A LSTM Based Bus Arrival Time Prediction Method

Bus arrival time prediction not only provides convenience for passengers, but also helps to improve the efficiency of intelligent transportation system. However, the low prediction accuracy becomes one of great puzzle. Considering both historic data and real-time traffic condition, in this paper, a new bus arrival time prediction method is proposed. A LSTM training model is used to get historic cruising speed, while two traffic factors are defined to illustrate real-time traffic state. Then a bus arrival time prediction is established based on speed values. Validation experiment results show that proposed method could predict the bus arrival time in special time span accurately.