Hi all,
I've been thinking about this for a while, any project is a long way off though! One day I would like to attempt to make a robot that could run about the yard cleaning it up.

First problem before I can do anything is location, I thought about gps but its not quite accurate enough, buried wire isn't a option unless I dig up the concrete!

Is there a reason I couldn't use a couple of ultra sounders in two corners of the yard activated by radio transmitters on the robot? So the robots control logic would know how far it is from each 'beacon' therfore where it is in the yard? I've not had anything to do with ultrasound before, at the moment its the idea that's interesting me.

I fear there are probably practicalitys that make this idea not workable? What do folks think?
Cheers Geoff

 

There are a number of ideas for your toolbox, GPS, digital compass, laser, IR, ultrasound. I would consider using as many as possible.

Your idea of beacons using utrasound is a sound idea. You can trigger the beacons using radio or IR and measure the time it takes to receive the signal. You can have three or four such beacons in the yard and use triangulation to determine your location. Resolution using ultrasound is very good, even down to 1 cm.

 

Hi all,
I've been thinking about this for a while, any project is a long way off though! One day I would like to attempt to make a robot that could run about the yard cleaning it up.

First problem before I can do anything is location, I thought about gps but its not quite accurate enough, buried wire isn't a option unless I dig up the concrete!

Is there a reason I couldn't use a couple of ultra sounders in two corners of the yard activated by radio transmitters on the robot? So the robots control logic would know how far it is from each 'beacon' therfore where it is in the yard? I've not had anything to do with ultrasound before, at the moment its the idea that's interesting me.

I fear there are probably practicalitys that make this idea not workable? What do folks think?
Cheers Geoff

If I was to implement the location finder in this manner, I would use RF instead of ultrasound. The ultrasound might have an interference from bouncing off objects; the RF would too, but I think it would be less.

Maybe an easier way is to have a home spot that the robot can reset it's position from. The home can be in a corner. To reset, it tries to move as far as it can into the corner.

 

...
Your idea of beacons using utrasound is a sound idea.
...

(pun)

And using InfraRed beacons would be a "step in the light direction"?

 

Thanks for the responses, are there any issues using ultrasound? As Mr Chips said I would get the robot to trigger the beacons in turn using a radio transmitter and work out distance from each. What is the practical maximum range for such a set up? Would weather have a significant effect? Ie if the winds howling would that cause any problems?

LCD3, how would you go about rf location? I'd sort of assumed rf would be to quick to be able to use practically, I'd considered using a sort of home spot but yards get slippery so it needs to 'know' if its wandering off line!
Cheers Geoff

 

LCD3, how would you go about rf location? I'd sort of assumed rf would be to quick to be able to use practically, I'd considered using a sort of home spot but yards get slippery so it needs to 'know' if its wandering off line!
Cheers Geoff

You're right, ultrasound is easier to determine distance. With RF, I think all you can measure is direction, so you would need to have at least 3 transmitters.

Now that I'm thinking about it:
You would need at least 3 transmitters for ultrasound as well. Direction might be difficult to provide accurately since you would need to turn the antenna and monitor signal strength, which will take several seconds (I think) for each one. Ultrasound could probably acquire the distance for all transmitters in less than a second.

 

Standard ultrasound sensors are reasonably directional i.e. the transmitter and receiver need to be pointing at each other. I guess you beacon could have an array of transmitters pointing in different directions but you would have to watch out for null points due to interference.

I have just built an ultrasound distance measuring unit where the transmitter and receiver are separate so that the receiver tells me how far I am from the receiver (I need to write this up). It works fine as long as the two modules are line-of-sight and pointing at each other. You would also need to know which signal came from which transmitter.

 

Standard ultrasound sensors are reasonably directional i.e. the transmitter and receiver need to be pointing at each other. I guess you beacon could have an array of transmitters pointing in different directions but you would have to watch out for null points due to interference.

I have just built an ultrasound distance measuring unit where the transmitter and receiver are separate so that the receiver tells me how far I am from the receiver (I need to write this up). It works fine as long as the two modules are line-of-sight and pointing at each other. You would also need to know which signal came from which transmitter.

An easy way to tell is to trigger them at different times.
i.e. Trigger 1 --> receive signal
then Trigger 2 --> receive signal
then Trigger 3 --> receive signal
No interference from each other since only one is one at a time.

 

I sure do hope you don't let the dog outside while this thing is on.

You don't have any kids yet, do you?

 

I love the bats we have in my neighborhood. They eat an incredible amount of insects. Not sure how they'd react to this!