From Anomaly Detection to Defect Classification.

This paper proposes a new approach to defect detection system design focused on exact damaged areas demonstrated through visual data containing gear wheel images. The main advantage of the system is the capability to detect a wide range of patterns of defects occurring in datasets. The methodology is built on three processes that combine different approaches from unsupervised and supervised methods. The first step is a search for anomalies, which is performed by defining the correct areas on the controlled object by using the autoencoder approach. As a result, the differences between the original and autoencoder-generated images are obtained. These are divided into clusters using the clustering method (DBSCAN). Based on the clusters, the regions of interest are subsequently defined and classified using the pre-trained Xception network classifier. The main result is a system capable of focusing on exact defect areas using the sequence of unsupervised learning (autoencoder)-unsupervised learning (clustering)-supervised learning (classification) methods (U2S-CNN). The outcome with tested samples was 177 detected regions and 205 occurring damaged areas. There were 108 regions detected correctly, and 69 regions were labeled incorrectly. This paper describes a proof of concept for defect detection by highlighting exact defect areas. It can be thus an alternative to using detectors such as YOLO methods, reconstructors, autoencoders, transformers, etc.

Mechatronic Device Control by Artificial Intelligence.

Nowadays, artificial intelligence is used everywhere in the world and is becoming a key factor for innovation and progress in many areas of human life. From medicine to industry to consumer electronics, its influence is ever-expanding and permeates all aspects of our modern society. This article presents the use of artificial intelligence (prediction) for the control of three motors used for effector control in a spherical parallel kinematic structure of a designed device. The kinematic model used was the "Agile eye" which can achieve high dynamics and has three degrees of freedom. A prototype of this device was designed and built, on which experiments were carried out in the framework of motor control. As the prototype was created through the means of the available equipment (3D printing and lathe), the clearances of the kinematic mechanism were made and then calibrated through prediction. The paper also presents a method for motor control calibration. On the one hand, using AI is an efficient way to achieve higher precision in positioning the optical axis of the effector. On the other hand, such calibration would be rendered unnecessary if the clearances and inaccuracies in the mechanism could be eliminated mechanically. The device was designed with imperfections such as clearances in mind so the effectiveness of the calibration could be tested and evaluated. The resulting control of the achieved movements of the axis of the device (effector) took place when obtaining the exact location of the tracked point. There are several methods for controlling the motors of mechatronic devices (e.g., Matlab-Simscape). This paper presents an experiment performed to verify the possibility of controlling the kinematic mechanism through neural networks and eliminating inaccuracies caused by imprecisely produced mechanical parts.

Waste Management with the Use of Heuristic Algorithms and Internet of Things Technology.

Studies have been performed to improve the process of waste management. They were fulfilled by changing the base of waste logistics management using a combination of intelligent algorithms and the IMPACT IoT platform instead of a human factor. The research was carried out on the example of real data with respect to waste management in a given area. The proposed solution includes a program that simulates the filling of specific waste containers located in various areas. The determined aspects are inconveniences on the routes, affecting the time of moving between the receiving points and the distances between the containers. The variability of the speed and intensity of the containers filling up over time is an additional factor taken into account. The proposed methods yielded the performance of the control of the containers' filling status in real time, which apparently results in the possibility of a reaction to the current demand just in time. The proposed solution enables the improvement of the waste logistics management process, including avoiding the too-frequent emptying of containers or overfilling them. The combination of the device prototype, the simulation program, and the developed algorithms opens the possibility for further research in the smart city and optimization areas.

Analysis of the Possibilities of Tire-Defect Inspection Based on Unsupervised Learning and Deep Learning.

