A Personalized Deep Learning Approach for Trajectory Prediction of Connected Vehicles 2020-01-0759
Forecasting the motion of the leading vehicle is a critical task for connected autonomous vehicles as it provides an efficient way to model the leading-following vehicle behavior and analysis the interactions. In this study, a personalized deep time-series modeling approach for human-like leading vehicle trajectory prediction considering different driving styles is proposed. The method enables a precise, personalized trajectory prediction for leading vehicles with limited inter-vehicle communication signals, such as the vehicle speed, acceleration, space headway, and time headway of the front vehicles. Based on the learning nature of human that human always tries to solve problems based on grouping and similar experience, three different driving styles are first recognized based on an unsupervised clustering with a Gaussian Mixture Model (GMM). The GMM generates a specific driving style for each vehicle based on the speed, acceleration, jerk, time, and space headway features of the leading vehicle Then, a personalized joint time-series modeling (JTSM) method based on the Long Short-Term Memory (LSTM) Recurrent Neural Network model (RNN) is proposed to predict the front vehicle trajectories. The JTSM contain a common LSTM layer and different fully connected regression layers for different driving styles. The proposed method is tested with the Next Generation Simulation (NGSIM) data on US101, and I-80. The JTSM is tested for making prediction one second ahead. Results indicate that the proposed personalized JTSM approach shows a significant advantage over the other algorithms.
Yang Xing, Chao Huang, Chen Lv, Yahui Liu, Hong Wang, Dongpu Cao
Nanyang Technological University, Tsinghua University, University of Waterloo