A Robust Optimization Method for Underwater Vehicle Positioning Based on Switch Constraints

Abstract

The invention relates to an underwater vehicle positioning robust optimization method based on switch constraint. For the defect that a conventional underwater vehicle navigation control method cannotachieve robustness to an error closed loop, a navigation control method based on General Graph Optimization (g2o) and switch constraint is disclosed, a switch constraint back-end optimization methodis adopted, a node factor graph is adopted, and a method of eliminating an error closed loop generated in a map construction process is designed. The limitation that an existing underwater vehicle generates an error closed loop in the processes of synchronous positioning and map construction to cause errors of other constructed maps is solved, thus the vehicle processes the error closed loop in the map construction process and the error closed loop is weighted to be zero, and switch prior constraint is added to eliminate adverse effects brought by the error closed loop. The theory proves the robustness of the method, an experiment verifies the effectiveness of the method, and the method has a good effect in reducing an error rate of underwater vehicle map construction.

Description

technical field [0001] The invention belongs to the field of underwater vehicle navigation control and relates to a robust optimization method for underwater vehicle positioning based on switch constraints. Background technique [0002] Since the underwater vehicle does not have the support of the GPS signal, it needs to continuously form a closed loop through its own path to eliminate the odometer error and optimize the map construction and its own positioning. The traditional optimization method is to use the least squares optimization method, including the incremental smoothing and mapping method (iSAM) and the consistent attitude estimation method, that is, the underwater vehicle solves the problem by combining the visual tree method with the Bayesian probability framework. This problem, thereby reducing location recognition to a recursive Bayesian estimation problem, or detection of potential closed loops with stereo cameras, and filtering of false loop closures based o...

AI Technical Summary

Problems solved by technology

[0003] From the above analysis process, it can be seen that the traditional underwater vehicle navigation control technology system can guarantee its robustness in an ideal environment, but it cannot realize the processing ability of the error closed loop, so it is very easy to cause map construction errors

Images