Description of Invention
This study develops a driver drowsiness detection system using Heart Rate Variability (HRV) signals and Recurrent Neural Networks (RNNs). Three models were tested: Simple RNN, Bidirectional RNN, and Deep RNN. The Deep RNN outperformed the others with 95% accuracy, 94% precision, and 93% recall. The results show that HRV data, processed through RNN models, can effectively detect drowsiness in real-time, providing a low cost alternative to traditional methods.