Francesco Di Corato

The project V-fides: A new generation AUV for deep underwater exploration, operation and monitoring

The proposed work is in the framework of the V-Fides project, aiming at developing a new generation of agile, over-actuated, long endurance Autonomous Underwater Vehicles, for deep underwater exploration, operation and monitoring. The project is co-funded by Tuscany Region (Italy) and is developed by a team lead by WASS S.p.A. (Whitehead Sistemi Subacquei, Livorno) with the participation of several partners including two research institutions of the University of Pisa and Small-Medium Enterprises in the Pisa-Livorno area. The vehicle is a general purpose, 3000m depth rated underwater vehicle with highly maneuverability capabilities, which can operate both as AUV and ROV. The vehicle is equipped with seven thrusters, with asymmetric input-output characteristic, and with a sensors payload for autonomous navigation, composed by: a tactical grade Inertial Measurement Unit (IMU), a Doppler Velocity Logger (DVL), a depth sensor, a magnetic compass and an acoustic modem for underwater communication and localization. This contribution gives an overview of the developed general architecture of the Navigation and Control module of the vehicle, from the algorithmic and system implementation stand-points.

Fusing acoustic ranges and inertial measurements in AUV navigation: The Typhoon AUV at CommsNet13 sea trial

The paper presents some experimental results of autonomous underwater navigation, based on the fusion of acoustic and inertial measurements. The work is in the framework of the Thesaurus project, funded by the Tuscany Region, aiming at developing techniques for systematic exploration of marine areas of archaeological interest through a team of Autonomous Underwater Vehicles (AUVs). The test was carried out with one Typhoon vehicle, a 300m depth rated AUV with acoustic communication capabilities, during the CommsNet13 experiment, organized and scientifically coordinated by the NATO S&T Org. Ctr. for Maritime Research and Experimentation (CMRE, formerly NURC), with the participation of several research institutions. The fusion algorithm is formally casted into an optimal stochastic filtering problem, where the rough estimation of the vehicle position, velocity and attitude, are refined by using the depth measurement, the relative measurements available on the acoustic channel and the vehicle surge speed.

An Entropy—Like Approach to Vision—Aided Inertial Navigation

Abstract The paper presents a loosely coupled approach for the improvement of the state estimation in autonomous inertial navigation tasks, augmented via image–based relative motion estimation. The proposed approach uses a novel Pose Estimation technique based on the minimization of a Entropy–Like cost function which is robust by nature to the presence of noise and outliers in the visual features. A Indirect Kalman Navigation Filter is used, in the framework of stochastic cloning. The robust relative pose estimation given by our novel technique is used to feed a relative position fix to the navigation filter. Simulations results are presented and compared with the results obtained via the classical Iterative Closest Point approach.

A stochastic algorithm for automatic hand pose and motion estimation

In this paper, a novel, robust, and simple method for automatically estimating the hand pose is proposed and validated. The method uses a multi-camera optoelectronic system and a model-based stochastic algorithm. The approach is marker-based and relies on an Unscented Kalman Filter. A hand kinematic model is introduced for constraining relative marker’s positions and improving the algorithm robustness with respect to outliers and possible occlusions. The algorithm outputs are 3D coordinate measures of markers and hand joint angle values. To validate the proposed algorithm, a comparison with ground truths for angular and 3D coordinate measures is carried out. The comparative analysis shows the advantages of using the model-based stochastic algorithm with respect to standard processing software of optoelectronic cameras in terms of implementation simplicity, time consumption, and user effort. The accuracy is remarkable, with a difference of maximum 0.035rad and 4mm with respect to angular and 3D Cartesian coordinates ground truths, respectively.

