Abstract
Motivation is the key to academic success. In the case of engineering, autonomous project teamwork guided by ethics and sustainability concerns acts as a major student motivator. Moreover, it empowers students to become lifelong learners and agents of sustainable development. Engineering schools can thus address simultaneously these two essential education goals – learning and academic success – by challenging students to find innovative, sustainable solutions in a learner-centred set-up.This paper describes how the European Project Semester (EPS), a capstone engineering programme offered by the Instituto Superior de Engenharia do Porto (ISEP), combines challenge-based learning, ethics and sustainability-driven problem-solving, and international multidisciplinary teamwork to achieve both goals.

Abstract
Extreme summer temperatures in the Iberia Peninsula are analyzed from ERA5-Land reanalysis data based on an extreme value mixture model combining a Normal distribution for the bulk distribution (i.e. for the non-extreme values) and a Generalized Pareto Distribution for the extremes in the upper tail. This approach allows to treat the threshold of temperature exceedances as being one of model parameters rather than fixed a priori, enabling to take into account its corresponding uncertainty. Extreme value mixture models are estimated individually for each location, and the analysis is performed separately for two distinct periods, namely from 1981 to 2000 and from 2000 to 2019, respectively. The results show significant differences in the extreme value mixture models for the two periods, and in their corresponding 20-year return levels. The mean of the bulk distribution of summer maximum temperature increases significantly, particularly in Eastern Iberia. The largest differences in the tails of the data distribution between the two periods occur in the eastern Mediterranean area, and are characterized by a significant increase in the threshold for temperature exceedances and in their corresponding return levels.

Abstract

Abstract
<p><br>The automatic classification of peaks in gamma radiation time series is relevant for both scientific and practical applications. From the practical perspective, the classification of  peaks is fundamental for  early-warning systems for radiation protection and detection of radioactive material. From the scientific point of view, peaks in gamma radiation are often driven by precipitation  and consequent  scavenging of airborne radon progeny radionuclides to the ground (mainly Pb-214 and Bi-214). Thus measurements of gamma radiation at the earth's surface have the potential to provide information on micro-physical processes occurring high above in the clouds, as the dominant source of radon progeny is thought to be associated with in-cloud processes – nucleation scavenging and interstitial aerosol collection by cloud or rain droplets. </p><p>The present study addresses the classification of peaks in high-resolution (1-minute) gamma radiation time series from the GRM (Gamma Radiation Monitoring) campaign, which is being carried out since 2015 at the Eastern North Atlantic (ENA) station of the ARM (Atmospheric Radiation Measurements) programme. In addition to the gamma time series, precipitation information from laser disdrometer measurements is considered, including rain rate, liquid water content, median drop diameter and droplet concentration. Diverse machine learning algorithms are examined with the goal to identify and classify gamma peaks driven by precipitation events, and further examine the association between precipitation characteristics and the resulting gamma radiation peak on the ground.</p><p> </p>

Abstract
<p>Radon and its progeny are well-documented sources of natural radioactivity which can be used as benchmarks for testing a novel ionisation detector. The miniaturised ionisation detector was deployed aboard the NRP Sagres on a SAIL mission in July 2021 which travelled between the Açores and Lisbon in the North Atlantic Ocean. On its voyage, the detector profiled natural background radiation and in-directly detected cosmic ray muons, providing both spectroscopic energy discrimination and count rate data. The detector was simultaneously run with a NaI(Tl) gamma ray counter and other meteorological instruments.</p><p>The small form factor and low-power detector, composed of a 1x1x0.8 cm<sup>3 </sup>CsI(Tl) microscintillator coupled to a PiN photodiode, was able to identify gamma peaks from Bi-214 and K-40, having been calibrated using laboratory gamma sources up to 1.3 MeV. This research aims to investigate the performance of the ionisation detector and behaviour of discrete gamma energies over the duration of the voyage. Additionally, we will show a comparison of the CsI(Tl) based ionisation detector against the gamma ray counter which features a larger NaI(Tl) scintillator.</p>

Abstract
The use of robots is widely spread across the industry. It is paramount that the robot end-effector tracks a pre-defined trajectory with the lowest energy loss. To contribute to the solution of this problem, the robot trajectory is defined using a tracking parameter which is optimised using the Matlab (R) fminunc function and the Particle Swam Optimisation algorithm. This approach was tested for a case study with the energy loss being reduced in approximately 96.15%.