An intrinsic genetic program for autonomous differentiation of muscle cells in the ascidian embryo.

The B-line presumptive muscle cells of ascidian embryos have extensive potential for self-differentiation dependent on determinants prelocalized in the myoplasm of fertilized eggs. Ascidian larval muscle cells therefore provide an experimental system with which to explore an intrinsic genetic program for autonomous specification of embryonic cells. Experiments with egg fragments suggested that maternal mRNAs are one of the components of muscle determinants. Expression of larval muscle actin genes begins as early as the 32-cell stage, prior to the developmental fate restriction of the cells. The timing of initiation of the actin gene expression proceeds the expression of an ascidian homologue of vertebrate MyoD by a few hours. Mutations in the proximal E-box of the 5' flanking region of the actin genes did not alter the promoter activity for muscle-specific expression of reporter gene. These results, together with results of deletion constructs of fusion genes, suggest that muscle determinants regulate directly, or indirectly via regulatory factors other than MyoD, the transcription of muscle-specific structural genes leading to the terminal differentiation.

Lens complementation system for the genetic analysis of growth, differentiation, and apoptosis in vivo.

A genetic approach has been established that combines the advantages of blastocyst complementation with the experimental attributes of the developing lens for the functional analysis of genes governing cellular proliferation, terminal differentiation, and apoptosis. This lens complementation system (LCS) makes use of a mutant mouse strain, aphakia (ak), homozygotes of which fail to develop an ocular lens. We demonstrate that microinjection of wild-type embryonic stem (ES) cells into ak/ak blastocysts produces chimeras with normal ES-cell-derived lenses and that microinjection of Rb-/- ES cells generates an aberrant lens phenotype identical to that obtained through conventional gene targeting methodology. Our determination that a cell autonomous defect underlies the aphakia condition assures that lenses generated through LCS are necessarily ES-cell-derived. LCS provides for the rapid phenotypic analysis of loss-of-function mutations, circumvents the need for germ-line transmission of null alleles, and, most significantly, facilitates the study of essential genes whose inactivation is associated with early lethal phenotypes.

Parvovirus B19 promoter at map unit 6 confers autonomous replication competence and erythroid specificity to adeno-associated virus 2 in primary human hematopoietic progenitor cells.

The pathogenic human parvovirus B19 is an autonomously replicating virus with a remarkable tropism for human erythroid progenitor cells. Although the target cell specificity for B19 infection has been suggested to be mediated by the erythrocyte P-antigen receptor (globoside), a number of nonerythroid cells that express this receptor are nonpermissive for B19 replication. To directly test the role of expression from the B19 promoter at map unit 6 (B19p6) in the erythroid cell specificity of B19, we constructed a recombinant adeno-associated virus 2 (AAV), in which the authentic AAV promoter at map unit 5 (AAVp5) was replaced by the B19p6 promoter. Although the wild-type (wt) AAV requires a helper virus for its optimal replication, we hypothesized that inserting the B19p6 promoter in a recombinant AAV would permit autonomous viral replication, but only in erythroid progenitor cells. In this report, we provide evidence that the B19p6 promoter is necessary and sufficient to impart autonomous replication competence and erythroid specificity to AAV in primary human hematopoietic progenitor cells. Thus, expression from the B19p6 promoter plays an important role in post-P-antigen receptor erythroid-cell specificity of parvovirus B19. The AAV-B19 hybrid vector system may also prove to be useful in potential gene therapy of human hemoglobinopathies.

