Study of risky situations using experimental data coming from an instrumented bicycle

The aim of this master’s thesis is to study the risky situations of the cyclist when they interact with road infrastructure and other road users as well as the influence of speed on safety. This research activity is linked with the SAFERUP (Sustainable, Accessible, Resilient, and Smart Urban Pavement) European funded project where one of the doctoral candidate has performed experiments on the bicycle simulation at the Gustave Eiffel university in the PICS-L laboratory (Paris) and instrumented bicycle at the Stockholm (Sweden). The approach of the experiment was to hire a number of people who have participated in the riding of the Instrumented bicycle (Stockholm) and bicycle simulator (PICS-L) which were developed by attaching different sensors and devices to measure important parameters of the bicycle riding and their data was collected to analysis in order to understand the behavior of the cyclist to improve the safety. In addition, a mobile eye tracker wore by participants to record the real experiment scenario, and after the end of the trip, each participant shared their remarks regarding their experience of bicycle riding according to different portions of the road infrastructure. In this research main focus is to analyze the relevant data such as speed profiles, video recordings and questionnaire surveys from the instrumented bicycle experiment. In fact, critical situations, where there was a higher probability, were compared with the subjective evaluation of the participant to be conscious of the issues related to the safety and comfort of the cyclist in different road characteristics.

ICE Emissions Estimation through Machine-learning techniques to support real-world R&D experiments

The emissions estimation, both during homologation and standard driving, is one of the new challenges that automotive industries have to face. The new European and American regulation will allow a lower and lower quantity of Carbon Monoxide emission and will require that all the vehicles have to be able to monitor their own pollutants production. Since numerical models are too computationally expensive and approximated, new solutions based on Machine Learning are replacing standard techniques. In this project we considered a real V12 Internal Combustion Engine to propose a novel approach pushing Random Forests to generate meaningful prediction also in extreme cases (extrapolation, very high frequency peaks, noisy instrumentation etc.). The present work proposes also a data preprocessing pipeline for strongly unbalanced datasets and a reinterpretation of the regression problem as a classification problem in a logarithmic quantized domain. Results have been evaluated for two different models representing a pure interpolation scenario (more standard) and an extrapolation scenario, to test the out of bounds robustness of the model. The employed metrics take into account different aspects which can affect the homologation procedure, so the final analysis will focus on combining all the specific performances together to obtain the overall conclusions.

Characterisation of the sewage microbiome in four European cities throughout the COVID-19 pandemic by means of ecological and network modelling

The purpose of this work was to investigate possible patterns occurring in the sewage bacterial content of four cities (Bologna, Budapest, Rome, Rotterdam) over time (March 2020 - November 2021), also considering the possible effects of the lockdown periods due to the COVID-19 pandemic. The sewage metagenomics data were provided within VEO (Versatile Emerging infectious disease Observatory) project. The first analysis was the evaluation of the between samples diversity, looking for (dis)similarities among the cities, as well as among different time periods (seasonality). To this aim, we computed both similarity networks and Principal Coordinate Analysis (PCoA) plots based on the Bray-Curtis metric. Then, the alpha-biodiversity of the samples was estimated by means of different diversity indices. By looking at the temporal behaviour of the biodiversity in the four cities, we noticed an abrupt decrease in both Rome and Budapest in the Summer of 2020, that is related to: the prevalence of some species when the minimum occurred, and the change in correlations among species (studied via correlation networks), which is enriched in the period of minimum biodiversity. Rotterdam samples seem to be very different with respect to those from the other cities, as confirmed by PCoA. Moreover, the Rotterdam time series is proved to be stable and stationary also in terms of biodiversity. The low variability in the Rotterdam samples seems to be related to the species of Pseudomonas genus, which are highly variable and plentiful in the other cities, but are not among the most abundant in Rotterdam. Also, we observed that no seasonality effect emerged from the time series of the four cities. Regarding the impact of lockdown periods due to the COVID-19 pandemic, from the limited data available no effect on the time series considered emerges. More samples will be soon available and these analyses will be performed also on them, so that the possible effects of lockdowns may be studied.

Monitoring of macro- and microlitter in the Venice lagoon and coastal area

Marine litter and plastics are a significant and growing marine contaminant that has become a global problem. Macrolitter is subject to fragmentation and degradation due to physical, chemical and biological processes, leading to the formation of micro-litter, the so-called microplastics. The purpose of this research is to assess marine litter pollution by using remote sensing tools to identify areas of macrolitter accumulation and to evaluate the concentrations of microplastics in different environmental matrices: water, sediment and biota (i.e. mussels and fish) and to contribute to the European project MAELSTROM (Smart technology for MArinE Litter SusTainable RemOval and Management). The aim is to monitor the presence of macro- and microlitter at two sites of the Venice coastal area: an abandoned mussel farm at sea and a lagoon site near the artificial Island of Sacca Fisola; The results showed that both study areas are characterised by high amounts of marine litter, but the type of observed litter is different. In fact, in the mussel farm area, most of the litter is linked to aquaculture activities (ropes, nets, mooring blocks and floating buoys). In the Venice lagoon site, the litter comes more from urban activities and from the city of Venice (car tyres, crates, wrecks, etc.). Microplastics is present in both sites and in all the analysed matrices. Generally, higher microplastics concentrations were found at Sacca Fisola (i.e., in surface waters, mussels and fish). Moreover, some differences were also observed in shapes and colours comparing the two sites. At Sacca Fisola, white irregular fragments predominate in water samples, blue filaments in sediment and mussels, and transparent irregular fragments in fish. At the Mussel Farm, blue filaments predominate in water, sediment and mussels, while flat black fragments predominate in fish. These differences are related to the different types of macrolitter that characterised the two areas.