Low-altitude unmanned aerial vehicle logistics path risk intelligent management and control system

The invention relates to the technical field of unmanned aerial vehicle path planning, and discloses a low-altitude unmanned aerial vehicle logistics path risk intelligent management and control system, which comprises an initialization stage and an environment interaction training stage, and is characterized in that the initialization stage comprises a state initialization module used for initializing an unmanned aerial vehicle distribution environment and an interruption scene; the environment interaction training stage comprises the following steps: an interaction data acquisition module performs environment interaction and acquires interaction data as a basis of a learning process of a reinforcement learning algorithm; the functions of three core algorithms adopted by the strategy selection module are mutually linked, tuple parameters are set to be obtained through network updating, a new state is entered, and the process of the environment interaction training stage is continuously repeated; and obtaining an optimal path in the interruption scene through multiple times of training. The Markov decision process is combined with low-altitude unmanned aerial vehicle logistics path planning under the interruption situation, the real environment can be reflected more dynamically and accurately, and the method is suitable for diversified unmanned aerial vehicle distribution scenes.