Laser Object Detection

Recently, I got a green laser from EBay for only $20.00 (usually these lasers cost $50-$80), to use with my robots. As you can see in the above picture, it's REALLY bright - the range is something like two miles.

The main advantage to a laser is that it puts a very bright light in one precise place. This can be used for things like making two robots align themselves to point towards each other (which I'm thinking of doing for my next project).

Anyway, I found that if I put the laser right on top of a light sensor and pointed it towards the ground (while the light sensor was level), the light sensor would only detect the beam if it reflected off an object a short distance away into the sensor. So I made an attachment for this and put it on Trax, my first successful rover that uses the new treads (using spring-loaded sprockets works great - thanks Brian!):

Below is a video of the robot detecting a wall... notice how the beam moves along the ground until coming across an object. Then it moves up the object as the robot gets closer to the object, until the beam reflects into the light sensor.


Although it was a fun idea, I have to say it's very impractical. The one theoretical advantage it has over the US sensor is that it can detect curved/slanted objects no problem - the reflection of the beam is just as bright no matter how curved or slanted the object is (except of course when the object is a mirror or another highly reflective material). However, usually when the robot comes at a wall from an angle it can't get close enough to the wall for the laser beam to get high enough. It's disadvantages include having a very small range of detection, and being susceptible to ambient light conditions and object textures. It might, however, be useful when combined with the US sensor to provide additional information about potential obstacles.

-Jonathan

NOTE: Since lasers (especially green lasers) can harm your eyes if pointed directly at them, you should exercise caution when using them.

Comments

Unknown said…
As this blog is explicitely directed on young builders (also), it might be reasonable to put some big "Not suited for children" label on the post - as far as I know, laser beams can be harmful when one looks directly into it.
Good idea... I added a warning to the post. I wouldn't say lasers aren't suited for children at all, but they should definitely have their parents permission before using one.

-Jonathan
Ethan Steckmann said…
This comment has been removed by a blog administrator.
Unknown said…
Ethan,

That's a somewhat careless practice (and a careless talking moreover in a public blog frequented by readers that usually are physical laymen), I'd say:

"A person exposed to laser radiation [..] may be unaware that damage is occurring" - for details on laser safety, see here, for instance.
Ethan,

Your comment has been removed - sorry. Staring into a laser... any laser... is simply stupid.

We have young (VERY young) readers on this blog and your comment could give the wrong impression.

Jim
Jonathan, I like your creative contraptions and their wide range of uses!

Without trying to pull this topic further from the MAIN idea...I would like to add my personal experience.

I have used lasers at home and in the course of my career and I can detect some amount of damage to my eyes from the exposure. I've used several different types of the years. While, I cannot remember ever being flashed directly into my eyes...I do believe just looking at the beams and receiving dispersed reflective exposure without protection (yes, sometimes) has harmed my eyes over time.

Chris
Anonymous said…
It might be interesting to use your setup to detect retro-reflective tape, that redirect the received light to the sender. See my laser sensor for more details.

Also note that a green laser may not be the best for this use: while the human eye is most sensitive to the green (hence the perceived brightness) the LEGO sensor is mostly sensitive to infra-red and red.

Maybe also a minifig magnifying glass in front of the sensor might improve sensitivity (distance between glass and sensor will need to be carefully adjusted)
Micah E. said…
Theoretically, lowering the light sensor while keeping the angle and altitude of the laser constant should increase the range... but I'm assuming that you toyed a great deal with the configuration. Interesting project!
Philo,

Thanks for the ideas... I actually did get a red laser at first, but it wasn't very powerful, so I returned it and got the green one. The main thing I'm planning on using the laser for is to let the light sensor detect it from across a room so that one robot can find another. To do this, I'm going to put a piece of paper over the light sensor (not right up against it), such that the laser can hit any part of the paper and will illuminate the inside of it, which the light sensor can detect. Since the green laser has a relatively big "dot" (and therefore illuminates the paper more), it seems to work perfectly for this. I'm not sure if the red lasers have as big "dots".

Catpit,

Actually this design was pretty much the first thing I came up with... I just wanted to try it out, but I didn't try to improve it since it doesn't seem useful for anything yet.

Your idea would probably work, although the reflection detected by the light sensor probably wouldn't be as bright, and would therefore be more susceptible to ambient light conditions.

-Jonathan
Ethan Steckmann said…
Sorry if my comment was uncalled for.

I really do feel sorry.

I forgot about the age group who reads your blog.
Unknown said…
I dont know much about green lasers, but couldnt you put a filter on the light sensor so it only picks up the wavelength that the laser emits? This might cut down on interferance.

I have no idea if they make a filter like this, but there may be something out there.
That sounds like a neat idea, but I don't know of anything like that.

-Jonathan
Brian Davis said…
Hmm... my 2 cents as well I guess...

Like Chris, I've been "exposed" to lasers both at work and home. There are several classes of lasers to worry about. Many of the "public" lasers are Class III devices (< 5 mW output), which officially carry some risk, but that will depend partially on the output and partially on the focus. For this application, you really don't want a tight focus - you want a broad and somewhat diffuse beam to illuminate the approaching wall.

There are also Class II lasers (a bit harder to find), that are considered "safe" for at least brief eye exposure (the barcode scanners in many supermarkets use these, as direct & reflected light could hit patrons and much more importantly workers).

Personally, I'd avoid the green "high visibility" lasers for a couple of reasons, including often higher power (near the upper range for a Class II device), more energy at the green end of the spectrum where the light sensor is less sensitive... and the red are significantly cheaper.

We recently used very low-power lasers in my son's science fair project with great success. we arranged it so that every time someone walked down the hall, they interrupted a laser pointed at the light sensor, and the NXT count "count" them this way. This ran for about 60 hours... and no one ended up noticing it. Very interesting data as well.

--
Brian "Datalog it" Davis
Unknown said…
I emailed a company that deals with lasers and they said there is such a thing to only allow certain wave lengths through a filter.

They called it a Transmission Filter.

Say your Green laser is 532nm. You would need F10-532.0-4 on this web page:

http://www.cvilaser.com/Catalog/Pages/Template2.aspx?pcid=206

At $75 for a half inch filter it is pricy, but i guess the main thing is something like this does exist and would cut down on interference.
Anonymous said…
Wow, the idea of the robot is very cool, but i was wondering if you can make the robot "find its way around" using this method.

I also liked the tank treads, where did you get the materials for it?
Wow, that's interesting Chad... too bad it's so expensive.

Anonymous,

It can't consistently avoid obstacles, because of the variable light conditions and small range of the laser/light sensor. I got the tread links and sprockets from LEGO Education.

-Jonathan
Those links and sprockets are in bricklink now also.

Chris
Anonymous said…
This comment has been removed by a blog administrator.
Unknown said…
Would the NXT colour sensor detect direct laser light. I tried using a 100 mw laser pointer as a trip wire and setting the sensor in front of the point where the beam ends; but the sensor still does not detect the reflected laser light.
Unknown said…
Would the NXT colour sensor detect direct laser light. I tried using a 100 mw laser pointer as a trip wire and setting the sensor in front of the point where the beam ends; but the sensor still does not detect the reflected laser light.

Popular Posts