I am in possesion of two XV-11 LiDAR units. / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

I am in possesion of two XV-11 LiDAR units.

Move Thread

LAN_403

Patrick

Fri May 17 2013, 06:12PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

i have two lidars.
XV-11, neato robotic vacuum cleaners.

each one weighs in at 192 grams.
overall height is 2.2 inches.
width is 3.8 inches.
length is 4.6 inches.
range is 6 meters (or 19.5 feet.)

im thinking of using Parallax Serial Terminal, or Termite.

data rate is 115.2 kbps, RPM is 5 to 10 Hz, resolution is 360 points per rotation (1 Deg) and sequential unlike the Hykoyuo's. This device uses the DLIS-2k linear sensor, which I wish I could buy. And a IR dot laser. Im planning to use my logic sniffer to verify there working.

source sites:

wiki lidar, for supporting external devices.

motor control for Lidar.

Carbon_Rod

Sat May 18 2013, 06:10AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

These are not LIDARs, but distance sensors which operate on a similar principle of the Sharp GP2Y0A02YK0F.
(IIRC, these had a CCD element inside similar to old desktop optical scanners)

The repeatable measured resolution one would need for SLAM will likely be difficult to achieve.

Best of luck,
Rod

Patrick

Sat May 18 2013, 01:46PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Carbon_Rod wrote ...

These are not LIDARs, but distance sensors which operate on a similar principle of the Sharp GP2Y0A02YK0F.
(IIRC, these had a CCD element inside similar to old desktop optical scanners)

close enough, using laser light to measure distance = a lidar, in my book.

Carbon_Rod wrote ...

The repeatable measured resolution one would need for SLAM will likely be difficult to achieve.

nope, the vacuum navigates just fine, my C-slam (input) isn't more demanding than the original Vacuum claeaners'.

in fact the C-slam [source, onSTM32F4] worries me more than the lidar at this point. (I may need some forum members to help figure / advise me on how to implement the C-SLAM source)

my speed shoud be a game changer in the compititon, the hykoyos are expensive, and all the ones ive seen use USB, so Theres a support PC on board taking the lidar+USB, then transmitting wirelessly to ground PC base station, then SLAM and decision making made on the ground, with results being transmitted back to the craft in flight.

in the mean time due to latency, even the high $ Michigan university team, has crashed face first into walls, due to long loop time.

my system will be more primitive than the 6500 US$ Hykoyou, but entirely an embedded solution. (remember USB is evil, if you don't have a custom driver, or an existing one. I was told by an ace programmer and MS/PhD group, don't even try to write your own USB driver, or youll be psychotic soon enough.)

Carbon_Rod

Sun May 19 2013, 02:01AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Some $1.7k models by Hokuyo had a regular rs232 on the same plug as the power inputs.

SLAM is resource intensive, and I doubt an STM32F4 kit will have the capacity to do this successfully....

USB is easy if your board has a real operating system with a host stack....

Commonly, people's goal â€œto winâ€ often gets misinterpreted as â€œbuilding a better sensor platformâ€.

Sometimes complexity loses the challenge before it even starts:

Patrick

Sun May 19 2013, 02:21AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Carbon_Rod wrote ...

Some $1.7k models by Hokuyo had a regular rs232 on the same plug as the power inputs.

yes, but I don't think those are made any more, and they seem to go fast when they appear on ebay and such.

Carbon_Rod wrote ...

SLAM is resource intensive, and I doubt an STM32F4 kit will have the capacity to do this successfully....

I worry about this fact every night before I go to bed.

Carbon_Rod wrote ...

USB is easy if your board has a real operating system with a host stack....

only if the original device maker, wrote the USB driver, and its for a compatible system, the other groups use normal win or Linux driver on the "GumSticks" type computers.

Carbon_Rod wrote ...

Commonly, people's goal â€œto winâ€ often gets misinterpreted as â€œbuilding a better sensor platformâ€.

Sometimes complexity loses the challenge before it even starts:

I am open to better solutions, and would beg anyone for a solution that avoids traditional slam, in fact im wondering if I should just "hand hold" the wall all the way around the course.

EDIT: I must admit, maze theory does look interesting.... however, attrition and inefficiency of path planning worry me.

EDIT: heres a graphical example of the office space we must navigate, the "hostile foreign nations center for torture and ministry of defense".
maze

the drone takes off and lands at the red window, but the flash drive is the blue spot.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob