Autonomous underwater vehicles powered by fuel cells: Design guidelines Rafael d’Amore-Domenech a , Miguel A. Raso b , Antonio Villalba-Herreros a ,  Oscar Santiago a, c , Emilio Navarro c , Teresa J. Leo a, * a Dept. Arquitectura, Construcci on y Sistemas Oce anicos y Navales, ETSI Navales, Universidad Polit ecnica de Madrid, Avenida de la Memoria 4, Madrid 28040, Spain b Dept. Química Física I, Facultad de C.C. Químicas, Universidad Complutense de Madrid, Plz. de Ciencias 2, Madrid 28040, Spain c Dept. Fluid Mechanics and Aerospace Propulsion, ETS Ingeniería Aeron autica y del Espacio, Universidad Polit ecnica de Madrid, Plz. Cardenal Cisneros 3, Madrid 28040, Spain ARTICLE INFO Keywords: Fuel cell marine application AUV design Energy density Power density Hydrogen storage Principal component analysis ABSTRACT Current commercial Autonomous Underwater Vehicles (AUVs) are powered by conventional batteries. However, such technology has reached a point at which an increase of the endurance and range of operations would require increasing the size of existing AUV designs. Much attention has been paid to fuel cells as they have proven successfully installable in AUV prototypes and have shown good results regarding the increase of range and endurance of AUVs. Nevertheless, no commercial AUV powered by fuel cells has been made yet. In this work, the characteristics of well-known commercial AUVs have been studied using the Principal Components Analysis. Such study allows inferring the requirements and constraints for the implementation of fuel cells in commercial AUVs. Freedom in certain design parameters of both fuel cells and AUVs has been found. Regarding the necessity to store oxidant along with the fuel in underwater applications, a map of achieved energy densities by the combination of different means of storing hydrogen and oxygen has been obtained. Such map reveals the most benefiting combination of storage means for both reactants. Finally, a method for obtaining the initial design parameters of fuel cell powered AUVs is proposed. 1. Introduction Thanks to the great development achieved in these last years, un- manned vehicles have evolved from the research laboratories into com- mercial, military, and scientific applications (Vukic, 2013). Although unmanned marine vehicles were developed by the military industry, they are revolutionizing the access to the oceans in the civil industry, as they are allowing to reach depths beyond the limits of scuba divers (Leonard et al., 1998). The main reasons for using underwater unmanned vehicles (UUVs) are that they reduce human presence for safety, costs or acoustic signature reasons, and they increase operational ranges. Two different types of UUVs can be distinguished, regarding whether UUVs are remotely piloted or not, being both able to develop multitude of missions (Wang et al., 2012):  Remotely Operated Vehicles (ROVs) are usually employed for deep sea offshore tasks, as substitutes of scuba divers.  Autonomous Underwater Vehicles (AUVs) are usually employed in offshore industry for obtaining information about subsea environ- ment. They operate without any physical or electronic interaction with human beings, and therefore are able to navigate by themselves (National Oceanic and Atmospheric Administration, 2013; Alam et al., 2014). Whereas AUVs need media to store energy on board for the propul- sion and all the systems, ROVs normally do not need them, as they are connected to a mother ship, although sometimes they might have energy storing media to reduce their cable section (Brown, ). Current commercial AUVs use batteries for storing energy (ECA Group, 2017; Alam et al., 2014; Kongsberg, 2012a, 2012b, 2013, 2015, 2016, 2017a; Elektronik, 2012, 2013; Bluefin Robotics, 2016a, 2016b, 2016c, 2016d, 2017a, 2017b; Gavia, 2016; ECA Group, 2016; ECA Group, 2017a; ECA Group, 2017b; ECA Group, 2017c; ECA Group, 2017d; ECA Group, 2017e). Their main drawback is that current technology has reached a maturity state at which very small improvement is expected regarding their energy densities. This means that at the current state, battery pow- ered AUVs with extended range and endurance are set to be bigger than normal range and endurance AUVs. However, bigger AUVs are less prac- tical as they require heavier duty cranes to put them afloat and need more space to maneuver. The main challenge addressed by this article is to * Corresponding author. E-mail address: teresa.leo.mena@upm.es (T.J. Leo). Contents lists available at ScienceDirect Ocean Engineering journal homepage: www.elsevier.com/locate/oceaneng https://doi.org/10.1016/j.oceaneng.2018.01.117 Received 24 April 2017; Received in revised form 30 January 2018; Accepted 31 January 2018 0029-8018/© 2018 Elsevier Ltd. All rights reserved. Ocean Engineering 153 (2018) 387–398