According to ISTAT statistics, in 2015 the number of seriously injured people and fatalities in urban centres due to car accidents has increased respectively of 6% and 8,6% compared to the year 2014, in countertrend with the last years. There is a great number of crash dynamics but, statistically, human behaviour while driving is responsible for the most part of the accidents (it is estimated to be 92,75% of the total amount) so this is a factor that must be taken into account; the present work concerns a specific component of the driver's emotional state, that is the stress while driving. In particular, thanks to the employment of wearable technologies that measure driver's physiological parameters and through their comparison, the aim is to define a method useful to identify stress spots, to validate the same method through an experimental procedure and to recognize possible correlations between these stressing spots and the analysis of the crash data of specific roads. After the method has been described in details and the experimental procedure has been designed considering a urban itinerary in Milan, spots with a greater level of stress were attempted to be located through the measurement of some of the cardiac parameters, the respiratory rate, the galvanic skin response, a GPS and through a camera during the tests. In this way, areas with same stress problems for both of the participants involved in the experiments were determined and the spots marked through the devices were compared with spots characterized by the presence of several road accidents, depending on the available crash data. From data processing, it has been discovered that 50% of the stressing spots marked with the present method match with crash data showing a sort of correlation. This fact enables to obtain indications about critical zones of the road network when accidents have yet to occur, to sustain the method here presented as a valid support for prevention analysis of crash data and to achieve information that could complete these same crash data, if available.
Nel 2015, secondo le statistiche ISTAT, il numero di feriti gravi e di morti nei centri abitati a causa di incidenti stradali è salito rispettivamente del 6% e dell’8,6% rispetto al 2014, in controtendenza con gli anni precedenti. Cause e dinamiche degli incidenti sono molteplici, ma statisticamente il comportamento del conducente alla guida risulta causa, o comunque concausa, della quasi totalità degli incidenti (si stima corrisponda al 92,75% dei casi) ed è pertanto un elemento da analizzare. Il presente lavoro investiga una componente specifica dello stato del conducente di un veicolo, costituito dallo stress alla guida. In particolare, ricorrendo a dispositivi indossabili di misurazione dei parametri fisiologici del conducente ed al loro confronto, ci si propone qui di definire un metodo utile per l'individuazione di punti stressanti per l'utente alla guida, di validarlo attraverso un protocollo sperimentale e di identificare possibili correlazioni tra i punti di stress determinati e l'analisi di incidentalità condotta su alcune tratte stradali. Dopo aver esposto in dettaglio il metodo definito e aver progettato una campagna di misura su un percorso urbano nella città di Milano, si è cercato di localizzare i punti di maggiore stress attraverso la misura di alcuni parametri cardiaci, della frequenza respiratoria, della conduttanza cutanea, del GPS e di una videocamera durante la sperimentazione. Si sono così potute determinare zone con problematiche di stress comuni agli utenti sottoposti alle prove e confrontare la localizzazione dei punti rilevati grazie alla strumentazione con quella di punti caratterizzati da numerosi eventi incidentali, sulla base dei dati storici disponibili. L'elaborazione dei dati ha evidenziato come il 50% dei punti in cui si registrano condizioni di stress secondo il metodo proposto corrisponda anche a punti incidentali, evidenziando una possibile correlazione. Questo fatto permetterebbe di ottenere indicazioni sui punti critici della rete stradale prima ancora del verificarsi di un evento incidentale, configurando il metodo proposto come un valido supporto all'analisi di incidentalità di tipo preventivo e all’ottenimento di informazioni che possano integrare i dati storici incidentali, qualora disponibili.
Indagine dello stress alla guida attraverso parametri fisiologici come supporto all'analisi di sicurezza stradale
GALLO, FEDERICA;PAGLINO, VALERIA
2015/2016
Abstract
According to ISTAT statistics, in 2015 the number of seriously injured people and fatalities in urban centres due to car accidents has increased respectively of 6% and 8,6% compared to the year 2014, in countertrend with the last years. There is a great number of crash dynamics but, statistically, human behaviour while driving is responsible for the most part of the accidents (it is estimated to be 92,75% of the total amount) so this is a factor that must be taken into account; the present work concerns a specific component of the driver's emotional state, that is the stress while driving. In particular, thanks to the employment of wearable technologies that measure driver's physiological parameters and through their comparison, the aim is to define a method useful to identify stress spots, to validate the same method through an experimental procedure and to recognize possible correlations between these stressing spots and the analysis of the crash data of specific roads. After the method has been described in details and the experimental procedure has been designed considering a urban itinerary in Milan, spots with a greater level of stress were attempted to be located through the measurement of some of the cardiac parameters, the respiratory rate, the galvanic skin response, a GPS and through a camera during the tests. In this way, areas with same stress problems for both of the participants involved in the experiments were determined and the spots marked through the devices were compared with spots characterized by the presence of several road accidents, depending on the available crash data. From data processing, it has been discovered that 50% of the stressing spots marked with the present method match with crash data showing a sort of correlation. This fact enables to obtain indications about critical zones of the road network when accidents have yet to occur, to sustain the method here presented as a valid support for prevention analysis of crash data and to achieve information that could complete these same crash data, if available.File | Dimensione | Formato | |
---|---|---|---|
2016_12_Gallo_Paglino.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
11.44 MB
Formato
Adobe PDF
|
11.44 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/131463