Multi-UUV/USV Adaptive Cooperative Search Using Online State Information.

Unmanned Surface Vessel (USV) and Unmanned Underwater Vehicle (UUV) detect the same task plane while located in different planes, and have to avoid obstacles autonomously in cooperative search in unknown environments. This paper proposes an adaptive cooperative search method for a USV/UUV heterogeneous cross-domain swarm based on online state information. Considering the different planes of USV/UUV detection, we establish a USV/UUV detection model with a forward-looking sonar according to the cross-sectional view. In order to guarantee the safety of USV/UUV, an obstacle avoidance algorithm based on predictive guidance is designed. Finally, the task environment is rasterized to establish the state information structure of grids and USV/UUV. On this basis, USV/UUV adaptively plans the search path according to its own motion state. The state information of each grid is updated online until USV/UUV swarm completes the search task. Simulation results show that USV/UUV can accomplish cooperative search safely in an unknown environment. Compared with a UUV homogeneous swarm, the USV/UUV swarm shortens the search time and improves the cooperative search efficiency.