This thesis investigates the design, control, and testing of an active anti-roll bar (AARB) system for vehicle lateral stability enhancement, with a focus on control tuning using genetic algorithms and driver-in-the-loop (DiL) testing. Conventional passive anti-roll bar systems provide fixed roll stiffness, which often forces compromises between vehicle handling, ride comfort, and safety. In contrast, active anti-roll bar systems dynamically adjust the distribution of roll stiffness between front and rear axles, enabling improved stability, responsiveness, and ultimately enhancing vehicle safety. The primary goal of this research is to optimize the control logic of an AARB system to enhance lateral vehicle dynamics during high-speed cornering and emergency maneuvers, thereby improving overall safety. A genetic algorithm is employed for tuning the control gains. A detailed model of the AARB was developed in Matlab/Simulink and integrated with a 14-degree-of-freedom vehicle model using Car Real-Time (CRT) simulation environment to simulate various driving scenarios. The performance of the AARB system is assessed using both offline simulations and driver-in-the-loop tests. The results demonstrate that the proposed AARB system significantly improves lateral stability by dynamically adapting the roll stiffness distribution between front and rear axles, reducing understeer or oversteer tendencies, and enhancing the driver's ability to maintain control during critical situations. This adaptability translates directly to enhanced safety, particularly in high-speed maneuvers where maintaining vehicle stability is crucial. By providing a balanced solution between comfort, performance, and safety, as dictated by the optimization criteria of the genetic algorithm, this approach makes AARB systems particularly suitable for applications in vehicles where high stability and responsiveness are prioritized. The findings suggest that integrating genetic algorithms for control tuning and validating performance with DiL testing can yield an effective framework for developing advanced vehicle dynamics control systems with a strong emphasis on safety.
Questa tesi affronta la progettazione, il controllo e la sperimentazione di una barra antirollio attiva (AARB) per il miglioramento della stabilità laterale del veicolo, con particolare attenzione alla messa a punto del controllo tramite algoritmi genetici e simulazioni con il conducente al volante (DiL). I sistemi di barre antirollio tradizionali offrono una rigidezza al rollio fissa, comportando spesso compromessi tra maneggevolezza, comfort e sicurezza. I sistemi AARB, invece, regolano dinamicamente la distribuzione della rigidezza al rollio tra gli assi anteriore e posteriore, migliorando la stabilità, la reattività e aumentando la sicurezza complessiva del veicolo. L'obiettivo principale di questa ricerca è ottimizzare la logica di controllo di un sistema AARB per migliorare la dinamica laterale del veicolo durante curve ad alta velocità e manovre di emergenza, garantendo una maggiore sicurezza. Per la regolazione dei guadagni di controllo è stato impiegato un algoritmo genetico. È stato sviluppato un modello dettagliato della AARB in Matlab/Simulink e integrato con un modello veicolare a 14 gradi di libertà di Car Real-Time (CRT), al fine di simulare diverse condizioni di guida. Le prestazioni del sistema sono state valutate sia attraverso simulazioni offline che test con il conducente. I risultati dimostrano che il sistema AARB proposto migliora notevolmente la stabilità laterale adattando dinamicamente la distribuzione della rigidezza al rollio tra gli assi, riducendo la tendenza al sottosterzo o al sovrasterzo e migliorando la capacità del conducente di mantenere il controllo in situazioni critiche. Questa adattabilità si traduce in una sicurezza superiore, soprattutto durante manovre ad alta velocità in cui la stabilità è fondamentale. Offrendo un equilibrio ottimale tra comfort, prestazioni e sicurezza, come definito dai criteri di ottimizzazione dell'algoritmo genetico, questo approccio rende i sistemi AARB ideali per veicoli in cui la stabilità e la reattività sono prioritarie.
Active anti-roll bar for vehicle lateral stability: genetic algorithm for control tuning and driver in the loop testing
Verde, Raffaele;Pirchio, Francesco
2023/2024
Abstract
This thesis investigates the design, control, and testing of an active anti-roll bar (AARB) system for vehicle lateral stability enhancement, with a focus on control tuning using genetic algorithms and driver-in-the-loop (DiL) testing. Conventional passive anti-roll bar systems provide fixed roll stiffness, which often forces compromises between vehicle handling, ride comfort, and safety. In contrast, active anti-roll bar systems dynamically adjust the distribution of roll stiffness between front and rear axles, enabling improved stability, responsiveness, and ultimately enhancing vehicle safety. The primary goal of this research is to optimize the control logic of an AARB system to enhance lateral vehicle dynamics during high-speed cornering and emergency maneuvers, thereby improving overall safety. A genetic algorithm is employed for tuning the control gains. A detailed model of the AARB was developed in Matlab/Simulink and integrated with a 14-degree-of-freedom vehicle model using Car Real-Time (CRT) simulation environment to simulate various driving scenarios. The performance of the AARB system is assessed using both offline simulations and driver-in-the-loop tests. The results demonstrate that the proposed AARB system significantly improves lateral stability by dynamically adapting the roll stiffness distribution between front and rear axles, reducing understeer or oversteer tendencies, and enhancing the driver's ability to maintain control during critical situations. This adaptability translates directly to enhanced safety, particularly in high-speed maneuvers where maintaining vehicle stability is crucial. By providing a balanced solution between comfort, performance, and safety, as dictated by the optimization criteria of the genetic algorithm, this approach makes AARB systems particularly suitable for applications in vehicles where high stability and responsiveness are prioritized. The findings suggest that integrating genetic algorithms for control tuning and validating performance with DiL testing can yield an effective framework for developing advanced vehicle dynamics control systems with a strong emphasis on safety.File | Dimensione | Formato | |
---|---|---|---|
Executive_Summary_Pirchio_Verde.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Executive Summary
Dimensione
4.68 MB
Formato
Adobe PDF
|
4.68 MB | Adobe PDF | Visualizza/Apri |
Tesi_Pirchio_Verde.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Tesi
Dimensione
12.83 MB
Formato
Adobe PDF
|
12.83 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/10589/231305