High-speed Unmanned Underwater Vehicle (UUV) needs precise target information to complete the mission, but the target information is not always continuous or with high-level accuracy. To promote the efficiency of underwater attack missions, different from traditional fixed-graph searching trajectory or waypoint path planning, this paper proposes a target distribution based fast searching trajectory planning method to provide high-speed UUV with both fast planning ability and intelligent searching ability. First, the characteristic of the moving target is modeled into its position distribution with respect to time, which provides a feedback signal to the fast planning method. A parameterized graph searching trajectory is developed and optimized by particle swarm optimization, which maintains the expert knowledge in particular graph searching trajectory and the lightweighted intelligent optimization in a small number of parameters. Finally, the experiment on the target search mission with only initial target information validates the effectiveness of this planning scheme.
Discussion(0)
No comments yet. Be the first to comment.