The opportunistic network is a wireless sensor network in which nodes have a strong motion.The dynamic change of the network topology makes the path between two nodes break frequently.Therefore the traditional routing algorithm can not be suitable.This paper analyzes the classic opportunistic networks routing algorithm Epidemic. Based on the analysis, it proposes a new algorithm ERSE combined with the sleeping mechanism which target is to reduce the energy consumption of nodes. ERSE makes three steps to judge the node whether to fall into the sleeping state. The first step is to make the node do not miss the current opportunity to communicate with others. Only the nodes who do not meet other nodes for a period of time will enter the sleeping judge process. The second step is to prevent the node with heavy forwarding task from falling into a state of death.It makes a threshold for nodes to judge itself whether to enter the mandatory sleeping. The last step is making the nodes predict future scenarios based on the previous motion so that the nodes can not miss the vast majority of the communication opportunities with others. The simulation results show that compared with the classical routing algorithms,the ERSE algorithm ensures that the network performance is not obviously changed while the energy consumption of the nodes is greatly reduced.