2025 marks a turning point for autonomous trucking, moving from years of R&D into real-world commercial deployment. This talk introduces Plus’s three-layer architecture powering this shift – vision language models (VLMs), end-to-end models and safety guardrails. Learn how VLMs help with meta decisions and handle 'long tail' problems, end-to-end models enable scalable deployment across diverse regions and vehicle platforms, and safety guardrails provide sanity checks to prevent radical maneuvers – together forming a robust, scalable and interpretable foundation for safe and generalizable autonomous driving.

• Incorporating driving-specific VLMs to decode complex human interactions and signals or interpret variable messages on electronic signs
• Using highly optimized end-to-end models that take sensor inputs and provide the planned vehicle trajectory
• Ensuring safe driving through safety guardrails which are algorithmic constraints acting as a watchdog over model-based driving