EAGER/Cybermanufacturing: Just-In-Time Compilation of Product Manufacturing Data to Machine Instructions via an Industrial Machine Operating System

未来的工厂将由支持网络的软件和物理机器的无缝集成推动，以提高制造灵活性和敏捷性，并最终提高生产效率。在这种网络制造环境中，基本的实物资产是车间中的单个制造机器，它将原材料/在制品加工成最终产品。几十年来，这些机床的核心设计配置几乎没有变化，其专有硬件和控制配置因机床制造商而异。这种支离破碎和封闭的机器体系结构阻碍了用于网络物理机器的软件的开发。这一早期概念探索性研究(EARGER)奖支持将制造车间中的单个机床转变为软件定义的机器的基础研究，在软件定义的机器中，软件决定特定机器的操作特性。这项研究的结果将为开放源代码的工业机器操作系统(IMOS)奠定基础，该系统将为个人开发人员、研究人员和组织提供为各种制造过程编写软件应用程序的能力。本课题的研究涉及制造工艺规划、嵌入式计算、编译器优化和机床控制系统理论等多个学科。这项结合了制造和计算科学的多学科研究将培养新一代科学家和工程师，从而提高美国在网络制造方面的竞争力。工业机器操作系统将在机床的硬件和软件组件内提供水平和垂直资源管理。该操作系统将通过一个软件堆栈建立在开源的Linux内核上，该软件堆栈提供硬件抽象、打包的库、应用程序框架和安装在制造机器(如铣床)上的用户定义的应用程序。操作系统概念的价值将通过对网络制造至关重要的两个独特方面来展示：1)演示在使用国际公认的产品文档格式的同时，将设计和制造过程数据直接传输到嵌入式计算平台上的机器指令；以及2)在嵌入式计算系统内的运行时间存储器期间，建立制造过程数据的实时编译和代码优化背后的规则。上述优点将通过机器操作系统框架来展示，该框架旨在可扩展和适应各种内存体系结构。

The factory of the future will be driven by a seamless integration of cyber-enabled software and physical machines to increase manufacturing flexibility, agility and ultimately production efficiency. In this cybermanufacturing environment, the foundational physical asset is the individual manufacturing machine on a shop floor which processes raw materials/in-process workpieces into final products. The core design configuration of these machines has changed little over decades, with proprietary hardware and control configurations that vary among machine tool builders. This fragmented and closed machine architecture hinders the development of software for cyberphysical machines. This EArly-concept Grant for Exploratory Research (EAGER) award supports fundamental research into transforming the individual machine tool on a manufacturing shop floor into software defined machines, wherein software determines the operational characteristics of a particular machine. The results of this research will lay the foundation of an open source Industrial Machine Operating System (IMOS) which will provide the capability to individual developers, researchers and organizations to write software applications for various manufacturing processes. This research involves several disciplines which include manufacturing process planning, embedded computing, compiler optimization and machine tool control system theory. This multi-disciplinary research combining manufacturing and computing sciences will educate a new generation of scientists and engineers leading to improved US competitiveness in cybermanufacturing. The Industrial Machine Operating System will provide horizontal and vertical resource management within a machine tool's hardware and software components. This operating system will be built on the open sourced Linux kernel by a software stack that provides hardware abstraction, packaged libraries, an application framework and user defined applications installed on a manufacturing machine such as a milling machine. The value of the operating system concept will be demonstrated through two unique aspects critical to cybermanufacturing: 1) Demonstration of the direct transfer of design and manufacturing process data to machine instructions on an embedded computing platform while making use of internationally accepted product documentation formats, and 2) Establishment of the rules behind Just-In-Time compilation and code optimization of manufacturing process data during run-time memory within an embedded computing system. The above merits will be demonstrated through the machine operating system framework, intended to be scalable and adaptable to various memory architectures.