Obstacle Detection and Tracking for Safe Human-Robot Interaction Based on Multi-Sensory Point Clouds

Conference paper


Angelika Fetzner
Christian Frese
Christian Frey


Proceedings of the Workshop on Human-Friendly Robotics, 2013.


6th International Workshop on Human-Friendly Robotics (HFR), Rome, September 25 - 26, 2013

For close human-robot interaction without physical separation, safety has to be ensured by monitoring the humans while they are within the robot workspace. Here we discuss the problem of selecting and placing a suitable set of sensors for mobile manipulators. Furthermore, methods for detection and tracking of dynamic obstacles based on data from heterogeneous sensors are presented. The approach is validated by experimental results on real sensor data acquired by the robot.