TY  - JOUR
PY  - 2014//
TI  - Multi-parameter prediction of driver's lane change behavior based on real-world driving tests
JO  - Chang'an Daxue Xuebao (Ziran Kexue Ban)
A1  - Ma, Yong
A1  - Fu, Rui
A1  - Guo, Ying-Shi
A1  - Yuan, Wei
A1  - Wu, Hai-Bo
SP  - 101
EP  - 108
VL  - 34
IS  - 5
N2  - When lane change maneuver is executed, driver's wrong decision is easy to cause traffic accidents. In this paper, the parameters of vehicle running, driver manipulation and head movements, as well as road environment were collected in real-world driving tests. By comparing and analyzing the data in lane change intention stage and lane keeping stage, the parameters that can characterize driver lane change intentions and behavior were extracted. BP neural network model was established to predict lane change behavior by using different parameters as the input vector, and eventually determined the input feature vector. The research results indicate that the prediction accuracy in 2 seconds before driver's lane change is 94.4%. It shows that the model can effectively predict driver's lane change behavior with a high accuracy.<p />  <p>Language: zh</p>
LA  - zh
SN  - 1671-8879
UR  - http://dx.doi.org/
ID  - ref1
ER  -