Single unmanned aerial vehicle autonomous motion planning method based on imitation learning

The invention provides a single unmanned aerial vehicle autonomous motion planning method based on imitation learning. A simulation environment is constructed, and a corresponding virtual unmanned aerial vehicle model is constructed based on a real unmanned aerial vehicle in the simulation environment, and the method comprises the following steps: S1, obtaining a training sample for trajectory planning of the virtual unmanned aerial vehicle based on the simulation environment; s2, training a neural network of the real unmanned aerial vehicle through the training sample; s3, collecting real-time environment sensing data and state data of the real unmanned aerial vehicle through a sensor of the real unmanned aerial vehicle; s4, generating a real-time prediction track point sequence based on the real-time environment sensing data and the real unmanned aerial vehicle state data by adopting a neural network in combination with the target point data; s5, converting the real-time prediction track point sequence into