Predicting Motion Trajectory of Bedload Pebble based on LSTM_Attention_Kalman Model

Predicting the motion trajectory of bedload pebbles is of great significance for studying riverbed evolution and regulation, reservoir sedimentation, and river management. This paper combines traditional method with neural network, and proposes a LSTM-Attention-Kalman model to predict the motion trajectory of pebbles based on the collected motion parameters. The experimental results show that the method proposed in this paper has good effectiveness in fitting and predicting the motion trajectory of bedload pebbles.