SPEED PLANNING METHOD DEVICE EQUIPMENT MEDIUM AND VEHICLE FOR AUTONOMOUS DRIVING

摘要

The present invention discloses a speed planning method, apparatus, equipment and medium for autonomous driving, and relates to artificial intelligence technology such as autonomous driving and deep learning. A specific implementation method is to obtain a current state of the vehicle, wherein the current state includes at least a remaining mileage and a current speed, and interpolates an operation according to the current state and a predetermined state table, so that the target operation of the current state is , wherein the state table is determined based on the learning reinforcement method, and includes a plurality of states and an operation executed in each state, wherein the operation includes at least an acceleration. The present invention replaces the neural network by using the lookup table method, first determines the state table offline, and then determines the target operation of the current state through the online inquiry method, a premise that does not occupy a large amount of computing resources Thus, it can be ensured to obtain a global optimal solution of the velocity plan.