Abstract
A sensing head based on a hollow-core photonic crystal fiber for in-reflection measurement of micro-displacements is presented. The sensing structure takes advantage of the multimodal behavior of a short segment of hollow-core photonic crystal fiber in-reflection, being spliced to a single mode fiber at its other end. A modal interferometer is obtained when the sensing head is close to a mirror, through which displacement is measured.

Abstract
Despite the benefits of precision agriculture and precision viticulture production systems, its rate of adoption in the Portuguese Douro Demarcated Region remains low. We believe that one way to raise it is to address challenging real-world problems whose solution offers a clear benefit to the viticulturist. For example, one of the most demanding tasks in wine making is harvesting. Even for humans, the environment makes grape detection difficult, especially when the grapes and leaves have a similar color, which is generally the case for white grapes. In this paper, we propose a system for the detection and location, in the natural environment, of bunches of grapes in color images. This system is able to distinguish between white and red grapes, and at the same time, it calculates the location of the bunch stem. The system achieved 97% and 91% correct classifications for red and white grapes, respectively.

Abstract
In this paper the integer-valued autoregressive model of order one, contaminated with additive outliers is studied in some detail. Moreover, parameter estimation is also addressed. Supposing that the timepoints of the outliers are known but their sizes are unknown, we prove that the conditional least squares (CLS) estimators of the offspring and innovation means are strongly consistent. In contrast, however, the CLS estimators of the outliers' sizes are not strongly consistent, although they converge to a random limit with probability 1. We also prove that the joint CLS estimator of the offspring and innovation means is asymptotically normal. Conditionally on the values of the process at the timepoints neighboring to the outliers' occurrences, the joint CLS estimator of the sizes of the outliers is also asymptotically normal.

Abstract
This paper is concerned with the effect of sediment accumulation on shore platform development. Boulder accumulations are common on the granitic shore platforms of Galicia, northwestern Spain. Boulders are produced by erosion of shore platforms and of cliffs consisting of cold-climate deposits from the last glacial period. Measurements were made of the long axis length of more than 800 boulders, and additionally of the short and intermediate axes of 340 of these boulders, as well as of their orientation and gradient. There were two study areas. The boulders on the Barbanza Peninsula are generally a little smaller than those in southern Galicia with, respectively; mean long axis lengths of 0.98 and 1.14, and masses of 1.06 and 1.59 t. There are also some isolated, very coarse boulders and megaclasts in southern Galicia. The distribution and extent of the deposits and boulder imbrication and orientation testify to the high levels of wave energy produced by northwesterly and westerly storms in this region. Although the boulders, as well as the underlying shore platforms, were inherited, in part, from previous interglacial stages, some boulder detachment and movement is occurring today during storms, when significant deep water wave heights exceed 8 to 10 m. Despite some abrasion of the shore platforms, the primary effect of large boulder accumulations is protective. The role of sediment on shore platforms has been neglected, but this study suggests that because of arrested development under thick accumulations, platform gradient in areas with abundant sediment increases with the grain size of the material. The occurrence and type of sediment on shore platforms may therefore help to explain the distribution of sloping and subhorizontal platforms under different morphogenic and geological conditions.

Abstract
It is becoming more and more common to use Autonomous Underwater Vehicles to perform tasks underwater. The use of this vehicles is affordable and its use doesn't raise any significant risk nor does it requires any human intervention. The traditional applications for the use of such vehicles were related with bathymetric tasks. But nowadays AUVs are being more and more used for variety of missions in open water environments, including the inspection of underwater structures and environmental monitoring in diverse oceanographic expeditions. Following some previous work, this paper addresses the problem of bottom following by an Autonomous Underwater Vehicle in an environment which is not previously known. In particular, the focus is on integrating a reactive behaviour based on environment sensing, with the on-board navigation software of the MARES AUV. For this, a guidance algorithm will provide the necessary pitch and depth references to the control layer of the vehicle. While the altitude towards the seabed can be measured with an altimeter, the pitch reference values are based on realtime estimation of the slope of the seabed. By doing so, it is possible to control the vehicle in a way that it will always maintain a constant attitude towards the bottom, and the trajectory followed will remain parallel to bottom, regardless of it's profile. © 2012 IEEE.

Abstract
The motivation for this work is the study and prediction of wind ramp events occurring in a large-scale wind farm located in the US Midwest. In this paper we introduce the SHRED framework, a stream-based model that continuously learns a discrete HMM model from wind power and wind speed measurements. We use a supervised learning algorithm to learn HMM parameters from discretized data, where ramp events are HMM states and discretized wind speed data are HMM observations. The discretization of the historical data is obtained by running the SAX algorithm over the first order variations in the original signal. SHRED updates the HMM using the most recent historical data and includes a forgetting mechanism to model natural time dependence in wind patterns. To forecast ramp events we use recent wind speed forecasts and the Viterbi algorithm, that incrementally finds the most probable ramp event to occur. We compare SHRED framework against Persistence baseline in predicting ramp events occurring in short-time horizons, ranging from 30 minutes to 90 minutes. SHRED consistently exhibits more accurate and cost-effective results than the baseline. © 2012 Springer-Verlag Berlin Heidelberg.