Pleasant sweeping grooming touch in rhesus monkey: effects on autonomic nervous system and its codification in posterior insular and secondary somatosensory cortex

Il tatto assume un'importanza fondamentale nella vita quotidiana, in quanto ci permette di discriminare le caratteristiche fisiche di un oggetto specifico, di identificarlo e di eventualmente integrare le suddette informazioni tattili con informazioni provenienti da altri canali sensoriali. Questa è la componente sensoriale-discriminativa del tatto. Tuttavia quotidianamente il tatto assume un ruolo fondamentale durante le diverse interazioni sociali, positive, come quando abbracciamo o accarezziamo una persona con cui abbiamo un rapporto affettivo e negative, per esempio quando allontaniamo una persona estranea dal nostro spazio peri-personale. Questa componente è la cosiddetta dimensione affettiva-motivazionale, la quale determina la codifica della valenza emotiva che l'interazione assume. Questa componente ci permette di creare, mantenere o distruggere i legami sociali in relazione al significato che il tocco assume durante l'interazione. Se per esempio riceviamo una carezza da un familiare, questa verrà percepita come piacevole e assumerà un significato affiliativo. Questo tipo di tocco è comunente definito come Tocco Sociale (Social Touch). Gli aspetti discriminativi del tatto sono stati ben caratterizzati, in quanto storicamente, il ruolo del tatto è stato considerato quello di discriminare le caratteristiche di ciò che viene toccato, mentre gli aspetti affettivi sono stati solo recentemente indagati considerando la loro importanza nelle interazioni sociali. Il tocco statico responsabile dell'aspetto discriminante attiva a livello della pelle le grandi fibre mieliniche (Aβ), modulando a livello del sistema nervoso centrale le cortecce sensoriali, sia primarie che secondarie. Questo permette la codifica a livello del sistema nervoso centrale delle caratteristiche fisiche oggettive degli oggetti toccati. Studi riguardanti le caratteristiche del tocco affiliativo sociale hanno messo in evidenza che suddetta stimolazione tattile 1) è un particolare tocco dinamico che avviene sul lato peloso delle pelle con una velocità di 1-10 cm/sec; 2) attiva le fibre amieliniche (fibre CT o C-LTMRs); 3) induce positivi effetti autonomici, ad esempio la diminuzione della frequenza cardiaca e l'aumento della variabilità della frequenza cardiaca; e 4) determina la modulazione di regioni cerebrali coinvolte nella codifica del significato affiliativo dello stimolo sensoriale periferico, in particolare la corteccia insulare. Il senso del tatto, con le sue due dimensioni discriminativa e affiliativa, è quotidianamente usato non solo negli esseri umani, ma anche tra i primati non umani. Infatti, tutti i primati non umani utilizzano la componente discriminativa del tatto per identificare gli oggetti e il cibo e l'aspetto emotivo durante le interazioni sociali, sia negative come durante un combattimento, che positive, come durante i comportamenti affiliativi tra cui il grooming. I meccanismi di codifica della componente discriminativa dei primati non umani sono simili a quelli umani. Tuttavia, si conosce ben poco dei meccanismi alla base della codifica del tocco piacevole affiliativo. Pur essendo ben noto che i meccanorecettori amilienici C-LTMRs sono presenti anche sul lato peloso della pelle dei primati non umani, attualmente non ci sono studi riguardanti la correlazione tra il tocco piacevole e la loro modulazione, come invece è stato ampiamente dimostrato nell'uomo. Recentemente è stato ipotizzato (Dunbar, 2010) il ruolo delle fibre C-LTMRs durante il grooming, in particolare durante il cosiddetto swepping. Il grooming è costituito da due azioni motorie, lo sweeping e il picking che vengono eseguite in modo ritmico. Durante lo sweeping la scimmia agente muove il pelo della scimmia