This page has only limited features, please log in for full access.
I can describe myself as a person that is passionated about his job and enthusiastic to learn new things. I have a strong curiosity for road safety and drivers' behavior. This led me to engage a PhD in such field, after earning the BSc and MSc in Civil and Environmental Engineering at the University of Udine. My main research topics concern the analysis and modeling of road safety by means of Machine Learning techniques, as well as the study of driving behaviors in different road traffic and environmental situations, also considering the driver engagement in cognitive secondary tasks.
Improving pedestrian safety at urban intersections requires intelligent systems that should not only understand the actual vehicle–pedestrian (V2P) interaction state but also proactively anticipate the event’s future severity pattern. This paper presents a Gated Recurrent Unit-based system that aims to predict, up to 3 s ahead in time, the severity level of V2P encounters, depending on the current scene representation drawn from on-board radars’ data. A car-driving simulator experiment has been designed to collect sequential mobility features on a cohort of 65 licensed university students who faced different V2P conflicts on a planned urban route. To accurately describe the pedestrian safety condition during the encounter process, a combination of surrogate safety indicators, namely TAdv (Time Advantage) and T2 (Nearness of the Encroachment), are considered for modeling. Due to the nature of these indicators, multiple recurrent neural networks are trained to separately predict T2 continuous values and TAdv categories. Afterwards, their predictions are exploited to label serious conflict interactions. As a comparison, an additional Gated Recurrent Unit (GRU) neural network is developed to directly predict the severity level of inner-city encounters. The latter neural model reaches the best performance on the test set, scoring a recall value of 0.899. Based on selected threshold values, the presented models can be used to label pedestrians near accident events and to enhance existing intelligent driving systems.
Matteo Miani; Matteo Dunnhofer; Christian Micheloni; Andrea Marini; Nicola Baldo. Surrogate Safety Measures Prediction at Multiple Timescales in V2P Conflicts Based on Gated Recurrent Unit. Sustainability 2021, 13, 9681 .
AMA StyleMatteo Miani, Matteo Dunnhofer, Christian Micheloni, Andrea Marini, Nicola Baldo. Surrogate Safety Measures Prediction at Multiple Timescales in V2P Conflicts Based on Gated Recurrent Unit. Sustainability. 2021; 13 (17):9681.
Chicago/Turabian StyleMatteo Miani; Matteo Dunnhofer; Christian Micheloni; Andrea Marini; Nicola Baldo. 2021. "Surrogate Safety Measures Prediction at Multiple Timescales in V2P Conflicts Based on Gated Recurrent Unit." Sustainability 13, no. 17: 9681.
An integrated approach based on machine learning and data augmentation techniques has been developed in order to predict the stiffness modulus of the asphalt concrete layer of an airport runway, from data acquired with a heavy weight deflectometer (HWD). The predictive model relies on a shallow neural network (SNN) trained with the results of a backcalculation, by means of a data augmentation method and can produce estimations of the stiffness modulus even at runway points not yet sampled. The Bayesian regularization algorithm was used for training of the feedforward backpropagation SNN, and a k-fold cross-validation procedure was implemented for a fair performance evaluation. The testing phase result concerning the stiffness modulus prediction was characterized by a coefficient of correlation equal to 0.9864 demonstrating that the proposed neural approach is fully reliable for performance evaluation of airfield pavements or any other paved area. Such a performance prediction model can play a crucial role in airport pavement management systems (APMS), allowing the maintenance budget to be optimized.
Nicola Baldo; Matteo Miani; Fabio Rondinella; Clara Celauro. A Machine Learning Approach to Determine Airport Asphalt Concrete Layer Moduli Using Heavy Weight Deflectometer Data. Sustainability 2021, 13, 8831 .
AMA StyleNicola Baldo, Matteo Miani, Fabio Rondinella, Clara Celauro. A Machine Learning Approach to Determine Airport Asphalt Concrete Layer Moduli Using Heavy Weight Deflectometer Data. Sustainability. 2021; 13 (16):8831.
Chicago/Turabian StyleNicola Baldo; Matteo Miani; Fabio Rondinella; Clara Celauro. 2021. "A Machine Learning Approach to Determine Airport Asphalt Concrete Layer Moduli Using Heavy Weight Deflectometer Data." Sustainability 13, no. 16: 8831.
