Need Some Advice On Simple Robotics Project

Projects General Guidance
1

hey guys.

i have a project where i have to build and simple robot that can carry a load across a course then stop at a gap, drop a bridge, pass over the bridge and then drive into the endzone.

after researching i found that i need a Arduino Uno and a motor drive shield in order to get started. i have a simple chassis in mind too. i just need some guidance on what else to buy to get the robot set-up for programming.

  • our group prefers that we map the course directly rather than using any sensors, since it is a pretty standard course.
  • we plan to use 4 DC motors
  • we are still largely uncertain on how we are to get the robot to hold a bridge then release it when it gets to the gap.

any help would be greatly appreciated.

thanks

2

If you have pictures of your robot and the bridge, we can suggest ways to combine the two.
Details are a must. How heavy is the bridge, the lenght/width is the gap you will be crossing, and more importantly, can the bridge support the robot?

Your most likely going to bring the bridge over the robot to then drop it across the gap. For that, you may need grippers, however, if the bridge has metal on it, you could use electromagnets. Of course you will need at least two servos on both sides, to lower the bridge first before dropping it in place.

So if you have any designs in mind, make a list of possible parts, and decide which one to make.

3

the bridge can be anything. we can shape it however we like. obviously it has to be wide enough and strong enough allow the robot to cross . any material can be used as long as the system all up is less than 8kg. we were thinking light plastic

the bridge basically has to rest on the robot until we get to the gap, then be lowered or dropped somehow.

here is the course: http://i.imgur.com/Y2pnoK9.png

the gap length is 40cm. the width of the gap 60cm.

4

Well, wide Popsicle sticks are great to build with, especially for small model bridges. Do you need to pick the bridge back up, once you cross it? Anyways, servos and either grippers or electromagnets will do nicely to drop and release the bridge.

5

I suppose this is some competition? Do the rules state that there absolutely has to be a deployable bridge, or do you just need to cross the gap and you thought the bridge would be a good idea?

If the rules allow it, it could be done without the bridge.

When the robot reaches the gap it stops, servos deploy the "training wheels", lifting entire robot. Extending arms are just over 40 cm long so they can reach over to the other side. Once over the gap, arms are raised and the robot continues normally. At least one arm would need to be powered (wheels spinning).

6

A couple of interesting aspects are that the bridge has to be (almost) as wide as the robot, presumably with side flanges as wheel guides, and that you will need to ensure the guidance sensors can see over/ under/ round it whilst being carried.

Did I say guidance sensors? Well, the problem with absolute navigation is that it requires that there be no wheel slip and that there be no "motor slip". You would for a start, use either stepper motors or DC motors with optical encoders fitted.

If you are going to cross a gap, either feelers (possibly optical) or an ultrasound sensor (for which I notice your course is specially designed) will be pretty much essential as one small miscalculation of the front wheel and - you're off!

If the rules permit it, the "spider" concept that would be an extension of the previous posting, could be worth considering.

7

I would not use a motor shield. I would buy 2 dual motor drivers. Also wait until you pick out your motors so that you buy one that can handle the current. How are you building the robot. From scratch? From a kit? And using dead reckoning is hardly ever a good Idea on a Robot. Use sensors!

Also I'm curious what competition this is.

Best of luck!

8

our group prefers that we map the course directly rather than using any sensors, since it is a pretty standard course.

Unlikely to work you do need to use sensors.

we plan to use 4 DC motors

Over kill two is far cheaper.

we are still largely uncertain on how we are to get the robot to hold a bridge then release it when it gets to the gap.

Google "bridge carrying tank" for ideas must it be a bridge could you not just fill the gap?

Mark

9

If it can be done with legos you can do it with an Arduino.

Watch the video and look at ALL OF THE SENSORS they use!

10

If the ditch is real and your not required to use (ie its no a must in the rules) then there are a number of other options.

One is to design something that can fill the ditch such as a fascine carrier (google) or something that can go into the ditch and get out again. The first tanks (WWI) tried to do this.

Mark

11

That. Was. Awesome.

If wanted to be really cool. Do you know how deep the gap is? Have robot extend "stilts" down into pit, sorta do a modified "pole vault" across (with 'simple' actuator pushing from ledge behind it), retract stilts, and keep moving....?

12

If you don't need to leave the bridge behind (or can do it without needing a real "bridge", in a manner similar to Shpaget's idea - you could try something like this interesting device (substituting wheels as needed):

http://cyberneticzoo.com/?p=8174

13

One problem I just noticed.

The topic.

Need Some Advice On Simple Robotics Project.

Simple?

Some sort of oxymoron.

14

^ hahahah. thanks for all the help guys.

just some notes:

15

Well aside from what ever sensor you will need for the bridge, you will two sensors. One in the front to detect the edge, a ultrasonic distance or IR sensor will work. And if you know the length of the bridge, you can use an encoder to track how far the robot goes compared to the bridge length.

16

abdullahkiran:
you don't really need to make a bridge, you just have to get a across

OK, so that's a different consideration.

abdullahkiran:
(also you lose marks if you leave any part of the system behind, but our team has already decided that retrieving the bridge isn't one of our priorities)

But a serious consideration.

abdullahkiran:
I have basically bought this: http://aud.dx.com/product/dual-layer-4-motor-smart-car-chassis-w-speed-measuring-coded-disc-black-yellow-961151803#.Uik7ET9KWlt,

That looks to have a seriously short wheelbase - hardly up to the job of carrying and deploying any sort of bridge.

I rather like those Russian "steppers". The principle sounds good - try this:

Have the vehicle about 1½ times as long as the gap (it will have to be at least as long as the gap anyway if it were carrying a bridge) with three wheel sets of which the central one can move from one end to the other, and a heavy top section which can also move from one end to the other, in fact from one extreme end to the other.

When advancing forward, have the ballast (top section) towards the rear and the middle wheels halfway along - the front wheels will be lightly loaded. The gap sensor is underneath, just behind the front wheels. When it senses the gap, the front wheels will be projecting over the gap. Lock the middle wheels just before the edge of the gap, move the ballast to the absolute rear and use the shift mechanism of the middle wheels and the rear wheels to drive the vehicle over the gap.

Once it straddles the gap (gap sensor behind the front wheels will indicate), drive the middle wheels and the ballast across, advance until the rear gap sensor (symmetrical - can be used in either direction) finds the gap at which point the middle wheels will be on the other side, lock them and move the rest of the vehicle across.

Strictly, the middle wheels do not need to drive, only lock. On further thought, you do not need wheels at all, only a jack which can be lowered and moved back and forth from front to back. You can allow the front wheels to drop (partially) off the gap so that it is resting on the central jack and perform the rest of the manoeuvre from there. The jack itself can have a force sensor which will adequately serve as the gap sensor - i.e., when the vehicle is resting on the jack in the retracted position, you know that the wheels have fallen in the gap. Now that simplifies matters!

Hmmm.

17

abdullahkiran:

How does that thing turn the tight corners in your course? I don't see a steering mechanism.

18

It has subtraction steering, like a tank.

19

HazardsMind:
It has subtraction steering, like a tank.

Ah, thanks. In that case I agree with earlier comments that sensors are needed to counter wheel slip. In fact I would try to find a way to sense not only the wall/edge but also in what direction the wall/edge is in, since a commanded 90° turn may result in a 91° or 89° turn and next thing you know you're trying to lay your bridge crooked and it tips into the chasm.

20

Its called differential steering, my bad.