Smart Workpiece Carriers for Production Simulations

Integration of Motor Control and Sensor Technology on an ESP32-S3 PCB

Abstract

In the context of AI-based predictive maintenance approaches, reproducible and seg-mentable time-series data are required in order to detect anomalies at an early stage and derive data-driven maintenance strategies. For the application scenario of a food production process, a model-based production workflow was implemented using a 1:24 slotcar track. The corre-sponding vehicle represents the workpiece carrier. For this purpose, motor control and a 9-axis inertial sensor system (acceleration, angular rate, magnetic field) were integrated on a custom-designed ESP32-S3 PCB.
The resulting intelligent workpiece carrier passes through defined process sections. Sec-tion markers enable the synchronization of the IMU time series as well as the derivation of segment and lap times. To generate segmentable time-series data, multiple datasets with more than 100 laps each were recorded both under normal operating conditions and under defined disturbance scenarios.
The evaluation shows a high repeatability of segment-specific signatures under reference conditions, while disturbances lead to measurable and locally limited deviations in timing and IMU-based features. Overall, the system provides a compact platform for the systematic gene-ration of datasets and the validation of predictive maintenance and anomaly detection approa-ches in research and education.