While MPC effectively handles structured, diverse, and low-level specifications, it lacks the capability to dynamically incorporate high-level contextual information such as social norms, user intent, or natural language instructions. To address this limitation, this manuscript introduces an agentic MPC framework that enables context-aware, semantically adaptive control synthesis by integrating with large language model-based agents. The agent interprets heterogeneous inputs, including natural language messages, environmental observations, and external knowledge, to resynthesize the control specifications. The effectiveness of the framework is demonstrated in an autonomous driving scenario, where the system aligns with personal preferences or responds to social situations such as emergency vehicle yielding.

翻译：尽管模型预测控制（MPC）能有效处理结构化、多样化的底层次规范，但其缺乏动态整合社会规范、用户意图或自然语言指令等高层次上下文信息的能力。为弥补这一局限，本文提出一种智能体MPC框架，通过集成基于大语言模型的智能体，实现上下文感知的语义自适应控制综合。该智能体能够解析异构输入——包括自然语言消息、环境观测及外部知识——以重新综合控制规范。本文通过自动驾驶场景验证了该框架的有效性，该系统既能适配个人偏好，也能应对诸如为紧急车辆让行等社会情境。