Decision Making and Trajectory Planning of Intelligent Vehicle’ s Lane-Changing Behavior on Highways under Multi-Objective Constrains

Discretionary lane changing is commonly seen in highway driving. Intelligent vehicles are expected to change lanes discretionarily for better driving experience and higher traffic efficiency. This study proposed to optimize the decision-making and trajectory-planning process so that intelligent vehicles made lane changes not only with driving safety taken into account, but also with the goal to improve driving comfort as well as to meet the driver’ s expectation. The mechanism of how various factors contribute to the driver’s intention to change lanes was studied by carrying out a series of driving simulation experiments, and a Lane-Changing Intention Generation (LCIG) model based on Bi-directional Long Short-Term Memory (Bi-LSTM) was proposed. The inputs of the Bi-LSTM were data fragments of several influencing factors including the relative velocity and the distance between the relative vehicles, the type of the preceding vehicles, and the average velocity of the adjacent traffic flows, that over a certain period of time, which was determined via examining subjects’ visual behaviors of the left view mirror or the right view mirror. By combining the LCIG model with a feasibility judgement model which was based on minimum safety spacing (MSS), a lane-changing decision-making model satisfying driving safety and drivers’ expectation was proposed. The model was trained with a part of trajectory dataset obtained from the simulation driving experiments. The jerk was taken into full consideration as boundary condition on the basis of seventh-degree polynomial trajectory planning. The proposed decision-making model were verified against a test dataset from the other parts of experimental data and the results show that the model resembles the lane-changing decision-making process of human drivers in real-world.