Abstract
This paper propose a real-time vision framework for mobile robotics and describes the current implementation. The pipeline structure further reduces latency and allows a paralleled hardware implementation. A dedicated hardware vision sensor was developed in order to take advantage of the proposed architecture. The real-time characteristics and hardware partial implementation, coupled with low energy consumption address typical autonomous systems applications. A characterization of the implemented system in the Robocup scenario, during competition matches, is presented.

Abstract
The coordinated use of multiple Autonomous Underwater Vehicles can provide important advantages for oceanographic missions. One important mission application scenario can be the search of underwater plumes such as sources of freshwater of hydrotermal vents. These plumes characterize the environment by creating a gradient field of some measurable physical quantity. An innovative integrated acoustic navigation system and coordination control maneuver for a formation of 3 AUVs and I surface craft to gradient search and following missions is proposed. The specific formation geometry and topology takes in account the navigation and coordination requirements. It was designed to achieve an efficient, low cost and technically feasible solution. The system can operate in 3 modes depending on formation distances. Varying pinging rates and offsets are used to communicate parameters and mode changing. No additional underwater communication systems neither acoustic transponder deployment are needed for the vehicle coordination. This way a high degree of energy efficiency and overall mission low cost and simpler logistics is achieved. The hybrid nature of the coordinating maneuver allows the formation gradient survey and following with the efficient exploitation of the environment structuring by the phenomena to be studied. The individual control laws were designed in order to minimize the inter-vehicle communication. The coordination factors are the knowledge by the vehicles of each other behavior (since all vehicles execute the same control laws) and the detection of formation distortions. These distortions are detected by the relative navigation system. The proposed approach allows the low cost implementation of a multiple AUV coordinating control for a large range of oceanographic missions.

Abstract
In this article, we describe the analysis, design and implementation of a control architecture for a mobile platform to autonomously carry out transportation, surveillance and inspection tasks in semi-structured industrial environments. Based on a hierarchical structure composed by the Organization, Coordination and Functional levels organized linguistically and structured according to the Principle of Increasing Precision with Decreasing Intelligence, this control architecture is permits the real-time parallel execution of tasks.

Abstract
In this article, we describe the effort being carried out in the analysis, design and implementation of the control architecture for a mobile platform for autonomous transportation, surveillance and inspection in structured and semi-structured industrial environments. The control architecture is based in a hierarchical structure organized linguistically permitting the real-time parallel execution of tasks. This, architecture is composed of three levels, Organization, Coordination and Functional, structured according to the Increasing Precision with Decreasing Intelligence Principle.

Abstract
This work addresses reconstructing an ancient mining site in three-dimensional (3D) modelling with robotic systems, processing the information from two visible spectrum cameras. The developed solution, GeoTec System, was validated in an underground environment in the Monastery of Tibães (Braga, NW Portugal). This study was developed under the MineHeritage project's scope, aiming to attain society on the importance of raw materials across a historical approach. The outputs acquired from the datasets developed in a successful 3D reconstruction of the main gallery and secondary tunnels of the Aveleiras mine in Tibães. However, the investigation is still ongoing to contribute to applying 3D reconstruction technologies, GIS-based mapping and geovisualization techniques in the underground heritage environment.

Abstract
<p>The UNEXUP project, funded under EIT Raw Materials, is a direct continuation of the Horizon 2020 UNEXMIN project. While in UNEXMIN efforts were made towards the design, development and testing of an innovative exploration technology for underground flooded mines, in UNEXUP the main goal is to push the UNEXMIN technology into the market, while further improving the system’s hardware, software and capabilities. In parallel, the aim is to make a strong business case for the improved UNEXUP technology, as a result of tests and data collection from previous testing. Improvements to the UX-1 research prototypes will raise technology readiness levels from TRL 6, as verified at the end of the UNEXMIN project, to TRL 7/8 by 2022. A "real service-to-real client" approach will be demonstrated, supporting mineral exploration and mine surveying efforts in Europe with unique data from flooded environments that cannot be obtained without high costs, or risks to human lives, in any other ways.</p><p>The specific purpose of UNEXUP is to commercially deploy a new raw materials exploration / mine mapping service based on a new class of mine explorer robots, for non-invasive resurveying of flooded mines. The inaccessibility of the environment makes autonomy a critical and primary objective of the project, which will present a substantial effort in resurveying mineral deposits in Europe where the major challenges are the geological uncertainty, and technological / economic feasibility of mine development. The robot’s ability to gather high-quality and high-resolution information from currently inaccessible mine sites will increase the knowledge of mineral deposits in Europe, whilst decreasing exploration costs – such as the number of deep exploration drillholes needed. This can potentially become a game changing technology in the mining panorama, where the struggle for resources is ever increasing.</p><p>On the technical side, a fourth robot, modular in nature, will be added to the current multi-robot platform, providing additional functionalities to the exploration system, including better range and depth performance. Hardware and software upgrades, as well as new capabilities delivered by the platform will greatly extend the usefulness of the platform in different environments and applications. Among these: rock sampling, better data acquisition and management, further downsizing, extended range, improved self-awareness and decision making, mature post-processing (such as the deployment of 3D virtual reality models), ability to rescue other robots, and interaction with the data will be targeted during the next years. Upgrading the overall technology with these tools, and possibly additional ones, will allow the system to operate with more reliability and security, with reduced costs.</p><p>These added functions arise from different stakeholders’ feedbacks from the UNEXMIN project. UNEXUP targets parties from the mining, robotics and mineral exploration sectors, as well as all other sectors that have any kind of underwater structure that needs to be surveyed – caves, underground reservoirs, water pipelines and fisheries are among them. For the purpose of exploitation of the technology, a joint company was founded, “UNEXMIN GeoRobotics Ltd”, which is part of the UNEXUP consortium, and is responsible for selling the service to the market.</p>