ricevente con un movimento a mano aperta, per poter vedere il preciso punto della pelle dove eseguire il picking, ovvero dove prendere la pelle a livello della radice del pelo con le unghie dell'indice e del pollice e tirare per rimuovere parassiti o uova di parassiti e ciò che è rimasto incastrato nel pelo. Oltre il noto ruolo igenico, il grooming sembra avere anche una importante funzione sociale affiliativa. Come la carezza nella società umana, cosi il grooming tra i primati non umani è considerato un comportamento. Secondo l'ipotesi di Dunbar l'attivazione delle C-LTMRs avverrebbe durante lo sweeping e questo porta a supporre che lo sweeping, come la carezza umana, costituisca una componente affiliativa del grooming, determinando quindi a contribuire alla sua codifica come comportamento sociale. Fino ad ora non vi è però alcuna prova diretta a sostegno di questa ipotesi. In particolare, 1) la velocità cui viene eseguito lo sweeping è compatibile con la velocità di attivazione delle fibre CT nell'uomo e quindi con la velocità tipica della carezza piacevole di carattere sociale affiliativo (1-10 cm/sec)?; 2) lo sweeping induce la stessa modulazione del sistema nervoso autonomo in direzione della modulazione del sistema vagale, come il tocco piacevole nell'uomo, attraverso l'attivazione delle fibre CT?; 3) lo sweeping modula la corteccia insulare, cosi come il tocco piacevole viene codificato come affiliativo nell'uomo mediante le proiezioni delle fibre CT a livello dell'insula posteriore? Lo scopo del presente lavoro è quella di testare l'ipotesi di Dunbar sopra citata, cercando quindi di rispondere alle suddette domande. Le risposte potrebbero consentire di ipotizzare la somiglianza tra lo sweeping, caratteristico del comportamento affiliativo di grooming tra i primati non umani e la carezza. In particolare, abbiamo eseguito 4 studi pilota. Nello Studio 1 abbiamo valutato la velocità con cui viene eseguito lo sweeping tra scimmie Rhesus, mediante una analisi cinematica di video registrati tra un gruppo di scimmie Rhesus. Negli Studi 2 e 3 abbiamo valutato gli effetti sul sistema nervoso autonomo dello sweeping eseguito dallo sperimentatore su una scimmia Rhesus di sesso maschile in una tipica situazione sperimentale. La stimolazione tattile è stata eseguita a diverse velocità, in accordo con i risultati dello Studio 1 e degli studi umani che hanno dimostrato la velocità ottimale e non ottimale per l'attivazione delle C-LTMRs. In particolare, nello Studio 2 abbiamo misurato la frequenza cardiaca e la variabilità di questa, come indice della modulatione vagale, mentre nello Studio 3 abbiamo valutato gli effetti dello sweeping sul sistema nervoso autonomo in termini di variazioni di temperatura del corpo, nello specifico a livello del muso della scimmia. Infine, nello Studio 4 abbiamo studiato il ruolo della corteccia somatosensoriale secondaria e insulare nella codifica dello sweeping. A questo scopo abbiamo eseguito registrazioni di singoli neuroni mentre la medesima scimmia soggetto sperimentale dello Studio 2 e 3, riceveva lo sweeping a due velocità, una ottimale per l'attivazione delle C-LTMRs secondo gli studi umani e i risultati dei tre studi sopra citati, ed una non ottimale. I dati preliminari ottenuti, dimostrano che 1) (Studio 1) lo sweeping tra scimmie Rhesus viene eseguito con una velocità media di 9.31 cm/sec, all'interno dell'intervallo di attivazione delle fibre CT nell'uomo; 2) (Studio 2) lo sweeping eseguito dallo sperimentatore sulla schiena di una scimmia Rhesus di sesso maschile in una situazione sperimentale determina una diminuzione della frequenza cardiaca e l'aumento della variabilità della frequenza cardiaca se eseguito alla velocità di 5 e 10 cm/sec. Al contrario, lo sweeping eseguito ad una velocità minore di 1 cm/sec o maggiore di 10 cm/sec, determina l'aumento della frequenza cardiaca e la diminuzione