Drivers are prone to distractions while driving, due to conversations they have with passengers on board, processing their thoughts or using their mobile phones. These distractions result in a mental workload that compromises driving safety and requires the implementation of risk compensatory behaviours. This study examines the effects of hands-free mobile phone conversations on young drivers' stopping manoeuvres when a pedestrian enters a zebra crossing. A cohort of seventy-eight university students, aged 20-30 years old, performed a driving task in a virtual urban environment, by means of a virtual car driving simulator. They formed a control and an experimental group, balanced on age and IQ level. The control group was left free to drive without any imposed cognitive task. The experimental group was asked to drive while making a phone call that was planned to diminish the amount of cognitive resources allocated to the driving experience. For both groups, the analyses focused on a specific moment, i.e., while a child suddenly entered a zebra crossing from a sidewalk. Throughout the simulation, the intensity of the participants' actions on the brake pedal, accelerator, and steering wheel were recorded with a time step of 250 ms. Before the virtual driving experiment, each participant completed a questionnaire on his/her daily driving style, involvement in road accidents, and general mobile phone usage even while driving. A mixed two-way ANOVA with Group as a between-subject factor (1. Control Group; 2. Experimental Group) and Gender (1. Male drivers; 2. Female drivers) as a within-subject factor was performed on the driving parameters as dependent variables. The results showed the presence of a significant difference for distracted and non-distracted drivers with the absence of gender-related differences across the two groups. Participants engaged in a hands-free phone-call while driving assumed lower initial speeds as an element of risk compensation and took the first action to stop at shorter distances from the pedestrian crossing. This suggests a delayed perception of the presence of the pedestrian. In addition, the fluctuation in speed after the distracted driver had released the accelerator pedal reached a statistical significance compared to the control group. These findings suggest that the distraction induced by the use of the mobile phone through the earphones may adversely affect driving behaviour and raise significant safety concerns.
Nicola Baldo; Andrea Marini; Matteo Miani. Effects of Cognitive Distraction on Driver’s Stopping Behaviour: A Virtual Car Driving Simulator Study. IOP Conference Series: Materials Science and Engineering 2020, 960, 022082 .
AMA StyleNicola Baldo, Andrea Marini, Matteo Miani. Effects of Cognitive Distraction on Driver’s Stopping Behaviour: A Virtual Car Driving Simulator Study. IOP Conference Series: Materials Science and Engineering. 2020; 960 (2):022082.
Chicago/Turabian StyleNicola Baldo; Andrea Marini; Matteo Miani. 2020. "Effects of Cognitive Distraction on Driver’s Stopping Behaviour: A Virtual Car Driving Simulator Study." IOP Conference Series: Materials Science and Engineering 960, no. 2: 022082.
Knowing the relationship between the stiffness modulus and the empirical mechanical characteristics of asphalt concrete, road engineers may predict the expected results of costly laboratory tests and save both time and financial resources in the mix design phase. In fact, such a model would make it possible to assess a priori whether the stiffness of a specific mixture, characterised in the laboratory only by the common Marshall test, is suitable for the level of service required by the road pavement under analysis. In this study, 54 Marshall test specimens of high modulus asphalt concrete were prepared and tested in the laboratory to determine an empirical relationship between the stiffness modulus and Marshall stability by means of shallow artificial neural networks. Part out of these mixtures was characterised by different types of bitumen (20/30 or 50/70 penetration grade) and percentages of used reclaimed asphalt (RAP at 20% or 30%); a polymer modified bitumen was used in the preparation of the remaining Marshall test specimens, which do not contain RAP. For the complex and laborious identification of the neural model hyperparameters, which define its architecture and algorithmic functioning, the Bayesian optimization approach has been adopted. Although the results of this methodology depend on the predefined hyperparameters variability ranges, it allows an unbiased definition of the optimal neural model characteristics to be performed by minimizing (or maximizing) a loss function. In this study, the mean square error on 5 validation folds was used as a loss function, in order to avoid a poor performance evaluation due to the small number of samples. In addition, 3 different neural training algorithms were applied to compare results and convergence times. The procedure presented in this study is a valuable guide for the development of predictive models of asphalt concretes' behaviour, even for different types of bitumen and aggregates considered here.
