Abstract
III
Abstract
Vehicle trajectory prediction is one of the key technologies in the research field
of unmanned vehicle technology.Accurate prediction of the motion trajectory of
vehicles in the surrounding environment of intelligent vehicles is conducive to the
safe and comfortable driving of intelligent vehicles.In this paper,in order to solve the
problem of insufficient prediction accuracy of existing methods,this paper first
proposes an improved LSTM（Long Short-Term Memory） classification model to
identify vehicle behavior, and then proposes a trajectory prediction model combined
with vehicle behavior recognition, which can fully consider vehicle state
information,behavior information and interactive information. On this basis, an
acceleration trajectory optimization algorithm is proposed to make the prediction
trajectory more accurate.The main work and innovations of this paper include:
(1)A vehicle behavior recognition method based on improved LSTM
classification model is proposed in this paper.In this paper,based on the traditional
LSTM classification model, the traditional LSTM classification model is improved,
an improved LSTM classification model is obtained, and a better vehicle behavior
recognition effect is obtained.Compared with the traditional machine learning
method,MLP model and traditional LSTM classification model,better recognition
results are obtained.
(2)Based on the vehicle behavior recognition model,a trajectory predictio
n model is proposed based on LSTM. Finally, an acceleration trajectory opti
mization algorithm is proposed, and simulation experiments are carried out o
n the NGSIM data set. Compared with the method without vehicle behavior
recognition and the method of direct output trajectory prediction, better expe
5星 · 超过95%的资源 176 浏览量
2023-08-29
10:30:19
上传
评论 2
收藏 2.79MB PDF 举报 
m0_748523222024-04-01资源内容详细，总结地很全面，与描述的内容一致，对我启发很大，学习了。
VIP享7折，此内容立减3.57元 
