Target search method for autonomous underwater vehicle system oriented to marine static targets

Abstract

The present invention proposes an autonomous underwater vehicle system target search method for ocean static targets, which is used for collaborative planning of multiple underwater vehicle search systems, including the following steps: based on the search history data of the task area, generating a prior The prior search probability map; extract the search sub-regions with higher search probability in the prior search probability map, and distribute them to multiple underwater vehicles; perform grade evaluation on the multiple sub-regions assigned to each underwater vehicle, according to the evaluation The level is high or low, and the search order of each sub-area for each underwater vehicle is determined; each underwater vehicle uses an adaptive helical trajectory to plan the search spiral path for each sub-area according to the search order, and the sub-areas of each sub-area The elliptical spiral coverage route is continuously connected with the Dubins transition route to obtain a complete search path. The invention proposes a four-layer search and solution framework, which can improve the search efficiency of the underwater vehicle.

Description

technical field [0001] The invention relates to the technical field of underwater vehicle control, in particular to a target search method for an autonomous underwater vehicle system facing ocean static targets. Background technique [0002] With the continuous exploration of the deep sea, various underwater accidents occur frequently. As an important tool for offshore operations, autonomous underwater vehicles (UAVs) are often used to perform accident target search tasks. Generally speaking, target search is the process of covering the relevant area to the maximum extent through the sensors carried by the underwater vehicle (such as side-scan sonar, camera, etc.). Searching for targets in a large area of ​​sea is like finding a needle in a haystack, which is very technically difficult. Compared with a single underwater vehicle, the competition and coordination mechanism of multiple underwater vehicles will often improve the search efficiency, but the underwater weak communi...

AI Technical Summary

Problems solved by technology

In order to solve the task allocation problem of multiple underwater vehicles, the traditional auction algorithm and various auction algorithms improved on the basis of it came into being, but the allocation process through the auction mechanism is more complicated; in addition, various biologically inspired neural networks The model has also been applied to the task allocation of multiple underwater vehicles, but this type of method is still a greedy strategy in nature, so the optimality of the final allocation results cannot be guaranteed

Images