Development of automated robotic vision inspection system for accurate quality control

开发自动化机器人视觉检测系统以实现精确的质量控制

• 批准号: 561153-2020
• 负责人: Alirezaee, Shahpour
• 金额: $ 3.1万
• 依托单位: University of Windsor
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=721903
• 关键词: Development; automated; robotic; vision; inspection

Quality control (QC) is an indispensable part of almost every manufacturing industry, especially those related to higher safety standards such as the automotive industry. Despite recent advances in automotive quality control, most of the QC processes were performed on randomly selected manufactured goods and outside of the manufacturer's product line. Spot-checking QC does not fully guarantee well-being and safety of the products, when meeting the stringent standards is essential. Off-site QC requires products measurement recording for later reference and reduces assembly line throughput. Most of the QC systems are still reliant on human operators to perform some parts of the inspection task. To address these shortcomings, this proposal aims to replace human operators with robots to perform inspection tasks automatically in the factory environment. The main goal of this project is to carry out: (i) dimensional measurement and (ii) flaw detection in bore cylinders. for the first objective, several features of the bore such as diameter, taper, and out-of-round will be inspected to verify if these parameters are within the tolerance set defined by the manufacturer. The second goal of this system will be to detect any flaws such as cracks, cavities or pinholes. Flaw detection is particularly challenging because defects may happen in any shapes or any sizes on the inside cylinder walls, and they are unnoticeable to human eye inspection. To the best to our knowledge, this is the first attempt to perform flaw detection automatically on-site. To employ robots in dynamic and uncontrolled workplaces, robots should cope with several unforeseen circumstances. Robot learning is a new paradigm which enables robots to handle variations in the working environment, such as differences in the item position on the product line or illumination variation. We will use learning from demonstration (LfD) technique and other cutting-edge technologies, such as computer vision, artificial intelligence, and pattern recognition. The outcome of this project will have considerable impacts on automotive sector as well as other critical industries.

质量控制（QC）几乎是每个制造业不可或缺的一部分，特别是那些与汽车行业等更高安全标准相关的行业。尽管最近在汽车质量控制方面取得了进展，但大多数质量控制过程都是在随机选择的制成品上进行的，并且不在制造商的生产线上。抽查QC并不能完全保证产品的健康和安全，当满足严格的标准是必不可少的。非现场QC需要产品测量记录以供以后参考，并降低了装配线的吞吐量。大多数质量控制系统仍然依赖于人类操作员来执行检查任务的某些部分。为了解决这些缺点，该提案旨在用机器人取代人类操作员，在工厂环境中自动执行检查任务。该项目的主要目标是：（一）进行尺寸测量和（二）进行镗孔圆柱体的缺陷检测。对于第一个目标，将检查孔的几个特征，例如直径、锥度和不圆度，以验证这些参数是否在制造商定义的公差范围内。该系统的第二个目标是检测任何缺陷，如裂缝、空洞或针孔。缺陷检测特别具有挑战性，因为缺陷可能以任何形状或任何尺寸发生在气缸内壁上，并且它们对于人眼检查是不可察觉的。据我们所知，这是首次尝试在现场自动进行缺陷检测。为了在动态和不受控制的工作场所使用机器人，机器人应该科普一些不可预见的情况。机器人学习是一种新的范例，它使机器人能够处理工作环境中的变化，例如生产线上物品位置的差异或照明变化。我们将使用从演示中学习（LfD）技术和其他前沿技术，如计算机视觉，人工智能和模式识别。该项目的成果将对汽车行业以及其他关键行业产生重大影响。