A Robust Hand Pose Estimation Algorithm for Hand Rehabilitation

During a rehabilitation session, patient activity should be continuously monitored in order to correct wrong movements and to follow patient improvements. Therefore, the application of human motion tracking techniques to rehabilitation is finding more and more consensus. The aim of this paper is to propose a novel, low-cost method for hand pose estimation by using a monocular motion sensing device and a robust marker-based pose estimation approach based on the Unscented Kalman Filter. The hand kinematics is used to enclose geometrical constraints in the estimation process. The approach is applied for evaluating some significant kinematic parameters necessary for understanding human hand motor improvements during rehabilitation. In particular, the parameters evaluated for the hand fingers are joint positions, angles, Range Of Motion and trajectory. Moreover, the position, orientation and velocity of the wrist are estimated.

Combined Vision – Inertial Navigation with Improved Outlier Robustness

This paper describes a loosely coupled approach for the improvement of state estimation in autonomous inertial navigation, using image-based relative motion estimation for augmentation. The augmentation system uses a recently proposed pose estimation technique based on a Entropy-Like cost function, which was proven to be robust to the presence of noise and outliers in the visual features. Experimental evidence of its performance is given and compared to a state-of-the-art algorithm. Vision-inertial integrated navigation is achieved using an Indirect Kalman Navigation Filter in the framework of stochastic cloning, and the proposed robust relative pose estimation technique is used to feed a relative position fix to the navigation filter. Simulation and Experimental results are presented and compared with the results obtained via the classical RANSAC – based Direct Linear Transform approach.

An Innovative Navigation Strategy for Autonomous Underwater Vehicles: An Unscented Kalman Filter Based Approach

Developing reliable navigation strategies is mandatory in the field of Underwater Robotics and in particular for Autonomous Underwater Vehicles (AUVs) to ensure the correct achievement of a mission. Underwater navigation is still nowadays critical, e.g. due to lack of access to satellite navigation systems (e.g. the Global Positioning System, GPS): an AUV typically proceeds for long time intervals only relying on the measurements of its on-board sensors, without any communication with the outside environment. In this context, the filtering algorithm for the estimation of the AUV state is a key factor for the performance of the system; i.e. the filtering algorithm used to estimate the state of the AUV has to guarantee a satisfactory underwater navigation accuracy. In this paper, the authors present an underwater navigation system which exploits measurements from an Inertial Measurement Unit (IMU), Doppler Velocity Log (DVL) and a Pressure Sensor (PS) for the depth, and relies on either an Extended Kalman Filter (EKF) or an Unscented Kalman Filter (UKF) for state estimation. A comparison between the EKF approach, classically adopted in the field of underwater robotics and the UKF is given. These navigation algorithms have been experimentally validated through the data related to some sea tests with the Typhoon class AUVs, designed and assembled by the Department of Industrial Engineering of the Florence University (DIEF) for exploration and surveillance of underwater archaeological sites in the framework of the THESAURUS and European ARROWS projects. The comparison results are significant as the two filtering strategies are based on the same process and sensors models. At this initial stage of the research activity, the navigation algorithms have been tested offline. The presented results rely on the experimental navigation data acquired during two different sea missions: in the first one, Typhoon AUV #1 navigated in a Remotely Operated Vehicle (ROV) mode near Livorno, Italy, during the final demo of THESAURUS project (held in August 2013); in the latter Typhoon AUV #2 autonomously navigated near La Spezia in the framework of the NATO CommsNet13 experiment, Italy (held in September 2013). The achieved results demonstrate the effectiveness of both navigation algorithms and the superiority of the UKF without increasing the computational load. The algorithms are both affordable for online on-board AUV implementation and new tests at sea are planned for spring 2015.Copyright

Experimental Evaluation of a Visual-Inertial Navigation System with Guaranteed Convergence

This contribution presents a constraints-based loosely-coupled Augmented Implicit Kalman Filter approach to vision-aided inertial navigation that uses epipolar constraints as output map. The proposed approach is capable of estimating the standard navigation output (velocity, position and attitude) together with inertial sensor biases. An observability analysis is proposed in order to define the motion requirements for full observability of the system and asymptotic convergence of the parameter estimates. Simulations and experimental results are summarized that confirm the theoretical conclusions.