The publication includes a review of information on the methods of pavement condition recognition using various methods. Measurement system has been presented that allows to determine the condition of the pavement using the Inertial Measurement Unit (IMU) and machine learning methods. Three machine learning methods were considered: random forest, gradient boosted tree and custom architecture neural network (roadNet). Due to the developed system the set of learning and validation data was created on 3 vehicles: Opel Corsa, Honda Accord, Volkswagen Passat. All of the listed vehicles have front wheel drive. The presented machine learning methods have been compared with each other. The best accuracy on the validation set was achieved by the artificial neural network (ANN). The study showed that asphalt condition classification is possible and the developed system fulfils its task.

This paper discusses a comprehensive study to develop a machine learning model for detecting unwanted vibrations during the 3D printing process. Undesired vibrations can significantly degrade print quality, leading to defects such as void formation, poor surface quality and improper layer...

This study focused on analyzing vibrations during waterjet cutting with variable technological parameters (speed, v_fi; and pressure, p_i), using a three-axis accelerometer from SEQUOIA for three different materials: aluminum alloy, titanium alloy, and steel. Difficult-to-machine materials often require specialized tools and machinery for machining; however, waterjet cutting offers an alternative. Vibrations during this process can affect the quality of cutting edges and surfaces. Surface roughness was measured by contact methods after waterjet cutting. A machine learning (ML) model was developed using the obtained maximum acceleration values and surface roughness parameters (Ra, Rz, and RSm). In this study, five different models were adopted. Due to the characteristics of the data, five regression methods were selected: Random Forest Regressor, Linear Regression, Gradient Boosting Regressor, LGBM Regressor, and XGBRF Regressor. The maximum vibration amplitude reached the lowest acceleration value for aluminum alloy (not exceeding 5 m/s²), indicating its susceptibility to cutting while maintaining a high surface quality. However, significantly higher acceleration amplitudes (up to 60 m/s²) were registered for steel and titanium alloy in all process zones. The predicted roughness parameters were determined from the developed models using second-degree regression equations. The prediction of vibration parameters and surface quality estimators after waterjet cutting can be a useful tool that for allows for the selection of the optimal abrasive waterjet machining (AWJM) technological parameters.

Artificial neural network models for the task of detection are used in many fields and find various applications. Models of this kind require adequate computational resources and thus require adequate energy expenditure. The increase in the number of parameters, the complexity of architectures, and the need to process large data sets significantly increase energy consumption, which is becoming a key sustainability challenge. Optimization of computing and the development of energy-efficient hardware technologies are essential to reduce the energy footprint of these models. This article examines the effect of the type of model, as well as its parameters, on energy consumption during inference. For this purpose, sensors built into the graphics card were used, and software was developed to measure the energy demand of the graphics card for different architectures of YOLO models (v8, v9, v10), as well as for different batch and model sizes. This study showed that the increase in energy demand is not linearly dependent on batch size. After a certain level of batch size, the energy demand begins to decrease. This dependence does not occur only for n/t size models. Optimum utilization of computing power due to the number of processed images for the studied models occurs at the maximum studied batch size. In addition, tests were conducted on an embedded device.