della variabilità di questa, quindi il decremento dell'attivazione del sistema nervoso parasimpatico; 3) (Studio 3) lo sweeping eseguito dallo sperimentatore sulla schiena di una scimmia Rhesus di sesso maschile in una situazione sperimentale determina l'aumento della temperatura corporea a livello del muso della scimmia se eseguito alla velocità di 5-10 cm/sec. Al contrario, lo sweeping eseguito ad una velocità minore di 5 cm/sec o maggiore di 10 cm/sec, determina la diminuzione della temperatura del muso; 4) (Studio 4) la corteccia somatosensoriale secondaria e la corteccia insulare posteriore presentano neuroni selettivamente modulati durante lo sweeping eseguito ad una velocità di 5-13 cm/sec ma non neuroni selettivi per la codifica della velocità dello sweeping minore di 5 cm/sec. Questi risultati supportano l'ipotesi di Dunbar relativa al coinvolgimento delle fibre CT durante lo sweeping. Infatti i dati mettono in luce che lo sweeping viene eseguito con una velocità (9.31 cm/sec), simile a quella di attivazione delle fibre CT nell'uomo (1-10 cm/sec), determina gli stessi effetti fisiologici positivi in termini di frequenza cardiaca (diminuzione) e variabilità della frequenza cardiaca (incremento) e la modulazione delle medesime aree a livello del sistema nervoso centrale (in particolare la corteccia insulare). Inoltre, abbiamo dimostrato per la prima volta che suddetta stimolazione tattile determina l'aumento della temperatura del muso della scimmia. Il presente studio rappresenta la prima prova indiretta dell'ipotesi relativa alla modulazione del sistema delle fibre C-LTMRs durante lo sweeping e quindi della codifica della stimolazione tattile piacevole affiliativa a livello del sistema nervoso centrale ed autonomo, nei primati non umani. I dati preliminari qui presentati evidenziano la somiglianza tra il sistema delle fibre CT dell'uomo e del sistema C-LTMRs nei primati non umano, riguardanti il Social Touch. Nonostante ciò abbiamo riscontrato alcune discrepanze tra i risultati da noi ottenuti e quelli invece ottenuti dagli studi umani. La velocità media dello sweeping è di 9.31 cm / sec, rasente il limite superiore dell’intervallo di velocità che attiva le fibre CT nell'uomo. Inoltre, gli effetti autonomici positivi, in termini di battito cardiaco, variabilità della frequenza cardiaca e temperatura a livello del muso, sono stati evidenziati durante lo sweeping eseguito con una velocità di 5 e 10 cm/sec, quindi al limite superiore dell’intervallo ottimale che attiva le fibre CT nell’uomo. Al contrario, lo sweeping eseguito con una velocità inferiore a 5 cm/sec e superiore a 10 cm/sec determina effetti fisiologici negativo. Infine, la corteccia insula sembra essere selettivamente modulata dallo stimolazione eseguita alla velocità di 5-13 cm/sec, ma non 1-5 cm/sec. Quindi, gli studi sul sistema delle fibre CT nell’uomo hanno dimostrato che la velocità ottimale è 1-10 cm/sec, mentre dai nostri risultati la velocità ottimale sembra essere 5-13 cm / sec. Quindi, nonostante l'omologia tra il sistema delle fibre CT nell'umano deputato alla codifica del tocco piacevole affiliativo ed il sistema delle fibre C-LTMRs nei primati non umani, ulteriori studi saranno necessari per definire con maggiore precisione la velocità ottimale di attivazione delle fibre C-LTMR e per dimostrare direttamente la loro attivazione durante lo sweeping, mediante la misurazione diretta della loro modulazione. Studi in questa direzione potranno confermare l'omologia tra lo sweeping in qualità di tocco affiliativo piacevole tra i primati non umani e la carezza tra gli uomini. Infine, il presente studio potrebbe essere un importante punto di partenza per esplorare il meccanismo evolutivo dietro la trasformazione dello sweeping tra primati non umani, azione utilitaria eseguita durante il grooming, a carezza, gesto puramente affiliativo tra gli uomini.

