Robust shelf detection from 3D point cloud data
Motivation and scope
The goal of this project is to improve on a recently developed method for shelf detection for mobile robots in a warehouse environments. Given a 3D map, the robot should automatically label all the shelves in the warehouse. Automatically detecting shelves is useful because it makes deploying robots easier, since the user does not have to manually annotate all points of interest in the map.
Specific tasks may include investigating different methods for detecting the parts that make up a shelf, as well as geometric verification methods to make sure that the output detections are consistent. Additional tasks may include long-term fine-adjustments of the shelf poses as the robot drives around.
Good programming skills, preferrably in C++. Being comfortable with 3D geometry processing is also an advantage.
Develop experience with 3D perception, and interact with ongoing research projects on industrial mobile robots.