Path Fallowing Method Based On Low Cost INS And Large Curvature Variable Path For Unmanned Surface Vehicles

In complex marine environmental tasks,a complete navigation,guidance,and control system is an important guarantee for unmanned surface vehicles(USVs)to achieve path tracking.Especially in the absence of Global Positioning System(GPS)support,relying on onboard Inertial Navigation Systems(INS)to complete navigation becomes crucial.This article conducts relevant research on the tracking problem of an unmanned surface vehicle based on low-cost INS devices and paths with high accuracy and large curvature changes.The main research content is as follows:(1)For the application of a cost sensitive unmanned surface vehicle,especially in the event of GPS signal failure,achieving high-precision navigation based on low-cost INS is an important research direction.This thesis proposes a low-cost INS algorithm based on model and event triggering to solve the problem of inaccurate navigation information caused by the accumulation of navigation errors over time in low-cost INS devices.By introducing the thrust model and event triggering mechanism of unmanned surface vehicles,an event trigger based on the thrust model is constructed.The integrated weight coefficients of acceleration and angular velocity measured by INS are assigned based on different triggering events,resulting in relatively accurate velocity V and heading angle ψ,reducing the impact of external factors on INS.Finally,combined with the event trigger,the velocity Vm and heading angle ψm output from the dynamic model were obtained,as well as the improved Particle Filter(PF)algorithm,which mainly corrected the weight coefficients determining the velocity measurement value zV and heading angle measurement value zψ,thereby achieving better filtering performance.(2)Aiming at the problem of large crossing-track error ye when USV fallows a given path with large curvature change,a guidance subsystem based on ISGL is designed to realize USV’s accurate following of a given path with large curvature change.In the thesis,Stanley guidance law can be unified into one form with the LOS by theoretical derivation and proof.It is,essentially,a kind of LOS method with time-varying lookahead distance in the path tangential reference frame.And a compensation angle(km/(1+|ye|))Δαm is integrated to improve the SGL,which can evaluate the path curvature in advance and compensate the desired heading angle.The aim is to improve the heading angle error and make the USV turn in advance.(3)Finally,based on existing unmanned surface vehicle platforms,a motion control experiment was constructed in an actual water surface environment,verifying the effectiveness of the proposed low-cost INS navigation algorithm based on model and event triggering.The feasibility of the ISGL algorithm for tracking large curvature given paths was verified,as well as its superiority over standard SGL and LOS algorithms.