This dataset captures sensor and physiological data collected during controlled wheelchair fall simulations, designed to support research on fall detection and human motion analysis. Participants wore GW4 smartwatches on both arms, a Polar® H10 heart rate monitor, and safety equipment including a helmet, elbow, and knee pads. Each session began with the participant seated in a wheelchair inclined at approximately 10° on a crash pad, with arms laterally positioned on the wheels. Participants performed three fall trials in four directions: forward, backward, left, and right. After each fall, they returned to the fundamental seated position for 10 seconds. Data collection was conducted using two wheelchair types: standard and exercise-adapted. The dataset includes synchronized motion data from the smartwatches and heart rate measurements, providing comprehensive coverage of the pre-fall, fall, and recovery phases. Due to privacy and ethical considerations, only a portion of the dataset may be publicly available. Researchers interested in accessing the full dataset must contact the authors and submit a formal request, which will be reviewed on a case-by-case basis to ensure compliance with privacy regulations and ethical standards. The PGD is made available here.