Consensus algorithms in a multi-agent framework to solve PTZ camera reconfiguration in UAVs

Abstract

A method for PTZ camera reconfiguration is presented. The objective of this work is to improve target tracking and surveillance applications in unmanned vehicles. Pan, tilt, and zoom configurations are computed transforming the visual constraints, given by a model of visual coverage, into geometric constraints. In the case of multiple targets the camera configurations are computed by a consensus algorithm. The approach is defined in a multi-agent framework allowing for scalability of the system, and cooperation between the cameras. Experimental results show the performance of the approach. © Springer-Verlag Berlin Heidelberg 2012.