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.

Stratigrafia fisica ed analisi di facies dei depositi torbiditici della Formazione delle Arenarie del Monte Cervarola tra la Val Secchia e la Val Scoltenna (Aquitaniano-Burdigaliano, Appennino Settentrionale)

The Cervarola Sandstones Formation (CSF), Aquitanian-Burdigalian in age, was deposited in an elongate, NW-stretched foredeep basin formed in front of the growing Northern Apennines orogenic wedge. The stratigraphic succession of the CSF, in the same way of other Apennine foredeep deposits, records the progressive closure of the basin due to the propagation of thrust fronts toward north-east, i.e. toward the outer and shallower foreland ramp. This process produce a complex foredeep characterized by synsedimentary structural highs and depocenters that can strongly influence the lateral and vertical turbidite facies distribution. Of consequence the main aim of this work is to describe and discuss this influence on the basis of a new high-resolution stratigraphic framework performed by measuring ten stratigraphic logs, for a total thickness of about 2000m, between the Secchia and Scoltenna Valleys (30km apart). In particular, the relationship between the turbidite sedimentation and the ongoing tectonic activity during the foredeep evolution has been describe through various stratigraphic cross sections oriented parallel and perpendicular to the main tectonic structures. On the basis of the high resolution physical stratigraphy of the studied succession, we propose a facies tract and an evolutionary model for the Cervarola Sandstones in the studied area. Thanks to these results and the analogies with others foredeep deposits of the northern Apennines, such as the Marnoso-arenacea Formation, the Cervarola basin has been interpreted as a highly confined foredeep controlled by an intense synsedimentary tectonic activity. The most important evidences supporting this hypothesis are: 1) the upward increase, in the studied stratigraphic succession (about 1000m thick), of sandstone/mudstone ratio, grain sizes and Ophiomorpha-type trace fossils testifying the high degree of flow deceleration related to the progressive closure and uplift of the foredeep. 2) the occurrence in the upper part of the stratigraphic succession of coarse-grained massive sandstones overlain by tractive structures such as megaripples and traction carpets passing downcurrent into fine-grained laminated contained-reflected beds. This facies tract is interpreted as related to deceleration and decoupling of bipartite flows with the deposition of the basal dense flows and bypass of the upper turbulent flows. 3) the widespread occurrence of contained reflected beds related to morphological obstacles created by tectonic structures parallel and perpendicular to the basin axis (see for example the Pievepelago line). 4) occurrence of intra-formational slumps, constituted by highly deformed portion of fine-grained succession, indicating a syn-sedimentary tectonic activity of the tectonic structures able to destabilize the margins of the basin. These types of deposits increase towards the upper part of the stratigraphic succession (see points 1 and 2) 5) the impressive lateral facies changes between intrabasinal topographic highs characterized by fine-grained and thin sandstone beds and marlstones and depocenters characterized by thick to very thick coarse-grained massive sandstones. 6) the common occurrence of amalgamation surfaces, flow impact structures and mud-draped scours related to sudden deceleration of the turbidite flows induced by the structurally-controlled confinement and morphological irregularities. In conclusion, the CSF has many analogies with the facies associations occurring in other tectonically-controlled foredeeps such as those of Marnoso-arenacea Formation (northern Italy) and Annot Sandstones (southern France) showing how thrust fronts and transversal structures moving towards the foreland, were able to produce a segmented foredeep that can strongly influence the turbidity current deposition.