Deep-learning-based automated driving vehicle dynamic trajectory planning and decision making

Aiming at the technical problem of dynamic scene trajectory planning and decision-making for self-driving cars in the increasingly complex traffic and road environments, we have applied adaptive from the aspects of intelligent control and emergency scene change, interactive predictive path planning, complex traffic scene perception, etc., we use adaptive cruise control, automatic driving motion planning algorithm, convolutional neural network and other technological methods to realize the safe and efficient driving of the self-driving car in the complex and changeable as well as unfamiliar dynamic scenes. The application of intelligent networked vehicles to collaboratively verify the multi-class algorithms and driving scenarios shows that it is possible to realize the safe navigation of vehicles in complex traffic environments.