Developing Autonomous Microgrids: A Theoretical Case Study

Microgrids are autonomously operated, geographically clustered electricity generation and distribution systems that supply power in closed system settings; they are highly compatible with renewable energy sources and distributed generation technologies. Mocrogrids are currently a serially underutilized and underappreciated commodity in the energy infrastructure portfolio worldwide. To demonstrate feasibility under poor conditions (little renewable energy potential and high demand) this capstone project develops a theoretical case study in which a renewable microgrid is employed to power rural communities of southern Montgomery County, Arkansas. Utilizing commercially manufactured 1.5-megawatt wind turbines and a 1-megawatt solar panel generation system, 4-megawatts of lithium ion battery storage, and demand response technology, a microgrid is designed that supplies 235 households with reliable electricity supply.

Decomposition of human behavior using model of competencies: application to robotic control system

For many years, humans and machines have shared the same physical space. To facilitate their interaction with humans, their social integration and for more rational behavior has been sought that the robots demonstrate human-like behavior. For this it is necessary to understand how human behavior is generated, discuss what tasks are performed and how relate to themselves, for subsequent implementation in robots. In this paper, we propose a model of competencies based on human neuroregulator system for analysis and decomposition of behavior into functional modules. Using this model allow separate and locate the tasks to be implemented in a robot that displays human-like behavior. As an example, we show the application of model to the autonomous movement behavior on unfamiliar environments and its implementation in various simulated and real robots with different physical configurations and physical devices of different nature. The main result of this work has been to build a model of competencies that is being used to build robotic systems capable of displaying behaviors similar to humans and consider the specific characteristics of robots.

A grammatical approach to the modeling of an autonomous robot

Virtual Worlds Generator is a grammatical model that is proposed to define virtual worlds. It integrates the diversity of sensors and interaction devices, multimodality and a virtual simulation system. Its grammar allows the definition and abstraction in symbols strings of the scenes of the virtual world, independently of the hardware that is used to represent the world or to interact with it. A case study is presented to explain how to use the proposed model to formalize a robot navigation system with multimodal perception and a hybrid control scheme of the robot. The result is an instance of the model grammar that implements the robotic system and is independent of the sensing devices used for perception and interaction. As a conclusion the Virtual Worlds Generator adds value in the simulation of virtual worlds since the definition can be done formally and independently of the peculiarities of the supporting devices.

Model of Competencies for Decomposition of Human Behavior: Application to Control System of Robots

Humans and machines have shared the same physical space for many years. To share the same space, we want the robots to behave like human beings. This will facilitate their social integration, their interaction with humans and create an intelligent behavior. To achieve this goal, we need to understand how human behavior is generated, analyze tasks running our nerves and how they relate to them. Then and only then can we implement these mechanisms in robotic beings. In this study, we propose a model of competencies based on human neuroregulator system for analysis and decomposition of behavior into functional modules. Using this model allow separate and locate the tasks to be implemented in a robot that displays human-like behavior. As an example, we show the application of model to the autonomous movement behavior on unfamiliar environments and its implementation in various simulated and real robots with different physical configurations and physical devices of different nature. The main result of this study has been to build a model of competencies that is being used to build robotic systems capable of displaying behaviors similar to humans and consider the specific characteristics of robots.

Carbonate system during the May 2013 MedSeA cruise

Most global ocean models are based on the assumption of a "steady state" ocean. Here, we investigate the validation of this hypothesis for the anthropized Mediterranean Sea. In order to do so, we calculated the mixing coefficients of the water masses detected in this sea via an optimum multiparameter analysis referred to as the MIX approach, using data from the BOUM (2008) and MedSeA (2013) cruises. The comparison of the mixing coefficients of each water mass, between 2008 and 2013, indicates that some of their proportions have significantly changed. Surface water mass proportions did not change significantly (Delta0.05-0.1), while intermediate and deep water mass mixing coefficients of both Eastern and Western basins were significantly modified (~Delata 0.35). This study clearly shows that the Mediterranean seawater is not in a "steady state".

