In-situ evaluation of hole quality and cutting tool condition in robotic drilling of composite materials using machine learning

The massive adoption of industrial robots in the manufacturing sector has significantly increased automation of installation and inspection procedures, particularly benefiting the aerospace industry, where large volumes of holes are drilled in each aircraft. However, mechanical drilling remains challenging when dealing with composite materials due to their inherently heterogeneous structure. This work presents a novel approach for in-situ hole quality inspection utilising integrated sensor data of an industrial robotic drill, combined with a machine learning model. Additionally, a novel classification approach for evaluating hole quality is proposed. This study employed a KUKA industrial robot, fitted with a multifunctional end-effector, to drill holes in a composite material used in aerospace applications. An ensemble neural network (ENN) model, which combines an artificial neural network with a genetic algorithm, was used to assess the quality of these drilled holes. The model was specifically developed and tested on the machined holes to relate process input parameters and drilling torques to hole quality. The model predictions were validated with six unseen datasets, of which five were predicted accurately. A full factorial study of the process parameters was conducted using analysis of variance (ANOVA) to investigate the relationship between tool condition and drilling torque. The results of the ANOVA show that tool condition is the largest contributor to drilling torque. The method proposed in this work, which allows real-time monitoring of hole quality, has the potential to improve manufacturing productivity of drilled components while ensuring high-quality end products.