Civilian Harbour Protection: Interception of Suspect Vessels with Unmanned Surface Vehicles ⋆ E. Simetti * A. Turetta * S. Torelli * G. Casalino * * DIST - University of Genova, 16145 Genova, Italy (e-mail: {simetti, turetta, torelli, casalino}@dist.unige.it). Abstract: In the last decade, the security of civilian transportation hubs has received an increasing focus. Civilian harbors are no exception, and to this aim Unmanned Surface Vehicles could be exploited for patrolling purposes and to investigate suspect situations, increasing the effectiveness of the harbor protection system while lowering the number of humans directly exposed to threats. The use of a team of USVs for the security of civilian harbors is the subject of an on-going research project between DIST and Selex-SI, a leading player for homeland security systems. This paper presents the latest results of the project, which are centered on the problem of intercepting a suspect vehicle. Both the off-line optimization of the positioning of the USV fleet and the on-line selection of the most suitable USV are tackled. Keywords: Marine systems, Autonomous vehicles, Optimization, Path planning, Obstacle avoidance 1. INTRODUCTION The problem of maintaining civilian harbours safeguarded against terroristic attacks, coming from the so-called “blue border” (i.e. the sea-side), is receiving an increasing inter- est, especially after September 11th. To this aim, the need for innovative and reliable technologies, improving the cur- rent harbour security infrastructures, has been expressed by different governments and international institutions. In this context, the use of a team of “protecting” Un- manned Surface Vehicles (USVs) certainly represents a promising solution for reducing the harbour vulnerability: indeed under normal conditions, the USVs can perform patrolling surveys of the more crucial waterways, provid- ing a remote operator with acoustic and/or optic images acquired through the on-board sensory equipments; when- ever instead a possible “menace” (i.e. an unauthorized vessel or a vessel moving in a suspect way, such as too fast or pointing towards restricted areas) is detected by the global security system, one of the USVs can be used for “intercepting” the menace (i.e. moving fast toward it), allowing to determine whether the suspect vessel is “hostile” or “friend” without exposing humans directly to threats. Previous experiments on USVs were mainly performed in open sea or in waterways in the absence of other unknown moving vessels (see Veers and Bertram (2006); Caccia (2006) for nice surveys on the topic). However civilian harbours are far different from those scenario and more crucial issues arise: the ship traffic is intense and the operations of tourist or merchant ships ⋆ This research was partially funded by EU FP7 project “CO3AUVs: Cognitive Cooperative Control of Autonomous Underwater Vehi- cles”, under Grant Agreement n. 23 1378. cannot be delayed or affected anyway by the security vehicles. Therefore the manoeuvres of USVs must not perturb at all the normal harbour activities. Further there is a concrete risk of collision with other vessels, with consequent risks of personal injuries and property damages. The above safety considerations have a clear impact on the required characteristics of the overall USVs-based security system. As a first consequence, the USVs have to be provided with good path-following capabilities, since they need to follow a reference path with a certain accuracy. Moreover, they need reliable emergency sensory devices enabling a prompt detection of any incipient obstacle, and suitable techniques for implementing reactive obstacle avoidance capabilities, in case an unforeseen obstacle is actually detected. Many of the above topics have been investigated and interesting results have been recently proposed in Larson et al. (2006); Healey et al. (2007); Caccia et al. (2008); Bibuli et al. (2009). Despite for some of the above issues consolidated results are not yet available Caccia (2006), for the sake of the present work the considered USVs will be assumed totally compliant with all the above requirements. Even under the hypothesis of almost-perfect USVs, the need of efficiently coordinating their operations still re- mains an important issue to be solved before obtaining an effective and reliable USVs-based harbour security system. In this perspective, DIST (University of Genova, Italy) and Selex Sistemi Integrati (a Finmeccanica Company, Italy), one of the international leading players in providing large systems for security and defence, are cooperating within an on-going joint research project, on the realization of the so- called Swarm Management Unit (SMU), a tool conceived for supervising the operations of a team of USVs per- 9th IFAC Conference on Manoeuvring and Control of Marine Craft, 2012 The International Federation of Automatic Control September 19-21, 2012. Arenzano, Italy ©2012 IFAC 10.3182/20120919-3-IT-2046.00074 435