Pedro J. Navarro

A cost-effective robotic solution for the cleaning of ships' hulls

Hull cleaning before repainting is a key operation in the maintenance of ships. For more than a decade, a means to improve this operation has been sought through robotization and the use of different techniques such as grit blasting and ultra high pressure water jetting. Despite this, it continues to be standard practice in shipyards that this process is carried out manually. This paper presents a family of robots that aims to offer important improvements to the process as well as satisfying, to a great extent, all the operative requirements of efficiency, security, and respect for the environment that shipyards nowadays demand. It is described the family of devices with emphasis on the mechanical design. This set consists of two vertical robotic towers and a robot climber. In addition, it is shown the control architecture of the global system. Finally, operative results are presented together with a comparison between the performance achieved in shipyards through the use of these robots and those obtained with a manual process.

A Machine Learning Approach to Pedestrian Detection for Autonomous Vehicles Using High-Definition 3D Range Data

This article describes an automated sensor-based system to detect pedestrians in an autonomous vehicle application. Although the vehicle is equipped with a broad set of sensors, the article focuses on the processing of the information generated by a Velodyne HDL-64E LIDAR sensor. The cloud of points generated by the sensor (more than 1 million points per revolution) is processed to detect pedestrians, by selecting cubic shapes and applying machine vision and machine learning algorithms to the XY, XZ, and YZ projections of the points contained in the cube. The work relates an exhaustive analysis of the performance of three different machine learning algorithms: k-Nearest Neighbours (kNN), Naïve Bayes classifier (NBC), and Support Vector Machine (SVM). These algorithms have been trained with 1931 samples. The final performance of the method, measured a real traffic scenery, which contained 16 pedestrians and 469 samples of non-pedestrians, shows sensitivity (81.2%), accuracy (96.2%) and specificity (96.8%).

From Teleo-Reactive specifications to architectural components: A model-driven approach

The Teleo-Reactive approach designed by N.J. Nilsson offers a high-level programming model that permits the development of reactive systems, such as robotic vehicles. Teleo-Reactive programs are written in a manner that allows engineers to define the behaviour of the system while taking into account goals and changes in the state of the environment. This article presents a systematic approach that makes it possible to derive architectural models, with structural descriptions and behaviour, from Teleo-Reactive Programs. The development of reactive systems can therefore benefit significantly from a combination of two approaches: (1) the Teleo-Reactive approach, which is oriented towards a description of the system from the standpoint of the goals identified and the state of the environment and (2) the architectural approach, which is oriented towards the design of component-based software, in which decisions are conditioned by the need to reuse already tested solutions. The integration of this work into a development environment that allows code to be generated via model transformations opens up new possibilities in the development of this type of systems. The proposal is validated through a case study that is representative of the domain, and a survey carried out with post-graduate students.

Machine Learning and Computer Vision System for Phenotype Data Acquisition and Analysis in Plants

Phenomics is a technology-driven approach with promising future to obtain unbiased data of biological systems. Image acquisition is relatively simple. However data handling and analysis are not as developed compared to the sampling capacities. We present a system based on machine learning (ML) algorithms and computer vision intended to solve the automatic phenotype data analysis in plant material. We developed a growth-chamber able to accommodate species of various sizes. Night image acquisition requires near infrared lightning. For the ML process, we tested three different algorithms: k-nearest neighbour (kNN), Naive Bayes Classifier (NBC), and Support Vector Machine. Each ML algorithm was executed with different kernel functions and they were trained with raw data and two types of data normalisation. Different metrics were computed to determine the optimal configuration of the machine learning algorithms. We obtained a performance of 99.31% in kNN for RGB images and a 99.34% in SVM for NIR. Our results show that ML techniques can speed up phenomic data analysis. Furthermore, both RGB and NIR images can be segmented successfully but may require different ML algorithms for segmentation.

An educational tool for implementing reactive systems following a goal-driven approach

This article presents a novel computer‐based tool that has proven useful for students to both implement and validate the design of reactive systems using the Teleo‐Reactive approach. The design and implementation of the tool and the proposed laboratory exercises are described. The tool, then, is a useful framework for tackling the synthesis of control systems for reactive systems. The results of the tests on efficiency with undergraduate students are also discussed.© 2012 Wiley Periodicals, Inc. Comput Appl Eng Educ 22:764–773, 2014; View this article online at wileyonlinelibrary.com/journal/cae; DOI 10.1002/cae.21568

A sensor system for detection of hull surface defects.

This paper presents a sensor system for detecting defects in ship hull surfaces. The sensor was developed to enable a robotic system to perform grit blasting operations on ship hulls. To achieve this, the proposed sensor system captures images with the help of a camera and processes them in real time using a new defect detection method based on thresholding techniques. What makes this method different is its efficiency in the automatic detection of defects from images recorded in variable lighting conditions. The sensor system was tested under real conditions at a Spanish shipyard, with excellent results.

Automated Visual Inspection of Ship Hull Surfaces Using the Wavelet Transform

A new online visual inspection technique is proposed, based on a wavelet reconstruction scheme over images obtained from the hull. This type of visual inspection to detect defects in hull surfaces is commonly carried out at shipyards by human inspectors before the hull repair task starts. We propose the use of Shannon entropy for automatic selection of the band for image reconstruction which provides a low decomposition level, thus avoiding excessive degradation of the image, allowing more precise defect segmentation. The proposed method here is capable of on-line assisting to a robotic system to perform grit blasting operations over damage areas of ship hulls. This solution allows a reliable and cost-effective operation for hull grit spot blasting. A prototype of the automated blasting system has been developed and tested in the Spanish NAVANTIA shipyards.

Industrial-Like Vehicle Platforms for Postgraduate Laboratory Courses on Robotics

The interdisciplinary nature of robotics allows mobile robots to be used successfully in a broad range of courses at the postgraduate level and in Ph.D. research. Practical industrial-like mobile robotic demonstrations encourage students and increase their motivation by providing them with learning benefits not achieved with traditional educational robotic platforms. This paper presents VEGO, an industrial-like modular vehicle platform for robotic education with an appropriate infrastructure that has been demonstrated to be very useful at the postgraduate level. Besides learning engineering concepts, in performing industrial-like exercises, students develop valuable skills such as teamwork and the capacity to solve problems similar to those they may encounter in a real industrial environment. The developed infrastructure represents a valuable platform for robotic education that can be used in many different disciplines as a way to demonstrate how to cope with the difficulties and challenges related to the development of industrial infrastructure systems. The platform evaluation proved its ability to inculcate the expected engineering skills. A novel approach is adopted through the use of multidisciplinary and close-to-industrial-reality platforms developed under an incremental approach and using an open and customizable structure.

Teleoperated Visual Inspection System for Hull Spot-Blasting

This paper presents a robotized teleoperated visual inspection system for spot-blasting applied to hull cleaning in ship maintenance tasks. It consists of a cleaning head, a robot body, a remote control unit, and a teleoperation platform connected to a computer vision system. This solution allows a reliable and cost effective operation for hull grit spot-blasting. A prototype of this robot has been developed and tested in the Spanish IZAR shipyards

Plant phenomics: an overview of image acquisition technologies and image data analysis algorithms

Abstract The study of phenomes or phenomics has been a central part of biology. The field of automatic phenotype acquisition technologies based on images has seen an important advance in the last years. As with other high-throughput technologies, it addresses a common set of problems, including data acquisition and analysis. In this review, we give an overview of the main systems developed to acquire images. We give an in-depth analysis of image processing with its major issues and the algorithms that are being used or emerging as useful to obtain data out of images in an automatic fashion.