Nicola Baldo; Jan Valentin; Evangelos Manthos; Matteo Miani. Numerical Characterization of High Modulus Asphalt Concrete Containing RAP: A Comparison among Optimized Shallow Neural Models. IOP Conference Series: Materials Science and Engineering 2020, 960, 022083 .
AMA StyleNicola Baldo, Jan Valentin, Evangelos Manthos, Matteo Miani. Numerical Characterization of High Modulus Asphalt Concrete Containing RAP: A Comparison among Optimized Shallow Neural Models. IOP Conference Series: Materials Science and Engineering. 2020; 960 (2):022083.
Chicago/Turabian StyleNicola Baldo; Jan Valentin; Evangelos Manthos; Matteo Miani. 2020. "Numerical Characterization of High Modulus Asphalt Concrete Containing RAP: A Comparison among Optimized Shallow Neural Models." IOP Conference Series: Materials Science and Engineering 960, no. 2: 022083.
In this study, a cohort of 78 university students performed a driving experience in a virtual urban scenario, by means of a car driving simulator, to examine effects of a planned hands-free mobile phone conversation on young drivers’ braking behaviors. To this aim, a control group was left free to drive without any imposed cognitive task. An experimental group faced the same scenario while engaged in a phone call. The conversation via earphones was arranged to diminish the amount of cognitive resources allocated to the driving task. For both groups, the analyses focused on the moment at which a child entered a pedestrian crossing from a sidewalk. The results of a mixed two-way ANOVA showed the presence of a significant difference for distracted and non-distracted drivers with the absence of gender-related differences across the two groups. Distracted participants assumed lower initial speeds, took the first action to stop at shorter distances from the zebra crossing, and had more difficulty in keeping speed variations under control. These findings suggest that the distraction induced by the use of earphones may induce risk compensation behaviors and delay pedestrian perception. Moreover, the effects on the participants' braking behavior suggest that the procedure adopted to increase cognitive load, based on a story retelling, is an effective method to analyze the impact of hands-free cellphone use on driving skills in a car simulation experiment.
Nicola Baldo; Andrea Marini; Matteo Miani. Drivers’ Braking Behavior Affected by Cognitive Distractions: An Experimental Investigation with a Virtual Car Simulator. Behavioral Sciences 2020, 10, 150 .
AMA StyleNicola Baldo, Andrea Marini, Matteo Miani. Drivers’ Braking Behavior Affected by Cognitive Distractions: An Experimental Investigation with a Virtual Car Simulator. Behavioral Sciences. 2020; 10 (10):150.
Chicago/Turabian StyleNicola Baldo; Andrea Marini; Matteo Miani. 2020. "Drivers’ Braking Behavior Affected by Cognitive Distractions: An Experimental Investigation with a Virtual Car Simulator." Behavioral Sciences 10, no. 10: 150.
The present paper discusses the analysis and modeling of laboratory data regarding the mechanical characterization of hot mix asphalt (HMA) mixtures for road pavements, by means of artificial neural networks (ANNs). The HMAs investigated were produced using aggregate and bitumen of different types. Stiffness modulus (ITSM) and Marshall stability (MS) and quotient (MQ) were assumed as mechanical parameters to analyze and predict. The ANN modeling approach was characterized by multiple layers, the k-fold cross validation (CV) method, and the positive linear transfer function. The effectiveness of such an approach was verified in terms of the coefficients of correlation (R) and mean square errors; in particular, R values were within the range 0.965–0.919 in the training phase and 0.881–0.834 in the CV testing phase, depending on the predicted parameters.
Nicola Baldo; Evangelos Manthos; Matteo Miani. Stiffness Modulus and Marshall Parameters of Hot Mix Asphalts: Laboratory Data Modeling by Artificial Neural Networks Characterized by Cross-Validation. Applied Sciences 2019, 9, 3502 .
AMA StyleNicola Baldo, Evangelos Manthos, Matteo Miani. Stiffness Modulus and Marshall Parameters of Hot Mix Asphalts: Laboratory Data Modeling by Artificial Neural Networks Characterized by Cross-Validation. Applied Sciences. 2019; 9 (17):3502.
Chicago/Turabian StyleNicola Baldo; Evangelos Manthos; Matteo Miani. 2019. "Stiffness Modulus and Marshall Parameters of Hot Mix Asphalts: Laboratory Data Modeling by Artificial Neural Networks Characterized by Cross-Validation." Applied Sciences 9, no. 17: 3502.