Model Compression Techniques for Scalable Deployment of Generative AI in Automotive Systems

The deployment of generative AI models within automotive systems is fundamentally constrained by the limited computational resources, memory capacity, and energy budgets of in-vehicle hardware. In a companion article \cite{part1}, we examined the scalability challenges, current applications, and evolving E/E architectures that frame this problem. The present paper builds on that foundation by providing an in-depth survey of model compression and optimization techniques that enable the deployment of large neural networks on resource-constrained automotive platforms. We first present an overview of deployment strategies including edge computing, hardware selection, optimized inference frameworks, AI compilers, dedicated accelerators, and knowledge distillation. We then conduct a detailed review of three core compression methodologies: pruning (both with and without retraining), quantization (post-training and quantization-aware approaches for both vision and language models), and low-rank tensor decomposition (including Canonical Polyadic, Tucker, Tensor Train, and Tensor Ring methods, as well as Neural Architecture Search-based compression). For each technique, we analyze the underlying principles, review state-of-the-art methods, and discuss their applicability to automotive deployment scenarios. A comparative analysis highlights the trade-offs among compression ratio, inference speedup, accuracy retention, and hardware compatibility. Our findings indicate that combining multiple compression strategies offers the most promising path toward practical, real-time generative AI deployment in vehicles, and we identify open research directions for the automotive AI community.