Driving Intention Identification Based on Long Short-Term Memory Neural Network

Abstract

In order to avoid frequent or accidental shift problems during the driving process, it is necessary to implement identification of driving intention based on vehicle driving data. In this study, the Long Short-Term Memory (LSTM) Neural Network is proposed to identify driving intentions in real time. First, according to the vehicle road test data, each driving intention to be identified is defined. Then, the intentions when driving on a straight and flat road are divided into acceleration, rapid acceleration, cruise, deceleration and rapid deceleration. Subsequently, a LSTM classification model is established to identify the driving intention with inputs of opening degree of the accelerator pedal, vehicle speed and brake pedal force. Identification results reveal that the highest accuracy of the proposed algorithm attains 95.36%, which is around 20% higher than that of the traditional back propagation neural network.