Laser navigation
Jonb
Posts: 146
For many months now I have been researching laser ranging devices.
Lasers are fast becoming the method of navigation,mapping etc. for use in robots.
Sonar and IR rangefinding modules have their place but giving a robotic device such limited obstacle detection range just doesnt cut it.
At this time, I have yet to find commercially available laser rangefinding modules(small and affordable).
My attention is currently turned towards commercially available products using this technology.
Example: Bushnell Yardage Pro 800 Compact Rangefinder
The range information is displayed in the viewfinder by some sort of LED display(I assume its like the HUD)
This indicates logic circuitry that may be accessed(hacked) somehow. I need to get my hands on one.
I·cant believe that in 2005 we cant mount lasers rangefinders on our robots at a reasonable cost·
Post Edited (Jonb) : 1/26/2005 11:23:42 PM GMT
Lasers are fast becoming the method of navigation,mapping etc. for use in robots.
Sonar and IR rangefinding modules have their place but giving a robotic device such limited obstacle detection range just doesnt cut it.
At this time, I have yet to find commercially available laser rangefinding modules(small and affordable).
My attention is currently turned towards commercially available products using this technology.
Example: Bushnell Yardage Pro 800 Compact Rangefinder
The range information is displayed in the viewfinder by some sort of LED display(I assume its like the HUD)
This indicates logic circuitry that may be accessed(hacked) somehow. I need to get my hands on one.
I·cant believe that in 2005 we cant mount lasers rangefinders on our robots at a reasonable cost·
Post Edited (Jonb) : 1/26/2005 11:23:42 PM GMT
Comments
lol, sorry, i just HAD to put that in there [noparse]:)[/noparse]
How do lazer range finders work? Send a pulse from the lazer, an wait for it to bounce back timing the pause?
Is it possible to make one usnig a cheap lazer pointer?
What happens if the surface is not flat, or if the lazer does not hit it at precise 90 degree angle.
I see im not helping here, acutally adding more questions. oh well.
I had an idea of using a CMUcam to detect where my laser was pointing.· It's a bright red/orange dot....stands out from most backgrounds.
Maybe play with some phototransistors and some optical filters....put a long slim tube leading to the photo-op (this might give you some more directional resolution) then step the sensor around until it sees the laser....then go in that direction!
I'm sure you could find some 'gel' filters that would get rid of ambient light issues and hopefully sun too (polarized?).· Not sure about scattered reflections!· the 2nd trip reflections might be brighter than the original reflection point....
And this doesn't give you an idea of distance....it's more of a beacon!· You might try one on of the IR range sensors after that (but they're fairly broad and susceptible to flourescents).
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·
Steve
http://members.rogers.com/steve.brady
"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."
http://www.seattlerobotics.org/encoder/200109/lasernav.htm
Stop me if I'm wrong, but laser rangefinders pulse a laser and time the receipt of the pulses (just like ATC radar).· If you want to measure something to the nearest foot, you'd need to be accurate down to the NANOSECOND.··I'm still amazed that radar equipment is that friggin' fast (and ATC radar can determine radial velocity based on the slight change in frequency too, which requires amazing frequency resolution).· The electronics for a laser rangefinder need to have to have·speed, accuracy, AND resolution.· You definately won't be able to build a range finder with a BS2.· But you could use a rangefinder module that interfaced the BS2 for data.
Also, you CAN mount a laser on a robot for·pretty cheap.··But mounting a·rangefinder requires more than just "mounting a laser".
Dave
http://www.seattlerobotics.org/encoder/200110/vision.htm
Bistatic uses a separate transmitter and receiver, where monostatic uses the same paths (single antenna).
Basically, on our radars we use one antenna for transmit and receive....so we fire our pulse then turn off the radar and wait for a receive pulse.
Because the pulse travels at the speed of light (near enough without all the fancy shmancy calculations) then depending on the width of your pulse, you are blind for that length of time.
For us, using a 0.8uS pulse where distance = C * t = 3x10^8m/s * 0.8uS = 240meters (hope I got the math right).· SO, we are blind for a quarter a km.· Now we have the transit times of some other gear which is VERY slow (on the order of 10uS or so which means we're blind for another 3kms).
So, with a single ranger finder (see the SRF's) you have a minimum distance you can measure.· and you need some pretty fast switching components to reference and measure the pulses....
