2
$\begingroup$

I am working with SICK lidars and have to mount/unmount them quite often on my robot. The mounting process is very tedious especially when it comes to making sure that the lidars are horizontal. I thought about using IR goggles (like the night vision ones) and some fog machine (like the one in nightclubs) in order to see the surface covered by the lidar's rotating laser ray. As a result I would expect to see something like this but planar.

Before thinking about trying to get my hands on such hardware I wanted to ask:

Do sick laser have enough intensity to be observed by such goggles? Does anybody tried such an approach?

Improve this question
$\endgroup$
5
  • $\begingroup$ I assume these are infra red - have you thought about using any digital camera (still or video) in a darkened room? I just guess it might work... $\endgroup$
    – Andy
    Commented Apr 20, 2016 at 16:21
  • 1
    $\begingroup$ Add a spirit level to it? $\endgroup$
    – Bending Unit 22
    Commented Apr 20, 2016 at 18:50
  • 1
    $\begingroup$ You can see lidar rays with proper equipment - see this video (I guess they use some kind of night vision): youtu.be/cc15Ox8UzEw?t=58s $\endgroup$
    – Kozuch
    Commented Apr 25, 2016 at 10:52
  • $\begingroup$ It sounds like you're trying to mount the LIDAR in such a way that you avoid finding the extrinsic calibration of the LIDAR with respect to the mount point. The way that you propose to do this sounds like it is more effort than it is worth. Instead, look up existing techniques for calibrating LIDARs and apply the inverse transform to account for this angular offset. $\endgroup$
    – Gouda
    Commented Aug 5, 2016 at 13:01
  • $\begingroup$ @Gouda due to the nature of 2d lidars, it is very hard to calibrate tilt (non horizontal lidars), I implemented some methods to compute the pitch and roll angles of a given lidar but the conclusion was that the lidar had to be mounted as horizontal as possible to avoid inconsistencies in the observations. The extrinsic calibration of lidar x, y, and yaw relative to the robot it is mounted on will totally fail if the other two angles are not zero. $\endgroup$
    – Mehdi
    Commented Aug 5, 2016 at 13:50

1 Answer 1

Reset to default
1
$\begingroup$

A digital camera, even on a cell phone will pick up the IR light. A darkened room will help it to show more prominently. No sense in buying any crazy equipment when the phone in your pocket will probably do what you want.

Improve this answer
$\endgroup$
2
  • 1
    $\begingroup$ +1 for going simple first, but I'd point out that some digital cameras have an IR filter over the aperture to prevent IR light from interfering with the image. A quick check would be to see if a TV remote LED lights up when you point it at the camera and push buttons. $\endgroup$
    – Chuck ♦
    Commented Aug 5, 2016 at 12:05
  • 1
    $\begingroup$ I ended up using a camera (without IR filter). The problem was that the sensitivity of most CMOS sensors drops to 10% at the wavelength of the lidars (904 nm). So I mounted a low-pass filter that filters everything under 800 nm. With this the lidar signal can be seen if the lidar is not too far from the wall it is projecting into. $\endgroup$
    – Mehdi
    Commented Aug 5, 2016 at 13:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.