CBANet: A Compact Attention-Based CNN-BiLSTM Network for Aggressive Driving Event Detection
Researchers have developed CBANet, a new deep learning framework designed to detect aggressive driving events using vehicle sensor data. The model addresses challenges like data imbalance and driver variability by constructing engineered dynamic features and employing a stable training strategy with oversampling and class-weighted loss. CBANet aims to improve road safety by more accurately identifying risky driving behaviors, outperforming existing baselines in minority-class recall and safety-critical metrics. AI
IMPACT This new model could enhance road safety systems by improving the detection of aggressive driving behaviors.