Miguel Aguiar's research while affiliated with KTH Royal Institute of Technology and other places

... For that reason, an area anomaly analysis was carried out by subtracting the sea outfall daily area of the respective month from the monthly sea outfall average area. Nevertheless, to comprehend the influence of the local meteo-oceanographic conditions on the outfall turbid plume's dispersion, sea surface elevation data were acquired for the Aveiro lagoon mouth from an existing hydrodynamic model [38], and hourly wind speed and direction were acquired from a WRF model (the Weather Research and Forecasting (WRF) Model is a mesoscale numerical weather prediction system used for atmospheric research and operational forecasting applications), with a spatial resolution of 12 km [39]. For this reason, one example of the wastewater plume dispersing inward toward the Ria de Aveiro coastal lagoon, observed through ocean color data, will be addressed. ...

... Assuming that the reference value δ ref is known, we used the previously developed control law as in [46]. There, we define the control law as ...

... The paradigm of physics-informed deep learning [13][14], which integrates the dual advantages of model driven model and data driven model, has received increasing attention since its introduction in recent years, and has also sparked a research wave in transportation fields such as traffic state estimation (TSE) [15][16]. Combining the basic laws of traffic flow and deep learning methods, such as Lighthill-Whitham-Richards (LWR) model and ANN model, PIDL can accurately and timely estimated traffic state utilizing limited and noisy sparse data [17][18]. ...

... A 3D implementation was established for the Sado estuary by applying the FLOW module and, for the first time, the WAQ module of the numerical model Delft3D (Deltares, 2021(Deltares, , 2014. A structured (curvilinear) spherical grid was generated based on the applications developed by Aguiar et al. (2020) and Ribeiro et al. (2016). The main differences to the previous implementations are the increase of the grid resolution (from ∼1500 m to ∼300 m on average in the offshore region and from ∼100 m to ∼80 m in the estuary body) and the inclusion of the intertidal areas in the grid. ...

... Many authors have proposed interesting controllers or observers such as [1][2][3][4][5][6][7][8][9]; however, in most of the studies, the robot models are assumed to be controllable or observable without any proof. A method to determine the controllability of a robot model is important because a robot model that is assured to be controllable can guarantee the existence of a controller to reach one of the objectives such as regulation, tracking, disturbance rejection, etc. ...

... Critically, our efforts are in the open ocean, well beyond the confines of the coastal zone, where there was no guarantee of being able to find the STF frontal zone. Equally, our work involves a rich legacy of software infrastructure used across multiple field experiments (Das et al., 2011;Faria et al., 2014b;de Sousa et al., 2016a;Py et al., 2016;de Sousa et al., 2016b;Chrpa et al., 2017;Fossum et al., 2018;Ferreira et al., 2019;Costa et al., 2018;Dias et al., 2020), and not a bespoke method, emphasizes propelled vehicles with substantial payload and computational abilities which makes for adaptations in situ. ...

... This work builds upon the results in Aguiar et al. [1], who presented an efficient method and implementation for computing time-optimal trajectories for AUVs using high-resolution water velocity models. Using the LSTS -Underwater Systems and Technology Laboratory -software toolchain [2], we incorporate the method into a networked vehicle system, optimizing vehicle deployments and enabling the operationalization of marine pollution incident in situ verification. ...

... Many authors have proposed interesting controllers or observers such as [1][2][3][4][5][6][7][8][9]; however, in most of the studies, the robot models are assumed to be controllable or observable without any proof. A method to determine the controllability of a robot model is important because a robot model that is assured to be controllable can guarantee the existence of a controller to reach one of the objectives such as regulation, tracking, disturbance rejection, etc. ...