Systems, Robotics & Vision Group

We are delighted to report that, between May and July this year, three of our fellow researchers have defended their doctoral theses. Miguel Martín, Joan Pep Company and Eric Guerrero have worked for several years in our laboratories as trainee researchers. During this time they have participated in different research projects while working on their doctoral theses. Happily, their efforts have paid off and, within days of each other, all three researchers have obtained their PhD. We are sure that a successful career awaits them. Congratulations!

---

• Miguel Martín Abadal
  Efficient implementation of Deep Nets for video processing to preserve marine ecosystem services
  Supervisor: Dra. Yolanda González Cid
  May 25th, 2023
• Juan Pep Company Córcoles
  Towards Robust Visual SLAM in Low-Textured Environments
  Supervisors: Dr. Alberto Ortiz Rodríguez and Dr. Emilio García Fidalgo
  June 23th, 2023
• Eric Guerrero Font
  Data-Driven Exploration of Benthic Environments with Autonomous Underwater Vehicles featuring Semantic Perception and Adaptive Navigation Intelligence.
  Supervisors: Dr. Gabriel Oliver Codina and Dr. Francisco Jesús Bonin Font
  July 7th, 2023

Today, 28/06/2023, field test in Port d’Andratx with the two marine robots and the stereo camera for the Plome lander. Follow us also in Linkedin !!!

On 19-20 June, the Marine Robotic Team of the SRV presented two talks at the International Workshop on MARine TECHnology, held at the University Jaume I, in Castellón de la Plana, Spain. Pol Baños and Fco Bonin presented their latest work in Underwater Mosaicing Improvements and Study of Nephrops Habitats using Deep Learning. The SRV representatives also announced the venue of the next edition of MARTECH 2025, to be held in Palma de Mallorca.

We look forward to seeing you all in Palma at MARTECH 2025 !!!!!

The Marine Robotics Team of the SRV gave 3 talks in the last edition of the IEEE Oceans Conference, held between 5-8 of june in Limerick, Ireland (https://limerick23.oceansconference.org/). The conference is the perfect forum to disseminate advances in oceanic enginering, technology, research and ideas, in general, concerning marine sciences. Antoni Burguera Burguera and Fco Bonin Font talked about image compression for underwater acoustic transmission and the application of Convolutional Neural Networks applied to detect fish fauna under the sea, all framed in the COOPERAMOS and PLOME projects, in which the SRV Group is involved.

The Master’s theses of Bo Miquel Nordfeldt-Fiol and Mateu Jover Mulet were awarded 1st and 2nd place in the “IV Convocatòria dels Premis Agenda 2030 per a la Transformació Social i el Desenvolupament Humà Sostenible per a TFG, TFM i tesis doctorals de la UIB”. The first Master’s thesis awarded is framed within the DETECPOS and COOPERAMOS projects, in which the Marine Robotics Team participates, and advances in the improvement of visual odometers for the navigation of AUVs in areas colonised by Posidonia oceanica. The Master thesis developed by Mateu Mulet is framed in the work of the Systems team and presents a tool set to improve the fault tolerance in automatized systems of the electrical networks.

The article entitled “Migrating Legacy Ethernet-Based Traffic with Spatial Redundancy to TSN networks”, from the research group in Systems, Robotics and Vision of the Department of Mathematics and Informatics of the UIB, has been awarded the Factory Automation Best Paper Award at the 27th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA 2022), an international congress held in Stuttgart, Germany, from September 6 to 9, 2022.

ETFA is one of the main conferences of the Industrial Electronics Society (IES) of the IEEE. In this conference, experts from academia and industry gather to disseminate, discuss and encourage the development and adoption of new methods, models and tools for the design and operation of automated manufacturing systems and automation systems in general.

The awarded article extends a set of tools designed to contribute to the adoption of a new technology of industrial communication networks based on Ethernet and known by the name Time-Sensitive Networking (TSN). Industry and academia are making great efforts to adopt TSN, as this technology will not only improve the real-time response and reliability of these industrial networks, but will also enable the integration of automation with information systems in the context of the Internet of Things (IoT). This integration is essential to enable new applications and digital services, such as intelligent automation systems in smart cities; intelligent electricity production and distribution networks (smart grids), or intelligent, connected and flexible factories (smart factories).

The transition to TSN must be gradual and it is of great importance that new automation installations based on TSN can be designed as easily as possible from existing installations based on old network technologies. The article takes as its starting point a set of tools that, in order to facilitate this transition to TSN of facilities, transforms traffic from legacy networks to TSN traffic. Unfortunately, this toolkit was not able to do this transformation when the traffic was redundant (when the traffic had replicated messages). This limitation was very relevant, given that many automation installations are highly reliable and critical, and need replicated messages to ensure that the different parts of the system exchange information correctly. The article extends this toolkit with a series of mechanisms that make it possible to transform this traffic containing replicated messages and therefore represents a significant step towards being able to adopt TSN in critical automation facilities.

