Research On Cooperative Positioning Navigation Method For Autonomous Underwater Vehicle

In recent years,with the rapid development of autonomous underwater vehicles in marine applications,collaborative positioning and navigation of multi-submersible robot systems has become a research hotspot.When long-haul,large-scale surveys or remote operations are required,superiority to the maneuverability and time-varying characteristics of the marine environment have led to the inability to apply traditional positioning methods.The emerging co-location technology is to solve various underwater submersibles.Positioning emerged.Cooperative positioning has gradually become the main research direction of underwater AUV clusters through the observational information between each AUV and the team's co-location.This paper analyzes and studies the existing co-location algorithms and experimental cases.The existing methods have problems such as the inability to simultaneously perform multiple AUV localization and the complexity of the algorithm with increasing number of AUVs.The covariance algorithm for the special variance estimation uses the distance between any two AUVs and the AUV's own velocity information as observations,and adds the AUV's own position error as an observation measurement to the adjustment equation,so that within a co-location period,All AUVs within the AUV team can correct their position and make their positioning accuracy consistent with the AUV standard with the smallest positional error in the team.The simulation of static co-location shows that the increase in the number of AUVs reduces the positioning error.When AUVs are paired,their position error is smaller than that in the AUV.The influences of random ranging errors,fixed ranging errors,dead reckoning errors and AUV dead reckoning accuracy are analyzed respectively.It is found that the fixed ranging error has obvious influence on the positioning error,and the two have a linear positive correlation.Random ranging error has little effect on positioning accuracy.The positioning error caused by dead reckoning error is much less than the positioning error caused by fixed ranging error.Therefore,it is concluded that ranging error is the main factor affecting positioning error.Moreover,the distribution area of the positioning accuracy is circular like the center of the main AUV diffusing outward,and the farther the distance from the main AUV is,the worse the positioning accuracy is.For the dynamic co-location,the position error of the linear track,V-shaped track and comb track of four auxiliary AUVs and eight AUVs under the single main AUV piloting condition was simulated and analyzed.It is proved that the AUV can be Co located at the same time,and with the increase of AUV number,the effect of algorithm is not decreasing.It can be seen that the comb track and the V-shaped track can correct the position error of the main AUV by the track,so that the positioning error of the entire AUV team is reduced.The method proposed in this paper can provide theoretical support for the field of co-location navigation,provide help for future engineering applications,and has a wide range of application scenarios.