Experimental multi-camera setup for perception of dynamic objects

Currently, perception and manipulation of dynamic objects represent an open research problem. In this paper, we show a proof of concept of a multi-camera robotic setup which is intended to perform coverage of dynamic objects. The system includes a set of RGB-D cameras, which are positioned and oriented to cover the object’s contour as required in terms of visibility. An algorithm of a previous study allows us to minimize and configure the cameras so that collisions and occlusions are avoided. We test the validity of the platform with the Robot Operating System (ROS) in simulations with the software Gazebo and in real experiments with Intel RealSense modules.