Software-Defined Vehicles: Architecting the Future of Intelligent and Connected Mobility

<div class="section abstract"><div class="htmlview paragraph">This paper presents a comprehensive technical review of the Software-Defined Vehicle (SDV), a paradigm that is fundamentally reshaping the automotive industry. We analyze the architectural evolution from distributed Electronic Control Units (ECUs) to centralized zonal compute platforms, examining the critical role of Service-Oriented Architectures (SOA), the AUTOSAR standard, and virtualization technologies in enabling this shift. A comparative analysis of leading High-Performance Computing (HPC) platforms, including NVIDIA DRIVE, Tesla FSD, and Qualcomm Snapdragon Ride, is conducted to evaluate the silicon foundation of the SDV. The paper further investigates key enabling technologies such as Over- the-Air (OTA) updates, Digital Twins, and the integration of Artificial Intelligence (AI) for applications ranging from predictive maintenance to software-defined battery management. We scrutinize the competing V2X communication standards (DSRC vs. C-V2X) and address the paramount challenges of functional safety (ISO 26262) and cybersecurity (ISO/SAE 21434) in this new landscape. Finally, we identify open research challenges, ethical considerations, and the future trajectory of SDVs toward fully AI-defined, intelligent, and connected mobility.</div></div>