Data-driven edge intelligence in electric mobility: from battery health forecasting to high-speed autonomous control

Chou, Ka Seng (2026) Data-driven edge intelligence in electric mobility: from battery health forecasting to high-speed autonomous control, [Dissertation thesis], Alma Mater Studiorum Università di Bologna. Dottorato di ricerca in Automotive engineering for intelligent mobility, 39 Ciclo.
Documenti full-text disponibili:
[thumbnail of P1204600-Ka Seng Chou-Final-Thesis-Unibo.pdf] Documento PDF (English) - Richiede un lettore di PDF come Xpdf o Adobe Acrobat Reader
Disponibile con Licenza: Salvo eventuali più ampie autorizzazioni dell'autore, la tesi può essere liberamente consultata e può essere effettuato il salvataggio e la stampa di una copia per fini strettamente personali di studio, di ricerca e di insegnamento, con espresso divieto di qualunque utilizzo direttamente o indirettamente commerciale. Ogni altro diritto sul materiale è riservato.
Download (92MB)

Abstract

Electric mobility is changing how people and goods move. Both electric cars and autonomous drones face a shared limit: their batteries. Lab tests cannot fully capture how these limits appear in real-world use. This thesis addresses that gap using data collected directly from the vehicles. Human driving behaviors were studied in passenger cars, with over 2.3 million data points collected from a Nissan Leaf using a custom data-logging device. The results show that rapid acceleration and other demanding behaviors raise battery temperature and accelerate cell aging. A Bidirectional Long Short-Term Memory model was developed to predict battery temperature, capturing the dynamic effects of real-world driving. These findings were then applied to autonomous drone racing. A vision-only system enables drones to fly at high speed without Global Positioning System localization, and a Reinforcement Learning controller achieves performance beyond human limits. However, such "bang-bang" aggressive control rapidly degrades batteries, with most cells failing in fewer than 20 flights. Building on these insights, the thesis introduces "Energy-Aware Autonomy". A framework that integrates battery prediction with control systems to maintain high performance while reducing the risk of battery damage.

Abstract
Tipologia del documento
Tesi di dottorato
Autore
Chou, Ka Seng
Supervisore
Dottorato di ricerca
Ciclo
39
Coordinatore
Settore disciplinare
Settore concorsuale
Parole chiave
Autonomous Vehicles, Electric Vehicles, Drone Racing, Battery Health Monitoring, Reinforcement Learning, Deep Learning, Edge Intelligence, Robotics
Data di discussione
23 Aprile 2026
URI

Altri metadati

Statistica sui download

Gestione del documento: Visualizza la tesi

^