抄録
We propose a method to predict trajectories of surrounding vehicles considering individual drivingcharacteristics. Trajectory prediction of surrounding vehicles is attracting a lot of attention now, and it is expected to apply to advanced driver assistance systems. However, previous methods perform the trajectory prediction based on common driving patterns even though each driver shows a different driving characteristic. The proposed method focuses on the following behavior behind the preceding vehicle and estimates a driving characteristic of each driver using machine learning techniques. Based on the estimation result, the proposed method adjusts the prediction model and appropriately generates a trajectory. As the result, the performance of trajectory prediction can be dramatically improved.
