Abstract: A third of an individual’s life is spent sleeping, and the quality of sleep is related to the recovery of energy in daily life, the maintenance of memory and thinking skills, concentration, and other processes. Sleep assessment provides an important basis for evaluating sleep quality and preventing or diagnosing sleep disorders. Currently, polysomnography (PSG) is the gold standard for sleep staging. However, PSG monitoring requires the attachment of a large number of electrodes to obtain physiological signals, which then affects the individual’s sleep quality. In clinical practice, sleep monitoring must be performed in a hospital and requires a sleep technician to keep vigil and manually record the sleep stages, which is time-consuming, labor-intensive, complex, and costly. The monitoring pad based on piezoelectric film collects piezoelectric signals without requiring that electrodes be attached to the individual, which is an unconstrained, non-contact, and non-disturbance monitoring method. The monitoring pad can measure the force from the thoracic motion generated by human respiration, the force caused by the heartbeat and blood flow, and the force generated by involuntary body movement during sleep. These physiological activities pass through the pad and other media to the sensor, which is then converted into electrical signals. The pad can be placed in different positions on the bed and can monitor physiological activities, such as heartbeat, breathing, body turning, and leg movement, without direct contact with the human body. Considering that the pad can be placed in the sleep environment for a long time, it can achieve long-term and discreet sleep monitoring. To date, most studies have focused on extracting features related to the heart rate, respiration, and body movement from piezoelectric signals. However, in actual daily sleep situations, the change of sleep position, body movement, and even the change of blood pressure at night affect the strength of heartbeat information obtained by the sensor, which can considerably affect the continuous heart rate detection. Meanwhile, continuous heart rate detection throughout the night when in the natural sleep state still requires a breakthrough. This study recruited healthy young adults as subjects and used a sleep monitoring mat equipped with a piezoelectric film sensor synchronized with PSG to collect data from 61 nights of sleep. A total of 53 features were extracted from signals such as cardiac activity, respiration, and body movement detected by the piezoelectric film, including 12 time-domain features, 36 frequency-domain features, and 5 higher-order statistical features. Using a bagging tree model, the study performed a four-stage classification (wake, N1+N2, N3, REM (rapidly eye movement)), three-stage classification (wake, N1+N2+N3, REM), and two-stage classification (wake, N1+N2+N3+REM) to predict sleep stages, and these predictions were validated against the PSG sleep stage labels. The final testing accuracies for the four-, three-, and two-stage classifications were 80.5%, 85.3%, and 96.3%, with Kappa values of 0.74, 0.78, and 0.93, respectively, thus demonstrating excellent performance compared to similar studies. This demonstrates the sleep monitoring mat’s capability for accurate sleep assessment and monitoring, thereby offering additional possibilities for home sleep monitoring and assessment. This method also reveals the sleep monitoring mat’s considerable effectiveness for sleep assessment when respiratory and heart rate measurements are disrupted.