Autonomous Vehicle Trajectory Combined Prediction Model Based on CC-LSTM)

There will be a complex driving environment formed by manned and unmanned vehicles with highly uncertain and dynamic interaction when autonomous vehicles enter actual traffic flow. Autonomous vehicles need to detect and analyze the movement of surrounding vehicles to make safe driving decisions. This paper proposes a Clustering Convolution-LSTM (CC-LSTM) vehicle trajectory prediction model. Using Fuzzy Clustering method, similar trajectories of surrounding vehicles are clustered to mine the temporal features of vehicles trajectory. The result is carried out to classify the features of historical trajectory by Density Clustering and find the similarity during segments which are used as the spatial features of target vehicles trajectory. The screened spatio-temporal features are fused by Las Vegas Wrapper (LVW) algorithm to obtain new input data of Convolution-LSTM network to make prediction. This model is trained by different features of data. Simulation shows that the CC-LSTM prediction model with surrounding vehicle interaction information and selected features can meet the real-time and accuracy requirements of prediction.

Automated summary

This paper proposes a CC-LSTM vehicle trajectory prediction model that combines clustering analysis and feature fusion, which can meet the real-time and accuracy requirements of autonomous vehicles in complex driving environments.