SBIR Phase I: An artificial intelligence system for autonomous numerical control programming for advanced manufacturing

这个小型企业创新研究(SBIR)第一阶段项目的更广泛影响是，通过帮助提高精密机械师的生产率，并激发新进入劳动力市场的人对精密制造领域的更大兴趣，来应对美国技术技能差距危机下制造业回流的挑战。计算机数控(Computer Numerical Control，简称CNC)是精密制造中的关键工序，是对机床操作进行计算机自动化管理的过程。对数控机床进行编程需要具有广泛而专业的技术技能的专业人员。随着技工学校毕业生的减少，数控编程技术已成为制造运营的瓶颈，创造了一个迫切的未得到满足的需求，即从根本上简化和自动化的数控编程工作流程，以改善精密制造的可及性。该项目旨在开发一种人工智能工具来自动化数控编程过程，从而帮助缓解数控编程瓶颈，并在减少制造时间和成本的同时填补技能缺口。该项目将扩大在精密制造中使用人工智能(AI)进行创成式数控编程的研究。人工智能具有自动化数控编程工作流程中的各个步骤的潜力；然而，这项研究一直局限于受限的加工条件，没有专注于优化过程，从而限制了其在现实世界中的适用性。本项目对这一研究进行了扩展，以开发出一种商业上可行的解决方案，用于数控编程的最佳刀具和加工参数选择阶段。简而言之，该项目将解决算法开发和优化的技术挑战，以提高刀具和加工参数的准确选择。结果将在数控机床上进行验证。在第一阶段结束时，将开发一个概念验证解决方案，以自动选择用于数控钻削和铣削的最佳刀具和加工参数。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact of this Small Business Innovation Research (SBIR) Phase I project is to address the challenge of reshoring manufacturing given the technical skills gap crisis in the U.S. by helping increase the productivity of precision machinists and sparking greater interest in the field of precision manufacturing among new workforce entrants. Computer Numerical Control (CNC), the computer-automated management of machining tools’ operations, is a key process in precision manufacturing. Programming CNC machines requires professionals with vast and specialized technical skills. With fewer graduates of trade schools, CNC programming know-how has become a bottleneck to manufacturing operations, creating an urgent unmet need for a radically simplified and automated CNC programming workflow to improve accessibility to precision manufacturing. This project aims to develop an artificial intelligence tool to automate the CNC programming process, thereby helping ease the CNC programming bottleneck and fill the skills gap at reduced manufacturing time and costs. This project will expand research on the use of Artificial Intelligence (AI) for generative CNC programming in precision manufacturing. AI has the potential to automate various steps in the CNC programming workflow; however, this research has been limited to constrained machining conditions and has not focused on optimization procedures, thereby limiting its real-world applicability. This project expands this research to develop a commercially viable solution for the optimal cutting tool and machining parameter selection stage of CNC programming. Briefly, this project will address technical challenges of algorithm development and optimization to improve accurate selection of tool and machining parameters. The results will be validated on CNC machines. At the end of Phase I, a proof-of-concept solution will be developed to automate the optimal cutting tool and machining parameters selection for CNC drilling and milling.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.