Bringing Industry 4.0 manufacturing to life: Digital shadows, optimized trajectories, structural controls, and advanced mechatronics

第四次工业革命（工业4.0）旨在建立具有内置虚拟化、监控、自我优化和控制功能的智能工厂，以提高整个过程链的制造效率和产品质量。每台生产机器都有一个数字模拟模型，称为“数字影子”，并行运行，以预测、监测、表征、优化、控制和提高其性能。然而，就目前的知识状况而言，还无法以有效和工业上可行的方式实现这样一套雄心勃勃的目标。Discovery计划旨在通过开发新技术来支持工业4.0中设想的某些关键功能，从而帮助缩小这一差距。其中包括：基于过程制造数据的动态模型估计和仿真、轨迹优化、通过先进的实时控制进行动态过程改进，以及用于提高制造生产率的新型机电一体化设备。******除了在每个主题中进行深入的基础研究外，还提出了一种整体和综合的方法，其中每个主题的进展和结果将使在补充主题中追求深远的新思想。例如，数字阴影将与轨迹优化算法和高效的3D实体建模技术相结合。因此，所提出的轨迹优化将直接考虑零件质量结果作为优化约束，而不是传统上使用的生产机器运动轴的运动学（速度、加速度和加速度）限制。这种直接的方法有望大大降低生成轨迹的保守性，从而减少制造周期时间，并保持所需的零件精度要求。在这个探索项目中，所有主题领域之间的类似协同效应将被积极利用，以取得既具有科学新颖性又具有工业创新性和优越性的成果。******总体而言，这个探索项目的目标是大胆而雄心勃勃的新步骤，将虚拟和智能制造的知识状态转变到一个新的水平，同时培养博士后、博士、硕士和本科水平的高素质人才。由Erkorkmaz教授创立的滑铁卢大学精密控制实验室在机电设备、建模、识别、轨迹优化、精密控制和过程仿真方面有着非常成功的研究记录。通过早期的探索项目开发的许多核心思想已经导致了与加拿大国内外的小型，中型和大型公司的后续合作研发项目，并产生了技术和解决方案，这些技术和解决方案已经转移到工业中，以及CNC制造中的新商业产品。

The 4th Industrial Revolution (Industry 4.0) aims to establish smart factories with built-in virtualization, monitoring, self-optimization, and control functionalities, to achieve improved manufacturing efficiency and product quality across complete process chains. Each production machine would have a digital simulation model, called a digital shadow', running in parallel to predict, monitor, characterize, optimize, control, and enhance its performance. The current state of knowledge, however, falls short of being able to achieve such an ambitious set of goals in an effective and industrially viable manner. The proposed Discovery program aims to help close this gap, by developing new technologies in support of certain critical functionalities envisioned in Industry 4.0. These include: dynamic model estimation and simulation from in-process manufacturing data, trajectory optimization, on-the-fly process improvement through advanced real-time controls, and novel mechatronic devices to boost manufacturing productivity.******In addition to pursuing deep fundamental research in each of the themes, a holistic and integrative approach is proposed in which the progress and results from every theme will enable far-reaching new ideas to be pursued in the complementary themes. For example, digital shadows will be integrated with trajectory optimization algorithms and with highly efficient 3D solid modeling techniques. Thus, the proposed trajectory optimization will directly consider part quality outcomes as the optimization constraints, rather than the traditionally used kinematic (velocity, acceleration, and jerk) limits of the production machine's moving axes. This direct approach is expected to enable a dramatic reduction in the conservativeness of the generated trajectories, thereby reducing the manufacturing cycle time and enabling the desired part accuracy requirements to be retained. Similar synergies between all theme areas will be actively capitalized upon within this Discovery program, in order to achieve results that are both scientifically new and also industrially innovative and superior.******Overall, this Discovery program targets bold and ambitious new steps towards transforming the state of knowledge in virtual and intelligent manufacturing to the next level, while simultaneously training new highly qualified personnel at postdoctoral, PhD, master's, and undergraduate levels. The University of Waterloo's Precision Controls Laboratory, founded by Prof. Erkorkmaz, has a highly successful record of research in mechatronic devices, modeling, identification, trajectory optimization, precision controls, and process simulation. Many core ideas developed through earlier Discovery programs have led to collaborative follow-up R&D projects with small, medium, and large-scale companies within and outside Canada, and have resulted in technologies and solutions that have been transferred to industry, as well as new commercial products in CNC manufacturing.