The aviation industry is associated with high precision and accuracy standards of the manufactured components, and thus the need to ensure precise quality control. Measurement processes, depending on the manufactured components, take place before, during and after the processing stage. Optical scanners can be used for these measurements, the measurement results of which can be displayed on the operator panel or used to prepare a report. The innovative approach is to measure, compare the results with a pattern, send the deviations to a neural decision-making system, select the forces and send the results to a robot controller for adaptive machining. The presented proprietary solution includes a data acquisition system, a neural decision-making system and a robot that carries out the machining process via force control. The proposed solution was verified on aviation components. During the process parameter optimization stage for the diffuser and ADT gearbox, the points describing the change in width of the chamfer being performed and the blade thickness in the control sections were approximated.

The article discusses the quality testing of a measuring system consisting of a CNC machine with measuring probes. The research was conducted in a broader context regarding the implementation of the closed door technology, i.e., production without human intervention, in an aviation plant manufacturing aircraft gearbox systems. This technology may involve automated measuring operations performed in machining centers, and not in measuring laboratories, provided that the quality of the measurements is appropriate. The aim of the study was to investigate whether the CNC machining device can be used to measure the geometric features of aircraft gearbox housing. For this purpose, measurement experiments were carried out with the use of three different probes. Measurements were carried out using four sequences of increasing complexity, so that, after error analysis, it was possible to find the causes of possible irregularities. A reference ring with known dimensions and position in the working space of the machine was used for the measurements performed as part of the assessment of the measurement system. The quality of the measurements was evaluated with the use of repeatability and reproducibility testing and statistical process control. The analysis results showed that the tested measurement system ensures adequate accuracy and repeatability, and the measurement process is characterized with adequate efficiency in relation to the manufacturing tolerance of the components produced using the machine. Thus, it was proven that the measurement process can be carried out on a machining device, which enables its integration into the closed door technology.

This paper deals with the development of dimensional control technology for the production of ADT (accessory drive train) gearbox housing. The project included the development of a robotic geometry inspection station for thin-walled aerospace casings. Laser profilometers from two brands, Cognex and Keyence, were used to measure. A proprietary software solution for arc measurements is presented. The obtained solutions were compared and verified in accordance with the requirements of the manufacturer Pratt and Whitney Rzeszów S.A. The results of the work indicated that correct solutions were obtained with a very large reduction in control time. In addition, the measurement is fully automated and transferable to the company’s electronic systems.

This paper investigated the design of a system to monitor the status of the robotic welding process of thin-walled components for an aircraft jet engine. Opportunities to measure and log processing parameters, such as welding speed, current, and voltage, on the existing production cell were taken. The acquired data were processed using elements of descriptive statistics. Obtained indicators were matched with physical inspection results of the weld’s quality. The adopted methodology was used to identify an essential parameter determining the presence of defined weld defects. The developed solution was implemented in the production cell.

The impulse for writing the paper is the observation of the works related to the implementation of robotization of processes such as machining, glue application, welding and painting. The abovementioned processes, in addition to the correct implementation of the trajectory, require the definition of various parameters (e.g., speed) in the robot’s software. In the trajectories where the reconfiguration of the robot arms is observed, there are significant errors in the implementation of the defined speed. Robotic technology suppliers, in the event of speed disturbances, manually increase the defined speed value or experimentally select other parameters. It is a cumbersome process, and the lack of information about the process parameters makes it time-consuming and inaccurate. In this paper, one representative process is selected, namely machining performed with various tools by ABB robots. In order for the robotic process to be controlled, it is necessary to compare the defined path with the speed profile. Then, the speed parameters can be controlled and corrected. The approach proposed in the paper allows for improving the quality of implemented robotic processes. It presents the available IT tools for station monitoring and how to use them. The advantages of the proposed solutions and their limitations are shown in the examples of implementation of robotic stations in the industry.

This paper deals with the development of dimensional control technology for the production of accessory drive train (ADT) gearbox housing, according to the closed door technology approach. The work presents the methodology of the final inspection of bearing seat position deviation by replacing the coordinate measuring machines (CMMs) with a computerized numerical control (CNC) machine and adaptive neuro-fuzzy inference system. The results of the work indicated that correct solutions were obtained. In addition, the technological process of manufacturing is fully automated and performed entirely on the production line.