Towards rope shovel automation operation information system

The road to electric rope shovel automation is marked with technological innovations that include an increase in operational information available to mining operations. The CRCMining Shovel Operator Information System not only collects machine operational data but also provides the operator with knowledge-of-performance and influences his/her performance to achieve higher productivity with reduced machine duty. The operator’s behaviour is one of the most important aspects of the man-machine interaction to be considered before semi- or fully-automated shovel systems can be realised. This paper presents the results of the rope shovel studies conducted by CRCMining between 2002 and 2004, provides information on current research to improve shovel performance and briefly discusses the implications of human-system interactions on future designs of autonomous machines.

Towards a standards-based autonomic context management system

Pervasive computing applications must be sufficiently autonomous to adapt their behaviour to changes in computing resources and user requirements. This capability is known as context-awareness. In some cases, context-aware applications must be implemented as autonomic systems which are capable of dynamically discovering and replacing context sources (sensors) at run-time. Unlike other types of application autonomy, this kind of dynamic reconfiguration has not been sufficiently investigated yet by the research community. However, application-level context models are becoming common, in order to ease programming of context-aware applications and support evolution by decoupling applications from context sources. We can leverage these context models to develop general (i.e., application-independent) solutions for dynamic, run-time discovery of context sources (i.e., context management). This paper presents a model and architecture for a reconfigurable context management system that supports interoperability by building on emerging standards for sensor description and classification.

Hippocampal models for simultaneous localisation and mapping on an autonomous robot

To navigate successfully in a novel environment a robot needs to be able to Simultaneously Localize And Map (SLAM) its surroundings. The most successful solutions to this problem so far have involved probabilistic algorithms, but there has been much promising work involving systems based on the workings of part of the rodent brain known as the hippocampus. In this paper we present a biologically plausible system called RatSLAM that uses competitive attractor networks to carry out SLAM in a probabilistic manner. The system can effectively perform parameter self-calibration and SLAM in onedimension. Tests in two dimensional environments revealed the inability of the RatSLAM system to maintain multiple pose hypotheses in the face of ambiguous visual input. These results support recent rat experimentation that suggest current competitive attractor models are not a complete solution to the hippocampal modelling problem.

Helicopter automation using a low-cost sensing system

This paper details the design of an autonomous helicopter control system using a low cost sensor suite. Control is maintained using simple nested PID loops. Aircraft attitude, velocity, and height is estimated using an in-house designed IMU and vision system. Information is combined using complimentary filtering. The aircraft is shown to be stabilised and responding to high level demands on all axes, including heading, height, lateral velocity and longitudinal velocity.

Multisensorial obstacles detection and classification for ADAS and autonomous driving

A reliable perception of the real world is a key-feature for an autonomous vehicle and the Advanced Driver Assistance Systems (ADAS). Obstacles detection (OD) is one of the main components for the correct reconstruction of the dynamic world. Historical approaches based on stereo vision and other 3D perception technologies (e.g. LIDAR) have been adapted to the ADAS first and autonomous ground vehicles, after, providing excellent results. The obstacles detection is a very broad field and this domain counts a lot of works in the last years. In academic research, it has been clearly established the essential role of these systems to realize active safety systems for accident prevention, reflecting also the innovative systems introduced by industry. These systems need to accurately assess situational criticalities and simultaneously assess awareness of these criticalities by the driver; it requires that the obstacles detection algorithms must be reliable and accurate, providing: a real-time output, a stable and robust representation of the environment and an estimation independent from lighting and weather conditions. Initial systems relied on only one exteroceptive sensor (e.g. radar or laser for ACC and camera for LDW) in addition to proprioceptive sensors such as wheel speed and yaw rate sensors. But, current systems, such as ACC operating at the entire speed range or autonomous braking for collision avoidance, require the use of multiple sensors since individually they can not meet these requirements. It has led the community to move towards the use of a combination of them in order to exploit the benefits of each one. Pedestrians and vehicles detection are ones of the major thrusts in situational criticalities assessment, still remaining an active area of research. ADASs are the most prominent use case of pedestrians and vehicles detection. Vehicles should be equipped with sensing capabilities able to detect and act on objects in dangerous situations, where the driver would not be able to avoid a collision. A full ADAS or autonomous vehicle, with regard to pedestrians and vehicles, would not only include detection but also tracking, orientation, intent analysis, and collision prediction. The system detects obstacles using a probabilistic occupancy grid built from a multi-resolution disparity map. Obstacles classification is based on an AdaBoost SoftCascade trained on Aggregate Channel Features. A final stage of tracking and fusion guarantees stability and robustness to the result.

