Abstract: A backup processing method and a server are provided. The method is applied to a backup system, and the backup system includes a plurality of backup servers. The method includes: dividing, by a master backup server in the plurality of backup servers, a backup task into a plurality of child backup tasks; allocating, by the master backup server, each child backup task to each of the plurality of backup servers; and sending, by the master backup server, the plurality of child backup tasks to respective corresponding backup servers. According to this method, a data backup rate can be improved.

Abstract: The invention relates to a method and system for path planning of a wave glider, comprising acquiring historical navigation data of the glider and an underwater vehicle via a shore-based monitoring center; fitting historical navigation data nonlinearly by a deep learning neural network to obtain a trained network; acquiring real-time navigation data of the glider at an off-line end, real-time navigation data and predetermined shipping track data of the vehicle; obtaining the set of off-line optimized path planning schemes of the glider by the above data and the trained network; and determining an optimal path planning scheme of the glider by the deep learning neural network according to real-time data and constraint data of the glider at the on-line end. The invention can reasonably plan the path of the glider and ensure continuous and reliable information interaction between the glider and the vehicle.

Abstract: The invention relates to a method and system for path planning of a wave glider, comprising acquiring historical navigation data of the glider and an underwater vehicle via a shore-based monitoring center; fitting historical navigation data nonlinearly by a deep learning neural network to obtain a trained network; acquiring real-time navigation data of the glider at an off-line end, real-time navigation data and predetermined shipping track data of the vehicle; obtaining the set of off-line optimized path planning schemes of the glider by the above data and the trained network; and determining an optimal path planning scheme of the glider by the deep learning neural network according to real-time data and constraint data of the glider at the on-line end. The invention can reasonably plan the path of the glider and ensure continuous and reliable information interaction between the glider and the vehicle.