UB-LSTM: A Trajectory Prediction Method Combined with Vehicle Behavior Recognition

Haipeng Xiao,Chaoqun Wang,Miguel Angel Sotelo,Rendong Wang,Cao Bo,Youchun Xu,Zhixiong Li

doi:10.1155/2020/8859689

Abstract

In order to make an accurate prediction of vehicle trajectory in a dynamic environment, a Unidirectional and Bidirectional LSTM (UB-LSTM) vehicle trajectory prediction model combined with behavior recognition is proposed, and then an acceleration trajectory optimization algorithm is proposed. Firstly, the interactive information with the surrounding vehicles is obtained by calculation, then the vehicle behavior recognition model is established by using LSTM, and the vehicle information is input into the behavior recognition model to identify vehicle behavior. Then, the trajectory prediction model is established based on Unidirectional and Bidirectional LSTM, and the identified vehicle behavior and the input information of the behavior recognition model are input into the trajectory prediction model to predict the horizontal and vertical speed and coordinates of the vehicle in the next 3 seconds. Experiments are carried out with NGSIM data sets, and the experimental results show that the mean square error (MSE) between the predicted trajectory and the actual trajectory obtained by this method is 0.124, which is 97.2% lower than that of the method that does not consider vehicle behavior and directly predicts the trajectory. The test loss is 0.000497, which is 95.68% lower than that without considering vehicle behavior. The predicted trajectory is obviously optimized, closer to the actual trajectory, and the performance is more stable.

Highlights

Trajectory prediction is an important research direction in the field of autopilot [1, 2]. e research on the decisionmaking characteristics of the driver shows that factors such as the relative speed and relative distance between the car and the surrounding moving vehicles will greatly affect the driver’s decision [3] and affect the driving safety
The input characteristics of each trajectory point are input into the vehicle behavior recognition model to get the behavior characteristics of the vehicle. en, the behavior characteristics, vehicle state characteristics, and interactive information characteristics are input into the trajectory prediction model together
A vehicle behavior recognition model is trained to make a prediction combined with the vehicle trajectory prediction model

Summary

Introduction

Trajectory prediction is an important research direction in the field of autopilot [1, 2]. e research on the decisionmaking characteristics of the driver shows that factors such as the relative speed and relative distance between the car and the surrounding moving vehicles will greatly affect the driver’s decision [3] and affect the driving safety. Intelligent vehicles need to improve their driving safety by predicting the trajectories of the moving vehicles around them in real time. Improving the accuracy of vehicle trajectory prediction is the most urgent problem to be solved. The accurate identification of vehicle behavior is very important to improve the accuracy of vehicle trajectory prediction, so this paper will take vehicle behavior recognition into consideration. E advantage of Kalman filter [13] is that the prediction in a short time (1 step or 2 steps) can be judged stably and accurately, but the trajectory prediction for a long time (such as more than 3 seconds or more than 5 steps) will seriously affect the prediction accuracy due to the increase of prediction error, and the model is very complex and easy to be affected by external noise Its advantage is that it can Journal of Advanced Transportation calculate the probability of maintenance capability and multiple degraded state systems, but it is not suitable for long-term prediction and is affected by the external environment. e advantage of Bayesian model [12] is that it is not sensitive to missing data but needs to know a priori probability, and a priori probability often depends on assumptions, and there can be many hypothetical models; at some point, the prediction effect is poor due to the hypothetical a priori model. e advantage of Kalman filter [13] is that the prediction in a short time (1 step or 2 steps) can be judged stably and accurately, but the trajectory prediction for a long time (such as more than 3 seconds or more than 5 steps) will seriously affect the prediction accuracy due to the increase of prediction error, and the model is very complex and easy to be affected by external noise

Methods

Findings

Discussion

Conclusion