We explore the usage of large language models (LLM) in human-in-the-loop human-in-the-plant cyber-physical systems (CPS) to translate a high-level prompt into a personalized plan of actions, and subsequently convert that plan into a grounded inference of sequential decision-making automated by a real-world CPS controller to achieve a control goal. We show that it is relatively straightforward to contextualize an LLM so it can generate domain-specific plans. However, these plans may be infeasible for the physical system to execute or the plan may be unsafe for human users. To address this, we propose CPS-LLM, an LLM retrained using an instruction tuning framework, which ensures that generated plans not only align with the physical system dynamics of the CPS but are also safe for human users. The CPS-LLM consists of two innovative components: a) a liquid time constant neural network-based physical dynamics coefficient estimator that can derive coefficients of dynamical models with some unmeasured state variables; b) the model coefficients are then used to train an LLM with prompts embodied with traces from the dynamical system and the corresponding model coefficients. We show that when the CPS-LLM is integrated with a contextualized chatbot such as BARD it can generate feasible and safe plans to manage external events such as meals for automated insulin delivery systems used by Type 1 Diabetes subjects.
DURA-CPS: A Multi-Role Orchestrator for Dependability Assurance in LLM-Enabled Cyber-Physical Systems
Cyrion Labs(Cyrion实验室);Carnegie Mellon University(卡内基梅隆大学);Cornell University(康奈尔大学);University of California San Diego(加州大学圣地亚哥分校);University of Montana(蒙大拿大学);Oakland University(奥克兰大学);University of Pennsylvania(宾夕法尼亚大学)
School of Automation, Huazhong University of Science and Technology(华中科技大学自动化学院);State Key Lab of Digital Manufacturing Equipment and Technology(数字制造装备与技术国家重点实验室);School of Mechanical Science and Engineering, Huazhong University of Science and Technology(华中科技大学机械科学与工程学院);Department of Cognitive Robotics, Delft University of Technology(代尔夫特理工大学认知机器人系);Department of Engineering, University of Cambridge(剑桥大学工程系);Luxembourg Centre for Systems Biomedicine, University of Luxembourg(卢森堡系统生物医学中心,卢森堡大学)
Proceedings of the The First Workshop on Verification and Validation of Cyber-Physical Systems
Reykjavík, Iceland(冰岛雷克雅未克);MITL Specification Debugging for Monitoring of Cyber-Physical Systems(网络物理系统监控的MITL规格调试);Automatic Synthesis of Controllers from Specifications using Control Certificates(使用控制证书从规范自动合成控制器);A Compositional Framework for Preference-Aware Agents(偏好感知代理的组合框架);Output Feedback Controller Design with Symbolic Observers for Cyber-physical Systems(网络物理系统符号观测器输出反馈控制器设计);Towards an Approximate Conformance Relation for Hybrid I/O Automata(混合I/O自动机近似一致性关系);On Nonlinear Prices in Timed Automata(时序自动机中的非线性价格);Towards the Verification of Safety-critical Autonomous Systems in Dynamic Environments(动态环境中安全关键自主系统的验证)
Federated Learning for Cyber Physical Systems: A Comprehensive Survey
School of Engineering, Deakin University(德克萨斯大学工程学院);Department of Electrical and Computer Engineering, University of Alabama in Huntsville(阿拉巴马大学亨茨维尔分校电子与计算机工程系);The School of Engineering, Royal Melbourne Institute of Technology University(皇家墨尔本理工学院工程学院);Elmore Family School of Electrical and Computer Engineering, Purdue University(普渡大学埃尔莫尔家族电子与计算机工程学院);Department of Electrical and Computer Engineering, Princeton University(普林斯顿大学电子与计算机工程系)