Designing a Computer Vision System for Underwater Robotic Interventions

This thesis deals with the challenging problem of designing systems able to perceive objects in underwater environments. In the last few decades research activities in robotics have advanced the state of art regarding intervention capabilities of autonomous systems. State of art in fields such as localization and navigation, real time perception and cognition, safe action and manipulation capabilities, applied to ground environments (both indoor and outdoor) has now reached such a readiness level that it allows high level autonomous operations. On the opposite side, the underwater environment remains a very difficult one for autonomous robots. Water influences the mechanical and electrical design of systems, interferes with sensors by limiting their capabilities, heavily impacts on data transmissions, and generally requires systems with low power consumption in order to enable reasonable mission duration. Interest in underwater applications is driven by needs of exploring and intervening in environments in which human capabilities are very limited. Nowadays, most underwater field operations are carried out by manned or remotely operated vehicles, deployed for explorations and limited intervention missions. Manned vehicles, directly on-board controlled, expose human operators to risks related to the stay in field of the mission, within a hostile environment. Remotely Operated Vehicles (ROV) currently represent the most advanced technology for underwater intervention services available on the market. These vehicles can be remotely operated for long time but they need support from an oceanographic vessel with multiple teams of highly specialized pilots. Vehicles equipped with multiple state-of-art sensors and capable to autonomously plan missions have been deployed in the last ten years and exploited as observers for underwater fauna, seabed, ship wrecks, and so on. On the other hand, underwater operations like object recovery and equipment maintenance are still challenging tasks to be conducted without human supervision since they require object perception and localization with much higher accuracy and robustness, to a degree seldom available in Autonomous Underwater Vehicles (AUV). This thesis reports the study, from design to deployment and evaluation, of a general purpose and configurable platform dedicated to stereo-vision perception in underwater environments. Several aspects related to the peculiar environment characteristics have been taken into account during all stages of system design and evaluation: depth of operation and light conditions, together with water turbidity and external weather, heavily impact on perception capabilities. The vision platform proposed in this work is a modular system comprising off-the-shelf components for both the imaging sensors and the computational unit, linked by a high performance ethernet network bus. The adopted design philosophy aims at achieving high flexibility in terms of feasible perception applications, that should not be as limited as in case of a special-purpose and dedicated hardware. Flexibility is required by the variability of underwater environments, with water conditions ranging from clear to turbid, light backscattering varying with daylight and depth, strong color distortion, and other environmental factors. Furthermore, the proposed modular design ensures an easier maintenance and update of the system over time. Performance of the proposed system, in terms of perception capabilities, has been evaluated in several underwater contexts taking advantage of the opportunity offered by the MARIS national project. Design issues like energy power consumption, heat dissipation and network capabilities have been evaluated in different scenarios. Finally, real-world experiments, conducted in multiple and variable underwater contexts, including open sea waters, have led to the collection of several datasets that have been publicly released to the scientific community. The vision system has been integrated in a state of the art AUV equipped with a robotic arm and gripper, and has been exploited in the robot control loop to successfully perform underwater grasping operations.