Drivable path detection for a mobile robot with differential drive using a deep Learning based segmentation method for indoor navigation
Oğuz Mısır · 2024
Read Drivable path detection for a mobile robot with differential drive using a deep Learning based segmentation method for indoor navigation by Oğuz Mısır (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11639217.txt.
Description
Oğuz Mısır presents in 2024 a vision system that lets an indoor robot see where it may safely drive. The work leans on deep learning image segmentation: each camera frame is painted into floor versus obstacle at pixel level. A strong known pairing, a segmentation network on a deep backbone, does the painting with high accuracy.
Raw paintings are then warped by a perspective transform into real world floor coordinates, and the floor is cut into a grid. On that grid a sampling based planner, an improved random tree method, threads a route to the target that dodges obstacles. A smoothing pass trims needless turns so the differential drive robot rolls a natural line. The chain from pixels to path ran in tests as one pipeline. Careful seeing first, the author argues, is what makes the later planning safe and close to optimal.
Catalog record
| CREATOR | Oğuz Mısır |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 74 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |