ArXiv TLDR

Agentic Driving Coach: Robustness and Determinism of Agentic AI-Powered Human-in-the-Loop Cyber-Physical Systems

🐦 Tweet
2604.11705

Deeksha Prahlad, Daniel Fan, Hokeun Kim

cs.AIcs.CLcs.ROeess.SY

TLDR

It introduces a reactor-MoC approach with Lingua Franca to restore determinism in agentic AI-powered human-in-the-loop cyber-physical systems.

Key contributions

  • Addresses unpredictable behavior in LLM-powered human-in-the-loop cyber-physical systems.
  • Proposes a reactor-model-of-computation (MoC) approach using the open-source Lingua Franca framework.
  • Demonstrates the approach with an agentic driving coach as a human-in-the-loop cyber-physical system.
  • Identifies practical challenges and pathways to reintroduce determinism into agentic HITL CPS.

Why it matters

Foundation models in human-in-the-loop cyber-physical systems face nondeterminism. This paper offers a novel reactor-MoC approach with Lingua Franca to address this, making these systems more robust and predictable. It's crucial for reliable AI integration in critical cyber-physical applications.

Original Abstract

Foundation models, including large language models (LLMs), are increasingly used for human-in-the-loop (HITL) cyber-physical systems (CPS) because foundation model-based AI agents can potentially interact with both the physical environments and human users. However, the unpredictable behavior of human users and AI agents, in addition to the dynamically changing physical environments, leads to uncontrollable nondeterminism. To address this urgent challenge of enabling agentic AI-powered HITL CPS, we propose a reactor-model-of-computation (MoC)-based approach, realized by the open-source Lingua Franca (LF) framework. We also carry out a concrete case study using the agentic driving coach as an application of HITL CPS. By evaluating the LF-based agentic HITL CPS, we identify practical challenges in reintroducing determinism into such agentic HITL CPS and present pathways to address them.

📬 Weekly AI Paper Digest

Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.