Semi-active suspension control is the most widely adopted technology in electronically controlled suspension systems. By regulating the damping characteristics, such systems aim to achieve an optimal balance between ride comfort and road-holding performance. To further enhance the performance of these systems, it is crucial to incorporate information about future road conditions. While the concept of preview-based suspension control has been explored in the literature, relatively few studies have addressed the use of preview information for big obstacles, such as speed bumps. Moreover, the impact of perception errors on the performance of such systems has received limited attention. This thesis addresses these gaps by first presenting an algorithm capable of estimating the profile of a bump ahead of a vehicle using only images from a monocular camera and the vehicle’s longitudinal velocity. The accuracy of the proposed estimation method is assessed using ground truth data obtained from a high-resolution 3D laser scanner, which captures the road surface ahead of the vehicle. Subsequently, a control algorithm is developed to determine the optimal damping force for a semi-active suspension system using the previewed bump information. The control design and performance evaluation are conducted using a semi-active quarter-car model. The proposed control strategy is benchmarked against the classical SkyHook controller across various road profiles. Simulation results demonstrate that, under ideal bump preview conditions, the proposed algorithm significantly improves ride comfort over bumps. When estimation errors are introduced, the controller’s effectiveness is shown to be influenced by the accuracy of the previewed road profile.
Il controllo semi-attivo della sospensione è la tecnologia più ampiamente adottata nei sistemi di sospensione elettronici. Regolando le caratteristiche di smorzamento, tali sistemi mirano a ottenere un equilibrio ottimale tra comfort di marcia e aderenza al suolo. Per migliorare ulteriormente le prestazioni di questi sistemi, è fondamentale integrare informazioni sulle condizioni stradali future. Sebbene il concetto di controllo della sospensione basato su anteprima sia stato esplorato in letteratura, pochi studi hanno affrontato l'utilizzo delle informazioni di anteprima per ostacoli di grandi dimensioni, come i dossi rallentatori. Inoltre, l’impatto degli errori di percezione sulle prestazioni di tali sistemi è stato finora poco investigato. Questa tesi affronta tali lacune presentando innanzitutto un algoritmo in grado di stimare il profilo di un dosso posto davanti al veicolo utilizzando esclusivamente immagini acquisite da una camera monoculare e la velocità longitudinale del veicolo. L’accuratezza del metodo di stima proposto è valutata utilizzando dati di riferimento (ground truth) acquisiti tramite uno scanner laser 3D ad alta risoluzione, capace di rilevare la superficie stradale davanti al veicolo. Successivamente, viene sviluppato un algoritmo di controllo per determinare la forza di smorzamento ottimale per un sistema di sospensione semi-attivo, sfruttando le informazioni di anteprima del dosso. Il progetto e la valutazione delle prestazioni del controllo sono condotti utilizzando un modello quarter-car semi-attivo. La strategia proposta è confrontata con il classico controllore SkyHook su diversi profili stradali. I risultati delle simulazioni dimostrano che, in condizioni di anteprima ideale del dosso, l’algoritmo proposto migliora significativamente il comfort di marcia. In presenza di errori nella stima, l’efficacia del controllo risulta dipendente dall’accuratezza del profilo stradale stimato.
Semi-active suspension control using mono camera-based bump preview
Lagalante, Antonello
2024/2025
Abstract
Semi-active suspension control is the most widely adopted technology in electronically controlled suspension systems. By regulating the damping characteristics, such systems aim to achieve an optimal balance between ride comfort and road-holding performance. To further enhance the performance of these systems, it is crucial to incorporate information about future road conditions. While the concept of preview-based suspension control has been explored in the literature, relatively few studies have addressed the use of preview information for big obstacles, such as speed bumps. Moreover, the impact of perception errors on the performance of such systems has received limited attention. This thesis addresses these gaps by first presenting an algorithm capable of estimating the profile of a bump ahead of a vehicle using only images from a monocular camera and the vehicle’s longitudinal velocity. The accuracy of the proposed estimation method is assessed using ground truth data obtained from a high-resolution 3D laser scanner, which captures the road surface ahead of the vehicle. Subsequently, a control algorithm is developed to determine the optimal damping force for a semi-active suspension system using the previewed bump information. The control design and performance evaluation are conducted using a semi-active quarter-car model. The proposed control strategy is benchmarked against the classical SkyHook controller across various road profiles. Simulation results demonstrate that, under ideal bump preview conditions, the proposed algorithm significantly improves ride comfort over bumps. When estimation errors are introduced, the controller’s effectiveness is shown to be influenced by the accuracy of the previewed road profile.File | Dimensione | Formato | |
---|---|---|---|
2025_07_Lagalante_Tesi_01.pdf
non accessibile
Descrizione: Tesi
Dimensione
29.84 MB
Formato
Adobe PDF
|
29.84 MB | Adobe PDF | Visualizza/Apri |
2025_07_Lagalante_Executive_Summary_02.pdf
non accessibile
Descrizione: Executive Summary
Dimensione
1.79 MB
Formato
Adobe PDF
|
1.79 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/10589/240199