backend
backend copied to clipboard
3d-Room recognition
It is possible to recognize 3d-environments using a simple camera or even simpler sensors. Here are some links:
- https://en.wikipedia.org/wiki/Simultaneous_localization_and_mapping
- http://ccwu.me/vsfm/
- https://en.wikipedia.org/wiki/Bundle_adjustment
- http://docs.opencv.org/2.4/modules/stitching/doc/motion_estimation.html#detail-bundleadjusterbase
- https://www.google.com/atap/project-tango/
- http://www.pirobot.org/blog/0015/
This should allow to recognize walls, floors, obstacles and such.
I also develop robots and would like to get in touch with you, if you want to talk about the projects, see if we may be able to merge efforts.
Hello Nicco, it would awesome to develop a SLAM algorithm in an inexpensive robot like CoderBot. As far as I know, the SLAM algo for a single camera robot is possible, but I couldn't fine a open source, documented, project to start with. It's a tough problem to solve, but the outcome is amazing, so let's join efforts!