Search

The physical-chemical monitoring of vast oceanic areas aims at assessing the status and evolution of the environmental resource for its exploration, protection and/or better understanding. However, the interpretation of monitoring data is affected by ocean seasonality and heterogeneity, and by the quality of sampling and characterization tools used to study the environment. All these factors contribute to the uncertainty of collected information that should be expressed in determined parameter values or trends. A trend of a studied parameter quantified by values difference is significant if the observed absolute value of the difference is larger than their expanded uncertainty. The correlation of studied parameters, useful for their interpretation, is equality affected by the mentioned sources of uncertainty. This work describes the metrologically sound evaluation of trends and correlations of physicochemical parameters of vast oceanic areas where all uncertainty sources affecting the information are considered by simulating their complex impact by the Monte Carlo Method. The described methodology was successfully used to study the impact of summer upwelling in an 800 km2 coastal area offshore two large cities in Portugal. Nutrients, conductivity, salinity and temperature trends and correlations are distinguished from system heterogeneity, sampling and sample analysis uncertainty for a 99% confidence level. [Display omitted] • Simulation of the impact of the heterogeneity, and sampling and analysis uncertainty on oceanic area characterisation. • Mean value of nutrients and fundamental oceanographic parameters from a vast oceanic area reported with uncertainty. • Evaluation of the variation of mean composition values between October 2019 and May 2018 for a 99% confidence level. • Correlations between nutrients and fundamental oceanic parameters evaluated taking information uncertainty into account. • Upwelling trends and correlations confirmed from modelled spatial variation of a vast oceanic area with uncertainty. [ABSTRACT FROM AUTHOR]

Vehicle localisation is an important and challenging task in achieving autonomous driving. This work presents a box particle filter framework for vehicle self-localisation in the presence of sensor and map uncertainties. The proposed feature-refined box particle filter incorporates line features extracted from a multi-layer Light Detection And Ranging (LiDAR) sensor and information from OpenStreetMap to estimate vehicle states. A particle weight balance strategy is incorporated to account for the OpenStreetMap positional inaccuracy, which is assessed by comparing it to a high definition road map. The performance of the proposed framework is evaluated on a LiDAR dataset and compared with box particle filter variants. Experimental results show that the proposed framework achieves respectively 10% and 53% localisation performance improvement with reduced box volumes of 25% and 41%, when compared with the state-of-the-art interval analysis based box regularisation particle filter and the box particle filter. • The FRBPF is proposed to deal with OpenStreetMap and sensor data uncertainties. • A contraction algorithm is developed to reduce the volume of box particles. • Theoretical proofs about the features-refined contractions are derived. • A weight balance strategy is designed to improve the performance of FRBPF. [ABSTRACT FROM AUTHOR]