Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

line laser

Joined
Jan 21, 2013
Messages
3
Points
0
Hello guys.

I am a student working on a computer vision system and I need some help for a lightweight laser system to be build on a small flying robot.

My idea is to horizontally project 50-100 points (splitter) which are to be seen on a robot's camera. The reason I don't want a line laser is that the intensity falls linearly and a 10-meter reflection won't be visible on the camera. I suppose that the wavelength is not so important and the laser power could compensate for the camera color specificity. I also guess that using a narrow filter (passing only the laser wavelength) in front of the camera would greatly clear the image but it looks much cheaper just to rise the laser power more.

I tested some Chinese 5mW 650nm laser with my current CMOS 720p camera. The point is visible from 10-15m so a 500mW laser divided on 100 beams looks reasonable without using any filters.

I don't know what and where to order because of the following problems:
-- I need a ~90 degrees of divergence but the Frankfurt laser (from above link) gives only ~20 degrees. Is there wider splitters or is it tricky to use lens in front of the splitter?
-- The "designed wavelength" of the splitter is written to be 633nm. Is it possible to use different wavelengths and what changes in such case?
-- Frankfurt lasers don't seem to be the cheapest but I hardly find alternatives;
-- I don't know how to choose compatible parts and how to arrange them (laser module + splitter + lens(?) + filter(?) + camera).

I will appreciate any suggestions about the design of the system as well as the shops to order from.

Thank you,
Petar Ivanov
 





Joined
Dec 31, 2012
Messages
904
Points
0
If the laser is to be mounted ON the flying robot, then the laser type, strength and run-time may be limited due to weight constraints. The payload capacity of flying robots is generally pretty limited. Although, I recall ロボット(Giant Robot) of Japanese movie fame had a rather large payload capacity, so your mileage may vary.

The 5mW laser diodes found in so many laser pointers can run for a long time without overheating. But as you increase output power, you introduce the need for cooling which involves a heat seat. And a heat sink will add (relatively) considerable weight.

Higher output power is usually accompanied by greater power requirements, meaning a larger heavier battery.

With regard to diffraction gratings, it is my understanding that 90-degrees is the absolute limit of what a grating is capable of producing. Also, the greater the number of orders (producing a wider dispersion) asked by the grating, the lesser its efficiency. If you want a 90-degree spread, you may find you’ll need even more output power from your laser to compensate. The following website discusses this 90-degree limit in more detail.
Optometrics: Diffraction Gratings

What is the robot and (presumed) stationary cameras using the laser light for? If it is positional telemetry, there has been some success in reversing the role, and shooting positional data from stationary positions, and placing the receptor on to the robot. The robot may then, if that is your aim, know where it is positioned in the arena. The following TED video is rather interesting and discusses this system in greater detail.
Vijay Kumar: Robots that fly ... and cooperate | Video on TED.com

At 6:54, the positional telemetry sent to the robot is mentioned only briefly.
At 12:06, they show a flying robot with a laser and a scanner, to map out the environment. It’s pretty cool.

The whole video is pretty cool. Though, I concede, there isn’t a great deal of laser content.
 
Joined
Jan 21, 2013
Messages
3
Points
0
Thank you very much for the answer, Pschlosser!

You are right that the payload is a significant problem. We are using AR.Drone 2.0 and the weight limit is about 100 grams for a stable flight.

I want to use a laser for enriching the texture of the scene in case of low texture objects. This way the telemetry calculation must be stable enough.

The video is really cool. The reason I am not using Kinect is that it works up to 4-6m. And the reason I am not using a laser scanner is its cost (>$1000) and weight (>130gram). It is generally needed to sustain an alternative technology so I don't think it is very pity I am not using these two.

As the heat of the laser is also a problem, I suppose that the filter is necessary for keeping the power low. If a filter can compensate for 2 orders of laser power, then it looks possible to split the beam to 50+ beams. Do you know what is the maximal power of a laser that can be used for at least 10min without cooling? Is there a problem with using lens in front of the splitter?

Cheers,
Petar
 
Joined
Dec 31, 2012
Messages
904
Points
0
Do you know what is the maximal power of a laser that can be used for at least 10min without cooling? Is there a problem with using lens in front of the splitter?

I don't know. My guess is <50mW, since many laser pointer pens can vary in output and emit >5mW, even though they advertise <=5mW. These pens are often measured in the 10mW to 50mW range. And these pens seem to operate continuously without much trouble. I am referring to the variety that runs on 2x AAA batteries. Stay in the red laser diode range, and I think you should be okay.

And a lens in front of the diffraction grating? I don't see any problem with that. Several 50mW red laser pointers discussed on this forum come with a focusing lens made of acrylic. However, the focusing knob and the accompanying threaded cylinder, into which the diode is often pressed, are generally made of metal, and that increases weight.

You may need to experiment to test the feasibility of using a laser diode on the drone. Cutting away some of the external foam from the drone may increase it payload.
 
Joined
Jan 21, 2013
Messages
3
Points
0
Pschlosser, thank you very much for enlightening me the basics.

I contacted with the sales manager of Frankfurt lasers and I got disappointed about their unwillingness in choosing or suggesting. This motivated me to take a look at all the Chinese shops I know for (cheap) alternatives. The biggest choice of lasers and filters I found on AliExpress so I write down what lasers, filters and graters caught my eye.

[the parts follow the order the photons go from the laser to the camera]

Dot 650nm laser modules:
250mW
200mW
100mW
50mW

Laser graters:
5-25% uniformity of the beam intensity, 19 dots and 99 dots in a line; ~15-20 degrees of pattern divergent
Dragonlasers beam fan
Starfield Diffraction Grating 500 lines/mm

Lens:
TODO

Camera filter:
Center at 650nm, 25nm Half bandwidth

Camera:
TODO

Does it look reasonable to you?

Can I expect that there won't be any serious problems with combining a greater from Frankfurt Lasers with a random Chinese laser? Frankfurt Lasers state that the input beam can be "any collimated (or nearly collimated) laser beam or white light source larger then 100µm and smaller than diffractive element aperture" and in the same time "large area input beam (e.g. 5 mm beam) will have overlapping points until the pattern has propagated enough to separate the points by 10 mm". (It must be that the diffractive element is big enough). Unfortunately, I couldn't find any other feasible line graters. Do you think it is okay to ask in a different thread about graters?

Thanks,
Petar
 
Last edited:




Top