At present, inspection systems process visual data captured by cameras, with deep learning approaches applied to detect defects. Defect detection results usually have an accuracy higher than 94%. Real-life applications, however, are not very common. In this paper, we describe the development of a tire inspection system for the tire industry. We provide methods for processing tire sidewall data obtained from a camera and a laser sensor. The captured data comprise visual and geometric data characterizing the tire surface, providing a real representation of the captured tire sidewall. We use an unfolding process, that is, a polar transform, to further process the camera-obtained data. The principles and automation of the designed polar transform, based on polynomial regression (i.e., supervised learning), are presented. Based on the data from the laser sensor, the detection of abnormalities is performed using an unsupervised clustering method, followed by the classification of defects using the VGG-16 neural network. The inspection system aims to detect trained and untrained abnormalities, namely defects, as opposed to using only supervised learning methods.

Analysis of Laser Sensors and Camera Vision in the Shoe Position Inspection System.

Inspection systems are currently an evolving field in the industry. The main goal is to provide a picture of the quality of intermediates and products in the production process. The most widespread sensory system is camera equipment. This article describes the implementation of camera devices for checking the location of the upper on the shoe last. The next part of the article deals with the analysis of the application of laser sensors in this task. The results point to the clear advantages of laser sensors in the inspection task of placing the uppers on the shoe's last. The proposed method defined the resolution of laser scanners according to the type of scanned surface, where the resolution of point cloud ranged from 0.16 to 0.5 mm per point based on equations representing specific points approximated to polynomial regression in specific places, which are defined in this article. Next, two inspection systems were described, where one included further development in the field of automation and Industry 4.0 and with a high perspective of development into the future. The main aim of this work is to conduct analyses of sensory systems for inspection systems and their possibilities for further work mainly based on the resolution and quality of obtained data. For instance, dependency on scanning complex surfaces and the achieved resolution of scanned surfaces.

Parameter Identification of Cutting Forces in Crankshaft Grinding Using Artificial Neural Networks.

The intensifying of the manufacturing process and increasing the efficiency of production planning of precise and non-rigid parts, mainly crankshafts, are the first-priority task in modern manufacturing. The use of various methods for controlling the cutting force under cylindrical infeed grinding and studying its impact on crankpin machining quality and accuracy can improve machining efficiency. The paper deals with developing a comprehensive scientific and methodological approach for determining the experimental dependence parameters' quantitative values for cutting-force calculation in cylindrical infeed grinding. The main stages of creating a method for conducting a virtual experiment to determine the cutting force depending on the array of defining parameters obtained from experimental studies are outlined. It will make it possible to get recommendations for the formation of a valid route for crankpin machining. The research's scientific novelty lies in the developed scientific and methodological approach for determining the cutting force, based on the integrated application of an artificial neural network (ANN) and multi-parametric quasi-linear regression analysis. In particular, on production conditions, the proposed method allows the rapid and accurate assessment of the technological parameters' influence on the power characteristics for the cutting process. A numerical experiment was conducted to study the cutting force and evaluate its value's primary indicators based on the proposed method. The study's practical value lies in studying how to improve the grinding performance of the main bearing and connecting rod journals by intensifying cutting modes and optimizing the structure of machining cycles.

Reasons for the Formation of Non-Fibrous Inclusions When Preparing Basalt Fibers by the Duplex Method.

Materials based on basalt fiber are widely used as thermal insulating material. These materials have a number of advantages, including their low thermal conductivity and fire resistance due to their natural composition. However, there is a significant drawback in that the material contain non-fibrous inclusions. The solution to this problem would significantly improve the working conditions of workers engaged in the production of materials from basalt fiber, as well as workers engaged in construction and installation works. In addition, the research will help to make completely new products, such as special fireproof paper and sterile medical materials. This article focuses on the reasons for the formation of non-fibrous inclusions in the production of this kind of material. The technology of producing canvases from superthin fiber in the duplex way is studied. The analysis of the production process is made. Certain technological and structural parameters of the influence on the formation of such inclusions are identified. Experiments are carried out and conclusions are drawn given formation of non-fibrous inclusions of various geometric shapes for various factors. A mathematical model of the process under consideration is built. The article draws conclusion on the application of these developments in the production cycle of creating materials based on basalt fiber.