23 research outputs found
Acoustic positioning of multiple AUVs by an ASV: experimental validation and performance characterisation
No full textCooperation among heterogeneous marine vehicles can offer several benefits to underwater applications, such as supporting the navigation of submerged robots by leveraging the information available to a surface vehicle. Within such context, this paper describes an acoustic positioning and communication protocol developed for the localisation of multiple Autonomous Underwater Vehicles (AUVs) by means of an Autonomous Surface Vehicle (ASV) equipped with an Ultra-Short BaseLine (USBL) device. To avoid packet collisions, access to the acoustic channel is managed through a time slot division mechanism. Unlike a classical Time Division Multiple Access (TDMA) protocol, the approach considered in this work consists of a centralised scheduling handled by the ASV, in which AUVs are localised at regular time epoch but can transmit only when queried. This allows to prioritise the number of position measurements taken by the ASV and then relayed back to the submerged vehicles, with the final goal of improving the navigation accuracy of the latter. The solution is also designed to handle the latency of acoustic communication and to correctly associate each position measurement with the corresponding acquisition time. Furthermore, it does not require any a priori synchronisation between the clocks of the vehicles involved and ensures complete decoupling between their navigation algorithms. Experimental activities in very-shallow waters, involving an ASV and two target nodes, were carried out to validate the multi-AUVs positioning system and to provide a characterisation of its performance
ISME activity on the use of Autonomous Surface and Underwater Vehicles for acoustic surveys at sea
No full textThe paper presents an overview of the recent and ongoing research activities of the Italian Interuniversity Center on Integrated Systems for the Marine Environment (ISME) in the field of geotechnical seismic surveying. Such activities, performed in the framework of the H2020 European project WiMUST, include the development of technologies and algorithms for Autonomous Surface Crafts and Autonomous Underwater Vehicles to perform geotechnical seismic surveying by means of a team of robots towing streamers equipped with acoustic sensors
AUV navigation, guidance, and control for geoseismic data acquisition
No full textThis chapter describes the main challenges, and the corresponding solutions, encountered during the development of the guidance, navigation, and control (GNC) systems for the autonomous underwater vehicles (AUVs) and the autonomous surface vehicles (ASVs) involved in the Widely scalable Mobile Underwater Sonar Technology (WiMUST) projec
North seeking in Motion: multi IMUs fusion in INS
No full textDevelop of an INS simulator using Matlab Object Oriented Programming.
Comparative analysis of different Kalaman Filter models.
Develop of a new model and method to improve the estimation performance of the Filter, inspired by the Carouseling technique
UNDERWATER SURVEILLANCE WITH AUTONOMOUS ROBOTS: CONSENSUS FROM BEARING-ONLY MEASUREMENTS
No full textDistributed systems and cooperative control are the main themes of the thesis work. The objective was the development of an algorithm capable of detecting and locating in a distributed way an underwater target by a team of Autonomous Underwater Vehicles (AUV). We are talking about passive detection and localization, so the sensors with which the vehicles are equipped, detect the target only when it emits a noise above a certain threshold (detection threshold), and they do not emit any acoustic signal. In fact, an additional objective was to detect the target without it being aware of the presence of the team. The innovation lies precisely in the use of a team of cooperating vehicles, which tow linear arrays in order to inspect a portion of an underwater area. The first arrays were in fact connected by winches and towed by individual boats and were hundreds of meters long, also had diameters of several centimeters and therefore the current autonomous underwater vehicles could not be able to withstand this load. During these inspections the first hydrophones that made up the array were mounted several meters from the stern of the ship, in order to avoid that the noise emitted by the engine compromised the measurements in acquisition. The advance of technology led to the development of linear arrays, whose diameter was less than 30 millimetres, drastically reducing the weight. It was therefore possible to think of dividing the overall length of the array and using a fleet of vehicles instead of a single vessel, thus introducing the problem of controlling a distributed system. Moreover, since the old arrays were several hundred meters long, in order to equalize the