Article reference:

M. Jover; M. Barranco; I. Álvarez i J. Proenza. «Migrating Legacy Ethernet-based Traffic with Spatial Redundancy to TSN». Proceedings of the 27th International Conference on Emerging Technologies and Factory Automation (ETFA 2022), Stuttgart, 2022.

Link to the article:

Migrating Legacy Ethernet-Based Traffic with Spatial Redundancy to TSN networks

‘- July 2022 – We have recently visited the Research Center in Robotics and Submarine Technologies, at the Universitat Jaume I de Castelló, where we have carried out experiments with our autonomous vehicles. We have also coordinated our research and development activity with the Robotics and Interactive Systems group, integrating systems from both groups and sharing knowledge. CIRTESU was inaugurated in 2020 and is currently one of the most advanced facilities in Europe for the study and development of underwater technologies. After the CIRTESU testing week, the UIB and UJI groups will join UTEK for sea trials near Cartagena, incorporating an autonomous vessel.

 

During the first week of February 2022, the UIB-Marine Robotics Team has been working in the Cabrera archipelago. The campaign was aimed at searching and mapping the invasive species Halimeda incrassata. This task is included in the INVHALIproject, funded by Fundación Biodiversidad. The Turbot AUV and the Xiroi ASV provided very good performance, autonomously recording large datasets of stereo images in shallow waters ranging from 1 to 20 meters depth. The data will be used for training automatic detection algorithms and also for building photomosaics to monitor the evolution of the threat.

On the 15th and 22nd of December we held an open day for students of Industrial Electronics and Automation Engineering at the Universitat de les Illes Balears, in which we gave technical-sales talks complemented with demonstrations of practical cases of integration of home automation systems with cutting-edge technology in our Control & Comfort offices at c/ Pedrera, 12 in Palma de Mallorca.

Click here to access the web page of Control & Confort company to see the complete product portfolio

Control & Comfort
B57749053
Pedrera, 12 bajos 2
07001 Palma de Mallorca
Tel: +34 971 495 582
controlandcomfort.com

Los días 15 y 22 de diciembre celebramos una jornada de puertas abiertas para estudiantes de Grado de Ingeniería Electrónica Industrial y Automática de la Universitat de les Illes Balears, en la que impartimos charlas de índole técnico-comercial complementadas con demostración de casos prácticos en integración de sistemas domóticos con tecnología de vanguardia presentes en nuestras oficinas de Control & Comfort sito c/ Pedrera, 12 de Palma de Mallorca.

Clicka aqui para acceder a la página Web de la empresa Control & Confort para ver el portafolio completo de productos

Author: Inés Álvarez Vadillo
Supervisors: Julián Proenza Arenas and Manuel Alejandro Barranco González
Date: 2nd December 2021

Inés Álvarez Vadillo successfully presented her Ph.D. thesis at the University of the Balearic Islands.

Abstract:

In the last years several novel industrial applications have emerged, such as Industry 4.0, autonomous driving or intelligent energy distribution. On top of the intrinsic characteristics of traditional industrial applications, i.e. real-time constraints and increased dependability; these novel applications exhibit new requirements such as adaptivity, high interconnectivity or integration of services with different operation requirements over a single infrastructure.

Because of their nature, the systems that support these applications are distributed, which means that the communication subsystem is fundamental to guarantee the proper operation of the overall system. Nonetheless, currently there is not a standardised communication protocol capable of fulfilling all the needs of the communication subsystems of these novel applications.

For this reason, the Time-Sensitive Networking (TSN) Task Group has been working on proposing a series of technical standards to provide Ethernet with hard real-time guarantees, on-line management of the traffic and fault tolerance mechanisms to increase reliability.

Specifically, TSN proposes using space redundancy to increase the reliability of Ethernet networks, which consists in transmitting each frame in parallel through several physical paths. Nevertheless, using space redundancy to tolerate temporary faults is not a cost-effective solution. For this reason, in this dissertation, we propose to use time redundancy to tolerate temporary faults in the links of TSN-based networks. Specifically, we present the Proactive Transmission of Replicated Frames mechanism, which consists in
transmitting several copies of each frame in a preventive manner to increase the chances of at least one copy reaching its intended destination even in the presence of temporary faults. This mechanism is specially tailored to the specific characteristics of TSN.

In this dissertation we show that we can increase the reliability of TSN-based networks using PTRF to provide proactive time redundancy to tolerate temporary faults in the links, while keeping the characteristics that make TSN an appealing technology for the communication subsystem of the aforementioned applications.