Development of an AUV control architecture based on systems engineering concepts Luciano O. Freire a , Lucas M. Oliveira a , Rodrigo T.S. Vale a , Mai a Medeiros b , Rodrigo E.Y. Diana a , Rubens M. Lopes b , Eduardo L. Pellini c , Ettore A. de Barros a, * a Unmanned Vehicles Laboratory (LVNT), Mechatronics Engineering Department, Polytechnic School, University of S~ ao Paulo. Av. Prof. Professor Mello Moraes, 2231, 05508-030, S~ ao Paulo, SP, Brazil b Laboratory of Plankton Systems, Oceanographic Institute, University of S~ ao Paulo. Praça do Oceanogr afico, 191, 05508-120, S~ ao Paulo, SP, Brazil c Lab. of Research on Protection and Automation of Electrical Systems (L.PROT), Department of Electrical Energy and Automation, Polytechnic School, University of S~ ao Paulo. Av. Prof. Luciano Gualberto, trav. 3, 158, 05508-010, S~ ao Paulo, SP, Brazil ARTICLE INFO Keywords: AUV CAN Control architecture Systems engineering Underwater vehicles ABSTRACT The very nature of a complex system does not allow that a single person be able to master all required compe- tences for its development or operation. Therefore, specialization, team work and coordination are required to achieve the desired goals. In order to assure positive synergy between every participant, system engineering concepts must be taken into account, like development methods, product decomposition, functional classification, design patterns, interface and compatibility assessment, configuration management, technological plans, tech- nical standards, integration policies, system commissioning and validation. These systems engineering concepts are shown in the presented work through the development of a fully functional AUV system and control archi- tecture. The control system and its components are properly described and compared against other state of the art architectures. It is also shown that it was possible to sustain the project development tasks among many successive group or generation of students, demonstrating the benefits of the proposed engineering methods. The proposed system was tested in field tests of the AUV during an oceanographic mission. 1. Introduction Research institutions working with complex systems typically have developed informal rules to allow parallel work of many researchers, and assure positive synergy between co-workers. Those rules are an essential part of the intellectual capital of the institution, yet they are typically neglected in the academic world, which focus on the physical phenomena themselves. Therefore, new departments or new institutions may face research continuity problems regardless of the academic excellence of the participants, because they do not know how to organize the institutional work to achieve wider and permanent goals. Similar problems are faced by other institutions developing complex systems, like NASA, which published its Systems Engineering Handbook (NASA, 2007). This publication provides general definitions, best prac- tices, guidelines and alternative approaches in product development, especially for long-lasting life cycles, with many intermediary milestones. In parallel, the Software Engineering Institute (CMMI, 2010) proposes a framework of tools, to improve product development and predictability of results. For questions related to safety, the generic IEC 61508 standard also discuss some important concepts that should be considered. A number of works have presented control architectures for mobile robots in general, that influenced also the AUV embedded system design. In many cases, the architecture main characteristic is rooted on some artificial intelligence paradigm, such as the deliberative approach (Nilsson, 1969; Bowen et al., 1990), reactive or behaviour-based control (Brooks, 1986; Kumar and Stover, 2000; Bellinghamet al, 1994), and other approaches biologically inspired (Arkin, 1990). More recently, hybrid approaches, combining deliberative and reactive control appli- cations are becoming usual in the robotics community (Gat, 1998; Brutzman et al., 1998; Valavanis et al., 1997; Palomeras et al., 2012; Sheikh et al., 2014; Ranganathan et al., 2001; Goldberg, 2011; Müller, 1996). On the other hand, with the advance of computer science, hardware resources, software tools and frameworks, the focus on organization of software with possibilities to combine the different paradigms (Hewitt and Inman, 1991; Kim and Yuh, 2004; Amianti and de Barros, 2008; * Corresponding author. E-mail address: eabarros@usp.br (E.A. de Barros). Contents lists available at ScienceDirect Ocean Engineering journal homepage: www.elsevier.com/locate/oceaneng https://doi.org/10.1016/j.oceaneng.2018.01.016 Received 27 October 2016; Received in revised form 7 November 2017; Accepted 3 January 2018 0029-8018/© 2018 Elsevier Ltd. All rights reserved. Ocean Engineering 151 (2018) 157–169