Opportunistic routing method for ocean underwater wireless sensing network based on reinforcement learning

The invention discloses an opportunistic routing method for an ocean underwater wireless sensing network based on reinforcement learning. The opportunistic routing method comprises the following steps: 1, randomly and uniformly deploying sensor nodes; 2, selecting a candidate forwarding set from a neighbor node set; 3, comprehensively taking various state information of the nodes into consideration by combining opportunistic routing and a reinforcement learning algorithm, and selecting a relay node in real time; 4, activating a recovery mechanism if no proper relay node exists in forwarding nodes; 5, setting a dynamic timer according to the priority of the node; and 6, judging whether the current node is in the communication range of the base station or not, and if so, directly communicating the node with the base station, or if not, continuing to repeat the step 2-4 to select the relay node until a data packet is transmitted to the base station. Aiming at the problem of a routing void, a corresponding recove