Sleep disorders represent significant challenges for public health, but also the quality of sleep is closely linked to overall well-being. Monitoring posture can provide highly relevant information for assessing sleep quality, particularly for improving it. This study utilizes pressure signals captured by “Smart BeddingTM”, a device with high-resolution sensors integrated into a textile sheet, to analyze sleep posture. Detailed pressure maps are constructed to visualize pres-sure distribution during sleep. Machine learning algorithms are applied to detect sleep positions based on the pressure data. The methodology involves data collec-tion from healthy subjects using polysomnography equipment and “Smart BeddingTM”, statistical analysis of pressure signals, data preprocessing, feature extraction, and classification using machine learning techniques. The study employed two classification methodologies: a single-step classification and a two-step classification. The models showed promising performance in classifying sleep positions, with the two-step classification approach achieving higher performance in identifying sleep postures compared to the single-step classification.