This thesis presents estimation algorithms for bicycles aimed to define the road slope and the road irregularity level using the information coming from inertial measures. The slope estimation is realised by means of a Kalman filter based on the longitudinal kinematic model and on the information coming from the pitch rate. The most signficant weak points of this approach have been removed by means of suitable correction algorithms. The proposed solution guarantees both high dynamic performance (reaction to fast slope variations in less than 1 s) and good accuracy (the slope estimation error is around 0.5 deg). The road irregularity level is defined according to an index dependent on the power of the vertical acceleration. The latter, weighted with respect to the bicycle speed, allows to classify the road confugurations in three differet categories.
In questa tesi vengono presentati algoritmi di stima per biciclette volti ad individuare la pendenza e l’irregolarità del fondo stradale sfruttando le informazioni provenienti dai segnali inerziali misurati da accelerometri e giroscopi. La stima di pendenza viene realizzata da un filtro di Kalman basato sul modello cinematico longitudinale e sull’informazione proveniente dal pitch rate. Le principali carenze di questo metodo sono state limitate attraverso opportuni sistemi di correzione. La soluzione proposta garantisce elevate prestazioni dinamiche (risposta a rapide variazioni di pendenza in meno di 1 s) e buona precisione (incertezza di 0.5 deg). Il riconoscimento dei livelli di irregolarità stradale viene effettuato tramite un indice basato sulla potenza dell’accelerazione verticale che, pesato rispetto alla velocità, permette di classificare le varie configurazioni stradali in tre categorie differenti.
Design of inertial estimation algorithms for bicycles
SPEZIALI, LUCA
2015/2016
Abstract
This thesis presents estimation algorithms for bicycles aimed to define the road slope and the road irregularity level using the information coming from inertial measures. The slope estimation is realised by means of a Kalman filter based on the longitudinal kinematic model and on the information coming from the pitch rate. The most signficant weak points of this approach have been removed by means of suitable correction algorithms. The proposed solution guarantees both high dynamic performance (reaction to fast slope variations in less than 1 s) and good accuracy (the slope estimation error is around 0.5 deg). The road irregularity level is defined according to an index dependent on the power of the vertical acceleration. The latter, weighted with respect to the bicycle speed, allows to classify the road confugurations in three differet categories.File | Dimensione | Formato | |
---|---|---|---|
2106_12_Speziali.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
15.32 MB
Formato
Adobe PDF
|
15.32 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/10589/131772