Robotic sensor networks for security

The development of intelligent surveillance systems is an active research area. In this context, mobile and multifunctional robots have been recently adopted as successful means to reduce fixed installations and the number of devices needed to cover a given area. On the other hand, modern techniques for data fusion and decision making can significantly increase the information content extracted from sensors both mounted on the robots and on the infrastructure. The use of many heterogeneous sensors, the number and complexity of operational tasks required for monitoring and surveillance with autonomous components like robots makes the overall system design very challenging. In this paper we present some ideas and investigations ongoing in SELEX Sistemi Integrati to assess the capability of such a kind of robots-sensors systems to improve the monitoring of large and densely populated indoor areas. In particular a discussion on some of the problems arising in robot guidance and navigation, oriented to the reduction of missed and false alarms is firstly carried out. Some numerical simulations are reported to support the proposed investigations. Follows the description of a possible decision fusion algorithm to identify and track "risky" targets in a dynamic environment with the aid of robots.