Abstract
A significant challenge in asset management is the selection of investment projects for infrastructures, which often relies on subjective judgement and lacks structured decision support methods. This challenge is particularly complex in water systems due to the diverse and heterogeneous nature of the components requiring investment. While the infrastructure value index (IVI) is widely used to characterise assets and support investment decisions in the water sector, its application in optimisation models for generating efficient project portfolios remains unexplored. To address this research gap, this study introduces optimisation models for generating investment portfolio plans in water systems' asset management. The proposed approach includes two mixed-integer linear programming (MILP) models that determine optimal solutions and an evolutionary algorithm that offers sub-optimal alternative investment selection plans to provide decision-makers with additional choices for balancing optimal outcomes. The primary contribution of this research is the combined utilisation of MILP and evolutionary algorithms, integrating the IVI into the decision-making process. These tools provide decision-makers with structured methods for defining investment plans and minimising the subjective elements typically associated with such processes. To illustrate the effectiveness of the models, a case study is presented involving a pumping station of a Portuguese water company. The results demonstrate the practical application and benefits of the proposed approach in optimising investment decisions. This research contributes to advancing asset management practices by integrating quantitative optimisation techniques and leveraging the IVI, thereby enhancing the objectivity and efficiency of investment planning in water systems' asset management.

Abstract
In this paper, a solution to monitor the location of humans during their activity in the agriculture sector with the aim to boost productivity and efficiency is provided. Our solution is based on map-matching methods, that are used to track the path spanned by a worker along a specific activity in an agriculture culture. Two different cultures are taken into consideration in this study olives and vines. We leverage the symmetry of the geometry of these cultures into our solution and divide the problem three-fold initially, we estimate a path of a worker along the fields, then we apply the map-matching to such path and finally, a post-processing method is applied to ensure local continuity of the sequence obtained from map-matching. The proposed methods are experimentally evaluated using synthetic and real data in the region of Mirandela, Portugal. Evaluation metrics show that results for synthetic data are robust under several sampling periods, while for real-world data, results for the vine culture are on par with synthetic, and for the olive culture performance is reduced.

Abstract
Mobile robot platforms capable of operating safely and accurately in dynamic environments can have a multitude of applications, ranging from simple delivery tasks to advanced assembly operations. These abilities rely heavily on a robust navigation stack, which requires stable and accurate pose estimations within the environment. The wide range of AMR's applications and the characteristics of multiple industrial environments (indoor and outdoor) have led to the development of a flexible and robust robot software architecture that allows the fusion of different data sensors in real time. In this way, and in terms of localization, AMRs have greater precision when it comes to uncontrolled and unstructured environments. These complex environments feature a variety of dynamic and unpredictable elements, such as variable layouts, limited visibility, unstructured spaces, and uncertain terrain. This paper presents a multi-localization system for industrial mobile robots in complex and dynamic industrial scenarios, based on different localization technologies and methods that can interact together and simultaneously.

Abstract
The rapid proliferation of Internet of Things (IoT) systems, encompassing a wide range of devices and sensors with limited battery life, has highlighted the critical need for energy-efficient solutions to extend the operational lifespan of these battery-powered devices. One effective strategy for reducing energy consumption is minimizing the number and size of retransmitted packets in case of communication errors. Among the potential solutions, Incremental Redundancy Hybrid Automatic Repeat reQuest (IR-HARQ) communication schemes have emerged as particularly compelling options by adopting the best aspects of error control, namely, automatic repetition and variable redundancy. This work addresses the challenge by developing a simulator capable of executing and analysing several (H)ARQ schemes using different channel models, such as the Additive White Gaussian Noise (AWGN) and Gilbert-Elliott (GE) models. The primary objective is to compare their performance across multiple metrics, enabling a thorough evaluation of their capabilities. The results indicate that IR-HARQ outperforms alternative methods, especially in the presence of burst errors. Furthermore, its potential for further adaptation and enhancement opens up new ways for optimizing energy consumption and extending the lifespan of battery-powered IoT devices.

Abstract
By leveraging the advances in wireless communications networks and their ubiquitous nature, sensing through communication technologies has flourished in recent years. In particular, Human-to-Machine Interfaces have been exploiting WiFi IEEE 802.11 networks to obtain information that allows Human Activity Recognition. In this paper, we propose a classification model to perform Person Identification (PI) through Body Velocity Profile time series, obtained by combining Channel State Information containing gesture knowledge from multiple Access Points. Through this model, we investigate the impact of different gestures on PI classification performance and explore how informing the model about the input gesture can enhance classification accuracy. This information may enable the network to adjust to the absence of features capable of adequately characterizing the desired classes in certain gestures. A simplified stacking model is also presented, capable of combining the softmax outputs of K previously proposed individual models. By having the individual models' evaluations of a gesture and the gesture information relating to it, the number of gestures considered was shown to significantly improve the performance of the PI classification task. This enhancement increased 17% of the average F1 scores when compared to the individual model tested on the same data.

Abstract