inspections carried out by large vessels, it is necessary that the team of vehicles maintain a certain distance and therefore a certain formation, so as to cover a wide geographical area. A first step in this area was taken during the European WiMUST (Widely scalable Mobile Underwater Sonar Technology) project. This project involved the use of a small fleet of autonomous underwater vehicles to carry out geophysical inspections. The arrays consisted of hydrophones streamers of small aperture in order to collect environmental data. One of the advantages revealed in WiMUST is that by controlling the geometry of the team it was possible to change the shape of the arrays according to the required goal. The innovation of the WiMUST project is therefore represented by the use of a team of cooperating underwater vehicles working as a distributed system, acquiring data locally on board each vehicle. However, the amount of data to be exchanged, in the order of Gigabytes, was too high, which is why the data processing was carried out off-line. As in WiMUST, in my thesis work, the data acquisition takes place locally on board of the vehicles, while regarding the communication between the vehicles, this takes place in real time through an appropriate communication strategy, thanks to which the agents of the network can share the information acquired with the other members of the team
Signal Processing for an Autonomous Underwater Vehicle: an FPGA approach
No full textThe idea of this thesis comes out from the participation of the University of Central Florida to the Annual International Autonomous Underwater Vehicle Competition of 2007. The objective of this competition is to make the AUV to accomplish to a specific route. A part of this route expects the AUV to detect a ping and following it as a source. The objective of this thesis is to improve the performance of this trajectory tracking. A Field Programmable Logic Array will be used to perform an effective Digital Signal Processing
Autonomous marine vehicles: improving autonomy through cooperation
No full textOceans exploration and conservation are topics that have gained great attention in recent years, driven by the commercial interests of industries and the growing desire of nations and environmental agencies to protect the health of the seas. In this context, the scientific community has a responsibility to identify new sustainable solutions that can lead to economic growth in the sector while preserving the marine environment. Achievements over the past two decades in the fields of marine technologies and mobile robotics have enabled the development of Autonomous Marine Vehicles (AMVs) and their use in a variety of application fields. Both Autonomous Surface Vehicles (ASVs) and Autonomous Underwater Vehicles (AUVs) fall into this category, both of which are characterised by the key element of autonomy. Although the term traditionally denotes the ability to make decisions without being controlled by others, for a mobile robot it can best be described as a set of strategies and solutions it has to implement in order to operate without human intervention in an unknown and changing environment, optimising its behaviour to adapt to the situation while trying to achieve a predefined goal. These marine robots are in fact powerful tools through which data, essential for gaining a better understanding of oceanographic processes, can be collected efficiently and at a lower cost than traditionally employed methods such as large oceanographic ships or Remotely Operated Vehicles (ROVs). In addition, AMVs have the potential to be able to perform inspection, repair and maintenance (IRM) operations, on offshore plants and submerged structures, safely and without any intervention by a human operator.
An ambitious goal of researchers from academia and industry is to make these vehicles capable of persistent operation in the field, so that they can constantly monitor sea conditions or carry out routine IRM tasks. To this end, it is necessary for marine vehicles to achieve the so-called long-term autonomy, through progress in the respective areas of navigation, perception, planning, communication and endurance.
This thesis aims to lay the foundation for the realisation of a Marine System of Systems (MSoS) in which heterogeneous, low-cost AMVs cooperate to complete missions in an efficient and intelligent manner. In this regard, the thesis proposes algorithms and methods to increase the autonomy of marine robots by approaching the problem from two perspectives: (i) by improving the navigation, perception and planning capabilities of individual underwater vehicles; (ii) through a cooperative system designed to enable accurate and low-cost underwater navigation of a team of AUVs through the support of an ASV. These two approaches also define the structure of the elaborate, which is in fact divided into two parts.
