Reducing the noise introduced by the engine of a moving autonomous vehicle for recording underwater sounds María-Elena Fernández-García ⇑ , Jorge Larrey-Ruiz, José-Luis Sancho-Gómez Tecnologías de la Información y las Comunicaciones, Universidad Politécnica de Cartagena, 30202 Cartagena, Spain article info Article history: Received 10 October 2018 Received in revised form 22 May 2019 Accepted 3 June 2019 Keywords: Underwater acoustics Autonomous Underwater Vehicle Order Analysis Hydrophone recordings Acoustic filtering abstract In this work, a hydrophone has been integrated in a moving autonomous underwater vehicle (AUV) in order to have an autonomous controlled system capable of recording underwater audio. The performance of the system, in terms of best recording of the audio, has been analyzed by placing the hydrophone in two different positions: the front of the AUV and its lower part. From this analysis, it has been determined that the best position for the hydrophone (i.e., the one that receives the lowest engine noise) is the for- mer. In addition, a signal processing based on an order analysis has been performed for the purpose of filtering the noise introduced by the engine of the AUV. This technique achieves the filtering of most of this noise without removing contributions from other sources of interest whose frequency spectrum overlaps with the noise of the engine. This is a fundamental characteristic of the new system because it allows for cleaner underwater recordings. The aim is to increase the capabilities for detection of sources emitting in the range of the AUV acoustic contribution by reducing the same and not so much by com- pletely removing it. Ó 2019 Elsevier Ltd. All rights reserved. 1. Introduction During the last few years, a growing awareness has emerged in the international sphere on the need to protect the marine envi- ronment. As a consequence, different regulations focused on its effective preservation have been developed. Thus, in 2008, the European Commission promulgated the Marine Strategy Frame- work Directive (MSFD), which obliges Member States to develop strategies and programs so that their marine waters can reach and maintain a good environmental status by the year 2020 [1,2]. At the national level, in Spain, the MSFD is structured through Law 41/2010 on the Protection of the Marine Environment [3], as well as through several preceding documents on marine strategy promulgated by the Spanish Ministry of Agriculture and Environ- ment. These initiatives can be understood as part of Horizon 2020 Societal Challenge 2 (SC2): Food security, sustainable agricul- ture and forestry, marine and maritime and inland water research and the bioeconomy. At the regional level, the marine strategy for the Levantine-Baleric demarcation (which includes the coast of Murcia) is of special relevance, since it contains three zones cata- loged as Site of Community Importance (SCI) and one as Specially Protected Area of Mediterranean Importance (SPAMI), mainly due to the high presence of marine mammals. The coast of Murcia is hence located in an area of special biogeographical importance for the entire Atlanto-Mediterranean Region. The effective applica- tion of the MSFD entails the measurement and control of underwa- ter noise, as well as the detection of the presence of marine fauna susceptible to suffering the harmful effects of acoustic pollution. The impact of underwater ambient noise on the environment, fundamentally on marine mammals, and its legislative regulation, require that entities which operate in European seas have tools for the measurement and control of the polluting sources. There exist some precedents, such as the extension of the port of Rotterdam (Maasvlakte 2), where monitoring and analysis of underwater noise were carried out for several years [4–6] to assess both the effects produced on marine life and the predictions made in the corresponding reports on environmental impact. On the basis of all these premises, the idea of developing a sys- tem that allows for the measurement and recording of underwater noise arises. The proposed method is intended to enhance the detection capabilities of some species having their calls on the low frequency range (e.g., mysticetus), and therefore it is not focused on noise measurement within the MSFD framework. Among the different possibilities for deploying underwater audio sensors, the integration of a single hydrophone of reduced size and high performance (Ocean Sonics’ icListen HF 200 [7]) in an AUV (Ocean Server’s IVER2 EcoMapper) is proposed in this work. https://doi.org/10.1016/j.apacoust.2019.06.006 0003-682X/Ó 2019 Elsevier Ltd. All rights reserved. ⇑ Corresponding author. E-mail address: elena.fdez.ga@gmail.com (M.-E. Fernández-García). Applied Acoustics 155 (2019) 250–259 Contents lists available at ScienceDirect Applied Acoustics journal homepage: www.elsevier.com/locate/apacoust