Lane changing planning control method for autonomous vehicle in dense traffic scene

The invention discloses a lane changing planning control method for an automatic driving vehicle in a dense traffic scene, and the method comprises the steps: taking environment uncertainty, vehicle safety constraint and vehicle interaction heterogeneity into consideration by an interactive reinforcement learning module, and generating a target track of interactive lane changing of the automatic driving vehicle in the dense traffic scene based on a reinforcement learning algorithm; by introducing prior driving knowledge, an unsafe motion planning strategy generated by the interactive reinforcement learning module is identified and adjusted, and the corrected motion planning strategy is transmitted to the motion tracking control module; a local optimal control command is generated based on a nonlinear model predictive control algorithm, and a target trajectory provided by the interactive reinforcement learning module is tracked, so that vehicle dynamics constraints are met, and potential risks can be predicted