No abstract

The present work describes the testing and application of a low cost plastic optical fiber sensor on the monitoring of groundwater levels. Two sensors with different grove depth (½ and ¼ of the core thickness) and a resolution of 20 cm along 2 m of the optical fiber were produced and tested. The sensors were tested under two experimental setups: water level variation (increasing and decreasing of the water level) and groundwater increase simulation, in a soil column. The analysis of the optical signal's amplitude and its variations due to the increasing or decreasing of water level showed that both tested sensors presented an appropriate performance and adequate sensibility to groundwater level variation and, therefore, can be used for in situ applications of monitoring.

Advances in the development of sensors, data processing systems, and numerical models have motivated the implementation of structural health monitoring (SHM) specially focused on the assessment of structural safety. Thus, this work presents a literature review about SHM platforms, especially from 1993 to 2015. In this way, a short history review about the recent advances on SHM, mainly related with dynamic monitoring, was summarized, and a benchmark and the main guidelines related with SHM platforms were also included in this review. Some case studies are also described here. Special attention was given to SHM platforms, and a method for their classification (an extension of Rytter's method) is presented. In addition, experiences related with heritage constructions, specially focused on maintenance, were included in this work. In the final section, some observations are made about the new prospects for SHM. The recent advances on SHM platforms contributed to the development of adaptive systems and to the cost reduction of the monitoring systems implementation, allowing the increase of its application in real structures. However, the monitoring systems should be implemented, optimizing all the available sensing technologies.

Abstract. Agent programming is mostly a symbolic discipline and, as such, draws little benefits from probabilistic areas as machine learning and graphical models. However, the greatest objective of agent research is the achievement of autonomy in dynamical and complex environments -a goal that implies embracing uncertainty and therefore the entailed representations, algorithms and techniques. This paper proposes an innovative and conflict free two layer approach to agent programming that uses already established methods and tools from both symbolic and probabilistic artificial intelligence. Moreover, this method is illustrated by means of a widely used agent programming example, GoldMiners.

No abstract