 20 MPH would not
cut it, and 20 miles is a lot shorter than 200 km.
I just love this silly motorcycle. It is totally impractical, but
as a proof of engineering skill, it is so much more impressive
than making a Hummer go down a dirt road.
Doesn't this contradict the autonomous criteria?
--
Pat
Re: Ideas for mobile platform
No, I'm not underestimating... it's been almost one year I'm working on a
small scale (R/C car scale) autonomous thing and it's already hard enough. I
went to Primm to watch the GC, and that made me gain an special appreciation
for this type of effort.
I saw ghost rider in person (it's much smaller than I thought it was) and
being a motorcyclist myself, I admire that they could get that thing moving
by itself...
What I think I could have been misunderstood is that for someone that is
mechanically talented, building something like an offroad go-kart with off
the shelf components may not be as difficult at first glance. In my case I
would take ages to do it, as my handyman skills don't go past weekend around
the house projects...
Cheers
Padu
dan michaels posted on
Re: Ideas for mobile platform
I don't know how far that thing got this year, but last year as I
recall it went about 10' and fell over. Clearly top-heavy. OTOH, an
off-road motorcycle makes a nice light-weight and cheap platform that
can go 50 MPH, and get 40 MPG.
Suggestion - add some "training wheels", or an outrigger, etc. Or
better yet, turn it into a tricycle, like you see a lot of Harleys made
into.
Re: Ideas for mobile platform
It went a lot farther. Cheap didn't seem to be
much of a criteria in this event.
For years, the Baja 1000 was won overall by motorcycles.
They are nimble, fast, accelerate quickly and
Harley trikes are so the bike doesn't fall over when the rider
is too drunk to remember to put out their feet. :-)
Single track vehicles can avoid a lot of obsticals that
two or three track vehicles hit and have to absorb.
I assume that the ghostrider folks did it for the challenge.
There are lots of challenges in a robotic bike.
--
Pat
Brian Dean posted on
Re: Ideas for mobile platform
Actually, it didn't complete in the GCE.
I think they had this on this year's version - retractable "landing
gear" that would fold up when the vehicle started into motion and then
fold back down when stopped.
-Brian
Re: Ideas for mobile platform
The problem, of course, is that the gyro effect of the wheels is only
effective when the wheels are spinning at a high rate. If you need to
slow down for an obstacle, etc, then the top-heavy 2-wheeler falls
over.
That's why human beings have a leg on each side - to put down to hold
the bike from falling over!
I started reading the book Fab by Neil Gershenfeld last nite, and lo
and behold found a picture of a guy in India who sells $400 conversion
units for motorcycles that converts the rear-end into 2-wheels, like a
trike. These use it as a cheap tractor. You can quickly remove the
wheels and convert it back into a regular 2-wheeler. I'm surprised the
Berkeley guys didn't attempt something like this.
Re: Ideas for mobile platform
Then they wouldn't be doing anything different than the others...
Padu
Re: Ideas for mobile platform
Ever watch a trials rider? if not go rent On Any Sunday.
They can stop, back up and then go foreward again without
doing a "touch". A touch is touching the ground with any
part of your body.
Humans may have an advantage of balance, but the idea of
a robot doing trials speed work on a motorcycle is so cool.
Adding an pair of emergency legs so you could do a dab
might be a way to do best of both worlds. Walking robots
aren't easy and not clear that they are even desired in general.
A motorcycles is going to be fine at speed. Don't know
if the added weight and complexity makes sense.
--
Pat
Re: Ideas for mobile platform
Check out this video if you are interested in
slow speed balance.
www.ghostriderrobot.com/videos/superslow.wmv
Pretty impressive to me
--
Pat
Re: Ideas for mobile platform
They're active balancing, using subtle body movements, not using the
wheel gyro effect.
Re: Ideas for mobile platform
There was a movie out recently about the Baja 1000. What hoot. The guys
on the bikes were complete maniacs. 100-120 MPH on dirt roads.
The Berkeley team probably should have rounded up a harley driver for a
consultant.
Site Timeline
- » I got 2 books for presents
- — Next thread in » General Robotics Forum
-
- » Advice on switching microntrollers
- — Previous thread in » General Robotics Forum
-
- » evoMUSART 2013: First CFP (with correct dates)
- — Newest thread in » General Robotics Forum
-
- » Heat pump refrigerant change to R-22 substitute
- — The site's Newest Thread. Posted in » General Metalworking
-
- » DCC sound question
- — The site's Last Updated Thread. Posted in » Model Railroad Forum
-
XML Sitemap