Might the bistatic approach be easier? (we do it with all IR finders and the SRF's!).
Just need to have somewhat accurate control of firing the laser....
Sorry about the math heavy post and assorted rambling....we ran out of coffee at work here and I eating packets of sugar until lunchtime!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·
Steve
http://members.rogers.com/steve.brady
"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."
The best thing I could find would have been a fibre transceiver!· 2 separate paths for the light (1 in/1 out)....fire your laser diode and mark the start time.· You would now need a fairly stable oscillator (OCXO?) then hopefully you'd be able to detect the pulse on the return path (given the appropriate lenses!).
You'd probably have to find a way to superhet the oscillator or something....but, who knows!
Just a note: working with a weather agency, I get to play with laser ceilometers (range lasers -- lidars -- pointing straight up that give cloud ceiling height).· Keep an eye out for these things at auction.· They're fairly large and clunky but I know we've upgraded almost all of ours with ones that go to 25 and 75km heights (does anyone really care after 5km?!).
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·
Steve
http://members.rogers.com/steve.brady
"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."
The two main categories are TOF(time of flight) similar in operation to what achilles described, and angle of reflection. Two sub categories, visible and IF lasers can be used. The Bushnell product i mentioned used TOF, by pulsing an IR laser(probably a led laser but not certain). That seems to be the most energy efficient way to get a distance meaurement with lasers. They work on a few AAA batteries for hours and hours of continuous use.
Someone is obviously manufacturing modules used in this product and others. Bushnell doesnt seem to be willing to provide this information, or they dont know. I'm going to keep contacting them in hopes that I'll get someone with the right answers. All they were able to tell me is that the warranty would be null if I opened and modified their product hehe.
Steve: I dont have a scope
http://www.jameco.com
page 237 in the nov 2004 cat
laser diodes and driver boards 182932cr 161736cr under $20.
73
spence
k4kep
I've seen a few interesting laser modules for LED lasers. They are definately an integral part of low power laser systems. They can be contained is a small package. Ebay has a bunch too.
Laser rangefinding is not·a DIY project. Some people claim to be able to make one themselves but admit the resulting range might be highly inacurate.
Ideally you would modulate the laser and the receiver could then·analyse·to differentiate it from other possible laser/light sources. It is quite similar to sonar rangefinder operation. It is not likely that anything with any accuracy could be made with a standard laser or pointing device as the laser frequency and whatnot must be strictly controlled.
Much of the attention currently directed to laser technology stems from the idiots pointing lasers at aircraft. What are they thinking? Next pair of sunglasses i get will need to be laser safe
Orion: I was thinking of mounting the rangefinder vertically and tilting/rotating a mirror(optical) 360 degrees to scan continuously. I've seen examples that seemed to do the job. One thing that rules over sonic rangefinding is the ability to bend light. When moving forward an arc of 90 degress could be scanned for example.
Post Edited (Jonb) : 1/26/2005 11:36:04 PM GMT
You could pulse another laser with the stamp.· Then use a beam splitting mirror to redirect a sample of the main beam back in to the receiver of this module.· Then with the appropriate lenses you could "hopefully" pick up the reflected pulse.
There would be an issue with ambient light interference....so would you have to modulate?· Don't know....
The accuracy of this whole thing comes down to a stable oscillator/timer.· 555's wouldn't cut it (I doubt).· And with the fastest BS2(x) stamps timing interval of say 0.75uS (for the BS2P pulsin command)...that's equivalent of traveling over 200meters at the speed of light.
So in order to measure down to 1m, you'd need a clock with a resolution of 0.000000003seconds (0.003uS what's that? 3nS?)
Not an easy task for a DIY....but certainly doable.
Here a thought....and somebody has mentioned one of the ways to determine distance with with geometry....if the CMUcam will indicate the center of a 'glob'....then you could set the camera up to be in the center of your BOT (or module) and level as well.· Then have a stepper move the laser up and down until the 'glob' is centered on in the camera.· a little sin/cos/tan and you should be able to determine the distance from your LEVEL position to that other LEVEL position.
Certainly there become issues when the terrain drops away....but maybe have the whole rig (camera and laser/stepper) on an elevated do-hickey that will swivel down in order for the camera to see lower than level.· More math involved here though...yipes!
Anyhow...Monday morning ideas....by lunch I'll forget them! mm lunch!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·
Steve
http://members.rogers.com/steve.brady
"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."