Publication type: Conferences

A Laser-based Structured Light System (LbSLS) has been designed to perform underwater close-range 3D reconstructions even with high turbidity conditions and outperform conventional systems. The system uses a camera and a 532 nm green laser projector. The optical technique used is based on the projection of a  pattern  obtained placing a Diffractive Optical Element (DOE) in front of the laser beam. In the experiments described in this manuscript, the DOE used diffracts the laser beam in 25 parallel lines providing enough information in a single camera frame to perform a 3D reconstruction.

The use of saliency mechanisms for defect detection is discussed in this work. We consider defects on regular surfaces as conspicuous areas that catch the attention of the surveyors. Following this approach, we propose the use of the Bayesian framework SUN, devised to provide saliency information based on natural statistics, to combine information about the visual appearance of the surface under inspection, to finally indicate where the defects (if any) are located. The visual information is suggested to be based on features commonly used to predict human eye fixations: contrast and symmetry. We demonstrate that these two features provide a description of the surface that can be used to indicate whether it is defective or not. Our approach is assessed using a publicly available image dataset containing a variety of surfaces with defective areas. The performance of the defect detector is evaluated through cross-validation and successful results are obtained.

Side Scan Sonars (SSS) are frequently used by Autonomous Underwater Vehicles (AUV) to observe the sea floor. However, automatically processing the information they provide to perform tasks such as Simultaneous Localization and Mapping (SLAM) or mosaicking is difficult and error prone. Not only the literature is scarce on this subject, but also the number of software tools to help researchers is very reduced. This paper presents SSS-SLAM, a new Object Oriented software framework based on Matlab aimed at facilitating research on underwater SLAM using SSS. The framework includes both libraries and support tools.

This paper presents a new method to remove the effects of uneven ensonification in Side-Scan Sonar (SSS) acoustic images, facilitating their use in underwater robotics. A Lambertian model is used for the sonar scattering process, relating the received echo with the ensonification intensity, the sea bed reflectivity and the incidence angle. As for the ensonification intensity, a sensitivity pattern model is adapted to SSS sensors. By means of these models, SSS imagery  is significantly improved.

While commercially available autonomous underwater vehicles (AUVs) are routinely used in survey missions, a new set of applications exist which clearly demand intervention capabilities: the maintenance of permanent underwater structures as well as the recovery of benthic stations or black-boxes are a few of them. These tasks are addressed nowadays using manned submersibles or work-class remotely operated vehicles (ROVs), equipped with teleoperated arms under human supervision. In the context of the TRITON Spanish funded project, a subsea panel docking and an intervention procedure are proposed. The light-weight intervention AUV (I-AUV) Girona 500 is used to autonomously dock into a subsea panel using a funnel-based docking method for passive accommodation. Once docked, an autonomous fixed-based manipulation system, which uses feedback from a digital camera, is used to turn a valve and plug/unplug a connector. The paper presents the techniques used for the autonomous docking and manipulation as well as how the adapted subsea panel has been designed to facilitate such operations.

The Flexible Time Triggered (FTT) paradigm provides online flexible scheduling for distributed embedded systems but it does not present adequate fault tolerance mechanisms so as to reach a very high reliability. Adding the adequate fault tolerance mechanisms to FTT-based architectures would open room for adaptive yet highly dependable systems. In this work we present a fault-tolerant system architecture for control applications that adds a node replication scheme with voting on top of an FTT-based system. Using a previously proposed network-centric approach we show how to coordinate the execution of the different phases for a typical control application in our system architecture, i.e. we show how to trigger the execution of tasks in node replicas and the transmission of messages in the communication channel, using the underlying FTT protocol. At the end, we demonstrate how to apply this idea of coordinated dispatching to one concrete control application, ball-on-plate.

In the context of critical applications there is an increasing interest in having Distributed Embedded Systems (DESs) that are able to operate in dynamic environments, while at the same time reaching a high reliability. The Flexible Time-Triggered communication paradigm (FTT) is designed to support the QoS and real-time requirements of the traffic of these systems. However, FTT does not provide fault tolerance. This paper explains our on-going work towards designing a consistent and highly-reliable voting protocol which supports node replication on DESs that use FTT switched Ethernet. In particular, we propose a protocol for the node replicas to vote consistently on messages exchanged through an FTT Ethernet network that uses time redundancy, while trying to maximize the reliability that can be achieved thanks to the redundancy of the nodes and the communication subsystem itself.