Robot lawn mower video

[TimMooney]

Hey all, I finally have a video link for the lawn mower I mentioned last year in alain's "Lawn Mower Robot" thread. (Recall: my robot is the ugly pink-wheeled Barbie Jeep, with minimally intelligent control system and razor-blade cutters.) Well I added sensors for an electronic dog fence, which was quite a bit easier than I had thought it might be, and works really well. Have a look:

http://www.youtube.com/watch?v=zuXqg4q3cQ8

Tim

[TimMooney]

I mounted a camera underneath the lawn mower, so we could see grass actually getting cut. Unfortunately, there's not much grass left to cut in my yard. This is the really annoying aspect of having a robot lawn mower: when you run out of grass, you have to wait a week or so before you can cut it again.

Anyway, the video came out pretty well, on the second try. (On the first try, I had the camera mounted on the back flap, which --duh!-- flaps. Not so good. It was like watching CSI Miami, the picture was jumping around so much.)

Here's the link:
http://www.youtube.com/watch?v=giIqcmLtvWw

[D0n]

Tim, I wonder if you'd mind sharing some details about your system for using the radio fence, and controls.
I've got some work started on a hovercraft lawn mower, and a radio fence already in use on my dog.

[TimMooney]

I don't mind at all; I got the circuit off the net (http://www.electro-tech-online.com/, I think) in the first place. It's just a comparator with a coil and capacitor on one input, and a reference voltage on the other. I'm attaching an image, in which the basic sensor is followed by a timer, to veto very short output pulses produced by noise spikes. (Actually, the circuit I got off the net used the comparator's input bias voltage as a reference, but that's a little too small, and not controllable enough, for my purpose.)

I have the coils mounted about three inches off the ground, and the fence wire is buried an inch or less deep. I have to get the sensor within around 6-12 inches from the wire to stop the mower. I don't think the frequency of the dog fence is critical, but mine seems to vary periodically around 15 kHz, if I recall correctly.

As you might expect, the four sensors I have are not equally sensitive. I'm using 10% (or so) components, so the resonant frequencies of the coil/capacitor combination vary from sensor to sensor.

I think the idea of a hovercraft lawn mower is cool. Can't wait to see how it works out.

Tim

[D0n]

thanx!
will see if it helps my project!

[revoltage]

Timmooney do you schematics of your robot mower ? Could you send them to me please ?:)

[TimMooney]

Tim, I wonder if you'd mind sharing some details about your system for using the radio fence, and controls.
I've got some work started on a hovercraft lawn mower, and a radio fence already in use on my dog.

Hey D0n, check out this hover mower on youtube. http://www.youtube.com/watch?v=x7RcjT2LYUE

It's not a robot or anything, but it's a nice validation of your idea.

Tim

[TimMooney]

Timmooney do you schematics of your robot mower ? Could you send them to me please ?:)

There's a problem. I recently caught the mower going through the electronic fence, and I don't feel comfortable posting the circuit until I've tested an improved version without the defect.

The problem is that the circuit is basically just an RS flipflop (with some timers to debounce switches, etc). As you probably know, an RS flipflop has four input combinations, and its response to one combination is undefined. What happened is that the mower had both the front and rear sensors detect the fence simultaneously, because its path intersected the fence at a very shallow angle. It's been doing this safely for a year, but one time, the trailing sensor switched before the leading sensor (noise, or mismatched sensitivity can cause this), and the state of the trailing sensor prevented the leading sensor's transition from reversing the mower.

So, if you still want the circuit, I've given you enough information to reproduce it. If you can do it from the description, you know enough digital electronics to fix the problem, and I don't have to worry that I'm going to get someone in trouble.

[revoltage]

Hey TimMooney for the radio transmitter you gave above , what circuit are you using as receiver? You have a schematic for it?

[TimMooney]

Hey TimMooney for the radio transmitter you gave above , what circuit are you using as receiver? You have a schematic for it?

Scroll up to post #4 in this thread.

Tim

[revoltage]

Err would you mind telling me which one is the transmitter part and which one is the receiver part please?

[TimMooney]

Err would you mind telling me which one is the transmitter part and which one is the receiver part please?

The picture I posted was the receiver. I didn't make a transmitter; I just used the electronic dog fence that I already had. However, you can make a perimeter-wire driver by wiring a 555 timer in astable (oscillating) mode (if you search Google for "555 timer", you'll get at least a hundred web sites that show an astable 555 timer), and sending its output through a 2 kOhm resistor to the base of an NPN transistor, such as a 2N2222. Connect the emitter to ground, and connect one end of the perimeter wire to the collector. Connect the other end of the perimeter wire to +5V through a 20 Ohm resistor.

I haven't tried this, but I'm pretty sure it will work.

Tim