A Double-Layer Vehicle Speed Prediction Based on BPNN-LSTM for Off-Road Vehicles

The accurate prediction of vehicle speed is crucial for the energy management vehicles. existing (VSP) methods mainly focus on road vehicles and rarely off-road In this paper, a double-layer VSP method based backpropagation neural network (BPNN) long short-term memory (LSTM) proposed. First all, considering motion characteristics vehicles, problem established relationship between variables in carefully analyzed. Then, framework presented, which consists information update layers. layer by using LSTM to predict horizon, built BPNN information. Finally, with help mining truck loader operation scenarios, proposed compared analytical method, recurrent (RNN) terms accuracy. results show that, under premise ensuring real-time performance, average error BPNN-LSTM two scenarios reduces 48.14%, 35.82% 30.09% other three methods, respectively. provides new solution predicting effectively improving