Regarding the first approach, a low-cost and data-based procedure for identifying the parameters of a simplified dynamic model for an AUV is initially presented. The process was tested and validated in the field and allowed the definition of a state transition model to be used within a Kalman-based filter to improve the navigation performance of an AUV. Next, in order to increase the perception capabilities of an underwater robot, solutions are presented for the online and offline processing of data sensed by a Side-Scan Sonar (SSS). The first method described makes it possible to obtain an accurate acoustic map of the inspected area by means of vehicle motion compensation, georeferencing of the data and their mosaicing. The second proposed solution is an algorithm for Automatic Target Reconition (ATR) based on saliency filters applied to SSS images for the Mine CounterMeasures (MCM) scenario. Finally, the last contribution provided in this first part concerns the adaptive planning capabilities with which a marine vehicle must be endowed in order to carry out operations efficiently and adapt to the surrounding environment. In particular, a planning algorithm inspired by evolutionary processes is proposed and analysed, the aim of which is to generate an optimal route in terms of information gathered and coverage of the area to be inspected. The method exploits a priori knowledge of the environment and a Gaussian Process (GP) to model the spatial distribution of the variable of interest, which is then used by the method to plan the optimal trajectory.
Concerning the second approach, the cooperative system based on two heterogeneous vehicles, an ASV and an AUV, is initially presented in detail. The system is designed so that the surface vehicle can provide support to the AUV with regard to navigation and communication. In fact, the ASV is able to locate the AUV through the use of an Ultra-Short BaseLine (USBL) device and the definition of a communication and positioning protocol specifically designed for the scenario in question. The distinctive features of the developed protocol are that it prioritises the positioning frequency, does not require any a priori synchronisation between vehicles and allows the exchange of information between robots and a command and control station for monitoring purposes, without affecting the localisation procedure. The cooperative system was conceived as a low-cost alternative solution for accurate underwater navigation, and was therefore designed to take advantage of minimal and typically standard equipment for AMVs: GPS sensor, Attitude and Heading Reference System (AHRS) and acoustic modem. It should be noted that although the AUV installs a simple acoustic modem, the ASV is equipped with a USBL device to perform localisation, which also has communication capabilities.
Strategies were implemented to allow tracking and following of the AUV by the surface robot, so that it could move to keep the distance between vehicles limited. An excessive increase in distance is in fact problematic for the performance of the positioning system, due to the unreliability of acoustic communication and the inherent delay of transmissions through this channel. The cooperative platform was extensively tested during experiments at sea, which allowed a characterisation of its performance in terms of navigation accuracy, tracking and reliability of the positioning procedure. A particular testing campaign allowed the evaluation of its performance also for a scenario involving the presence of natural gas seeps. Indeed, it is believed that such gas leaks may have an impact on the readings of the acoustic sensors typically employed by an AUV to navigate. An experimental analysis of the impact of these gas seeps on DVL- and USBL-based navigation strategies was therefore carried out, and the main issues an AUV may encounter when performing operations in similar areas were identified.
A further simulative study was conducted to define a motion planning algorithm designed for the ASV. The purpose of the planner is to guide the surface vehicle to positions from which it can take measurements with the USBL sensor that are as informative as possible for the submerged vehicle. Therefore, a cost functional was defined, composed of the determinant of the estimated covariance matrix related to the AUV position error and some penalising factors regarding the minimum and maximum distance between vehicles. By minimising this functional, the method aims to reduce the uncertainty in the position estimate made by the AUV, thereby improving its navigation accuracy.
Finally, the communication and positioning protocol was extended to the multi-AUVs scenario, defining a protocol inspired by the Time Division Multiple Access (TDMA) method but retaining the distinctive features of the two-vehicle approach: priority to positioning and no synchronisation between vehicles. The strategy was then tested and validated in the field through sea trials with one surface vehicle and two nodes to be located.
