This is a 3D scanning and mapping test for my ERP. Its basically mapping obstacles, then choosing the most free path. It scans in only 30 points, so its fairly poor resolution.
To learn more about Sharp IR rangefinders, go here:
http://www.societyofrobots.com/sensors_sharpirrange.shtml
And for more on the robot, ERP:
http://www.societyofrobots.com/robot_ERP.shtml
@shebotnov Just those links, really. The source code is all available for download.
palmisano 11 months ago
@palmisano thats what I thought =) Thx. Didnt you write about this algo anywhere? Forums, site etc ?
Iam very interested in it.
shebotnov 11 months ago
@shebotnov Ah! Well, the ERP takes multiple 2D scans and treats each as 2D. But as each scan is at a different head angle, it is in effect a 3D scan. (a 3D scan using a 2D algorithm, if that makes sense)
palmisano 11 months ago
@palmisano Already did it yesterday :) You write about 2d mapping using array and wave front alg. You write then that 3d mapping is done with expensive 3d LIDARs. Thats why Iam curios how you do 3d mapping with simple IR finders.
shebotnov 11 months ago
@shebotnov See the links in the info box for more info, and a tutorial.
palmisano 11 months ago
@palmisano Is there a map I could see or an array where you store the data. Anywhere I get better description of your mapping algorithm? I will build my 1. robot soon and want it to 2d map my room. I would be very interested to know how you do it in 3d.
Serg
shebotnov 11 months ago
@shebotnov It's taking measurements of 3D space, meaning locating objects in X, Y, and Z.
palmisano 11 months ago
Why do you call it 3d scanning if its only 2d. Or am I mistaked?
shebotnov 11 months ago
you must fix the starting
technology49 1 year ago
@DarkmoonUK nah, a compass is very inaccurate inside like this. Yeap, I program in C. And yea, its very inefficient scanning, but its just a test for something more advanced later on.
palmisano 1 year ago