The research activity illustrated in the present work is the result of the cooperation with Department of Mechanical Engineering and STMicroelectronics team of Validation & Characterization. It covers also elements of psychology in data science necessary for recognizing relevant patterns related to driving behavior. The primary aspect enlightened was the description of the experimental setup used to perform tests with 26 users. Through Unity and Matlab software, it was possible to exploit a virtual reality scenario aimed at recreating in the same route both urban and highway paths in condition of real traffic, performed with a Battery Electric Vehicle. Then, the interface constituted by accelerator, brake pedal and steering wheel was modified to record significant data coming from MEMS sensors, using SolidWorks and SensorTile.box. The acquisition part was illustrated in its methodology in two different cases covering disturbance and non-disturbance performance. Moreover, the population was divided into different categories to establish a characterization which linked energy consumption and associated analysis to psychological traits of the driver and data coming from sensors. The theorical backbone presented in the first chapters shows significant elements necessary to build the overall setup and postprocessing process both for the electric vehicle and the sensor technology used, not neglecting the actual electrical mobility scenario in terms of environmental impact.
L'attività di ricerca illustrata nel presente lavoro è il risultato della collaborazione con il dipartimento di Ingegneria Meccanica e il team di validazione e caratterizzazione di STMicroelectronics. Esso comprende inoltre elementi di psicologia per la scienza dei dati necessari per riconoscere gli aspetti rilevanti relativi al comportamento di guida. Particolare attenzione è stata posta per la costruzione del setup sperimentale, utilizzato per eseguire test con 26 guidatori. Attraverso i software Unity e Matlab, è stato possibile configurare uno scenario di realtà virtuale volto a ricreare nello stesso percorso un tratto urbano e autostradale in condizioni di traffico reale mediante un veicolo elettrico a batteria. In seguito è stata modificata l'interfaccia costituita da acceleratore, pedale del freno e volante per registrare dati significativi provenienti dai sensori MEMS, utilizzando SolidWorks e SensorTile.box. La parte di acquisizione è stata illustrata nella sua metodologia in due diversi casi, riguardanti condizioni di disturbo e non disturbo. Infine, la popolazione è stata divisa in diverse categorie per stabilire una caratterizzazione che collegasse il consumo di energia e l'analisi associata ai tratti psicologici del guidatore con i dati provenienti dai sensori. Il supporto teorico presentato nei primi capitoli illustra elementi dirimenti necessari per la costruzione del setup e per la elaborazione dati sia considerando il veicolo elettrico che la tecnologia utilizzata per i sensori, senza trascurare l'attuale scenario di mobilità elettrica in termini di impatto ambientale.
Analysis of driving behaviors in an electric vehicle using virtual reality and MEMS sensors
DI ANTONIO, JACOPO ANDREA;LAMANUZZI, MARIKA
2018/2019
Abstract
The research activity illustrated in the present work is the result of the cooperation with Department of Mechanical Engineering and STMicroelectronics team of Validation & Characterization. It covers also elements of psychology in data science necessary for recognizing relevant patterns related to driving behavior. The primary aspect enlightened was the description of the experimental setup used to perform tests with 26 users. Through Unity and Matlab software, it was possible to exploit a virtual reality scenario aimed at recreating in the same route both urban and highway paths in condition of real traffic, performed with a Battery Electric Vehicle. Then, the interface constituted by accelerator, brake pedal and steering wheel was modified to record significant data coming from MEMS sensors, using SolidWorks and SensorTile.box. The acquisition part was illustrated in its methodology in two different cases covering disturbance and non-disturbance performance. Moreover, the population was divided into different categories to establish a characterization which linked energy consumption and associated analysis to psychological traits of the driver and data coming from sensors. The theorical backbone presented in the first chapters shows significant elements necessary to build the overall setup and postprocessing process both for the electric vehicle and the sensor technology used, not neglecting the actual electrical mobility scenario in terms of environmental impact.| File | Dimensione | Formato | |
|---|---|---|---|
|
Analysis of driving behaviors in an electric vehicle using virtual reality and MEMS sensors.pdf
non accessibile
Descrizione: Thesis text 7/4
Dimensione
4.04 MB
Formato
Adobe PDF
|
4.04 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/10589/153091