All the solutions proposed in this thesis represent an important first step towards the realisation of the MSoS, and progress towards the achievement of the long-term autonomy. The results also confirm how cooperation can be a game changer in defining low-cost strategies that pave the way for the use of AMVs in various application fields
Deformable-Medium Affordances for Interacting with Multi-Robot Systems
No full textIn this thesis, I address the issue of human-swarm interactions by proposing a new set of
affrodances that make a multi-robot system amenable to human control. An affordance,
as defined by Gibson, is a relation between an object and a user, where the object
explicitly allows the user to perform a particular action. The affordances we identify when
controlling a swarm, include stretching the swarm, molding it into a particular shape, splitting
and merging sub-swarms, and mixing of different swarms. The contribution beyond
the formulation of these affordances is the coupling of an image recognition framework
identified by an effective deformable-medium control interface, and the accompanying algorithms
needed to identify the appropriate inputs, and then turn those into decentralized
control laws for the individual robots. As result, the developed human-swarm interaction
methodology is applied to a team of mobile robots
RAMI competition: detection and classification of man-made objects for the autonomy of underwater robots
No full textThe Metrological Evaluation and Testing of Robots in International CompetitionS (METRICS) EU project, led by the National Metrology and Testing Laboratory (LNE), organizes robotics competitions in four priority areas identified by the European Commission: health, agri-food, inspection and maintenance of infrastructure and agile production. In particular, the first edition of the Robotics for Asset Maintenance and Inspection (RAMI) competition will take place and will be led by NATO Science and Technology Organization - Centre for Maritime Research and Experimentation (STO CMRE), one of the reference research centres in Europe devoted to marine robotics technologies. RAMI Cascade competition will focus on the classification, identification and localization of Object of Potential Interest (OPI) in underwater images.
In this thesis work an object detection and classification algorithm able, to deal with underwater imaging and to detect and classify different OPIs, is developed. The training dataset containing OPI images employed during the preparation of the algorithm is the one proposed by NATO STO CMRE to the participating teams as support to the development of software. The provided images have been collected in underwater environment and divided into five classes: colored buoys, black numbers on yellow pipes, black numbers over red background, red markers on yellow pipes and images without OPIs.
The algorithm has been developed according to the following main steps. Firstly, an effective color enhancement and color restoration procedure has been defined and implemented on the basis of the Jaffe-McGlamery computer image degradation model and by employing pixel intensity redistribution techniques. Red and Blue channels have been pre-compensated to balance the high attenuation suffered in underwater images; then Gray World Assumption (GWA), Dark Channel Prior (DCP) and Contrast Limited Adaptive Histogram Equalization (CLAHE) have been implemented and applied to the RGB images provided in the dataset to compensate the color cast and to increase the overall image contrast. After an initial phase of this work in which a color and shape based detection and classification algorithm has been developed and tested, a novel approach based on deep learning is proposed. A state-of-the-art Faster R-CNN network has been fine-tuned and tested on detection and classification of three OPIs classes: buoys, black digits (1-6) and red markers. Here, data augmentation techniques and Early Stopping have been employed to increase the performances of the network and to prevent over-fitting. For the detection of the black digits (1-6), the patch provided as output by the Faster R-CNN network has been processed employing a binarization based on the Otsu's adaptive thresholding method. Then, a Convolutional Neural Network (CNN), trained on the modified National Institute of Standards and Technology (MNIST) dataset and fine-tuned on a dataset created exploiting all the ground-truth boxes containing black digits, has been used to recognize and classify each digit. At last, the performance of the deep learning based novel approach has been evaluated through mean Average Precision (mAP), class Average Precision (AP) and accuracy metrics. The results reported in the last part of this work, show that the first stage of the algorithm, based on the Faster R-CNN network, achieved a mean Average Precision (mAP) of about 0.94, while the Digit Recognition CNN reached an accuracy of about 77%
Underwater acoustic source localization using a multi-robot system: the DAMPS project
No full textThis paper presents the funded project DAMPS, that addresses the development of a multi-robot system for the position tracking of noise sources in underwater environment. The robots composing the team collaborate in order to fuse the local information coming from the analysis of a Passive Acoustic Monitoring sensor, with the final aim of estimating the position of a the acoustic source. The team is thus a distributed, modular and reconfigurable sonar system that can be controlled remotely from a Control & Command center. In this paper we highlight the functional architecture that will be implemented within the project, the possible algorithmic approaches to be employed within the submodules and possible application scenarios
