3d Semantic Reconnaissance

October 2013 - October 2014
Levente Tamas

Within this project a novel approach in the field of scene understanding for mobile robots in urban outdoor scenarios based on fused 3D information from several sensors (lidar, color camera, time of flight camera, thermal ones) is investigated. The 3D perception is essential in real life application for autonomous robots such as the ones used in search and rescue activities, surveillance, guidance etc.

For the theoretical investigations and implementation of the proposed method an important aspect is the experimental validation of the proposed approach on a real robot (iRobot Packbot) in an outdoor environment such as the participation at ELROB 2014 in Warsaw, Poland.

Additional Links

video of bebot in urban reconnaissance trial of ELROB 2014
Proof of concept for indoor 3D object classification using a Swiss Ranger 4k camera and the VFH feature descriptors.
Object detection demo with PFH features, 3D data from Swiss ranger 4k camera
3D object detection with PFH using a Swiss Ranger 4k camera