This technical session offers a deep dive into the deployment and optimization of AI models powering autonomous driving systems. It walks through some industry-standard workflows, from model development to real-time deployment in embedded automotive platforms. The presentation will cover critical topics such as inference acceleration, quantization, pruning and runtime optimization on autonomous driving hardware. Attendees will gain insights through practical examples, toolchains and performance benchmarks, highlighting best practices and common pitfalls in deploying AI on autonomous vehicles.

• Model deployment
• Model optimization
• Inference acceleration