Research On Gravity Aided Inertial Navigation Matching Algorithm

Gravity-assisted inertial navigation technology makes use of the spatial distribution characteristics of the ocean gravity field to modify underwater navigation,which is dominated by inertial navigation,to achieve high-precision navigation with high autonomy,high concealment,passivity,anti-interference and other characteristics.The gravity-matched inertial navigation system is divided into three parts.It includes the output subsystem of the inertial navigation system's position,distance and course information;the subsystem of ocean gravity field anomaly map and the real-time acquisition of gravity data during navigation;the fusion matching subsystem of gravity anomaly map and inertial data.The matching algorithm is a key technology that needs urgent research.At present,the actual application of matching algorithm is still based on the correlation extreme value algorithm based on the correlation analysis,the Iterative Closet Contour Point algorithm,and the Sandia Inertia Terrain-aided navigation algorithm.Those algorithms have both high precision and timeliness of matching under specific conditions,but the adaptability of the gravity anomalies background field,and it is impossible to achieve a largescale gravity-matched inertial navigation.This article combines the existing traditional algorithms,analyzes its algorithm principle,advantages and disadvantages.Compare the adaptability of different algorithms in the background of multiple ocean gravity fields,which proposes a new method that combines two feature algorithm to complementary.The new method improves the precisions of Terrain Contour Matching algorithm.Researching the influence factors of the mismatch in the probabilistic neural network algorithm in the implementation of the gravity/inertial integrated navigation system.Theoretically discussed the high-precision features in the short time with the help of inertial navigation,and increased restrictions to come true the feasibility of improving the probabilistic neural network algorithm.This paper introduces the particle swarm optimization algorithm evolved from the bird swarm prey behavior.The idea of finding optimal solution through collaboration and information sharing among individuals in the group.A new submarine gravity aided inertial matching navigation algorithm based on particle swarm optimization is proposed.The main research results include: 1.The gravity matching navigation algorithm of TERCOM and ICCP fusion is studied.Analysis of the applicable characteristics of the two algorithms for inertial data and gravity data,put forward a new matching navigation based on the combination of the two.Based on the existing improved algorithm,the algorithm performs a joint comparative analysis of the sampling interval distance provided by inertial navigation,the ICCP matching distance and the improved TERCOM matching distance,sets the limit difference,searches for the larger TERCOM matching error,and ICCP replaces corresponding matching points,repeatedly searches and replaces,and finally generates a new matching trajectory combining the two algorithms.2.The characteristics of the probabilistic neural network algorithm are analyzed.By calculating the similarity probability between the trajectory to be matched and the measured gravity trajectory,the maximum probability of the same kind is selected for gravity matching navigation.Based on this,a new gravity-matching navigation algorithm based on correlation constraints is proposed by using the characteristics of inertial trajectories that can provide high-precision heading as well as distance in a short period of time to restrain the pending trajectories of probabilistic neural networks.Based on the existing algorithm,the algorithm adds trajectory orientation and distance information based on the inertial navigation system,which effectively exclude a large number of interference trajectories to be matched,improves matching efficiency,and improves matching accuracy.3.In view of the current demand for high matching positioning accuracy in gravitymatched inertial navigation,this paper introduces the particle swarm optimization algorithm evolved from the bird predation behavior,and seeks the optimal solution through the collaboration and information sharing between the individuals in the group.A new algorithm of submarine gravity-assisted inertial matching navigation based on particle swarm optimization is proposed.This study improves the particle swarm optimization algorithm to make it suitable for gravity-matched navigation,conducts algorithm implementation and simulation analysis,effectively improved the accuracy of gravity matching navigation,and analyzes that the algorithm is less affected by the resolution of the matched gravity field background map,more affected by the accuracy of the background field gravity value,and more affected by the observation gravimeter.4.The realization of gravity matching inertial navigation put forward higher requirement on the algorithm and matching environment.How to effectively analyze the relationship between the two has important implications for matching navigation.Using TERCOM algorithm,Probabilistic neural network algorithm and ICCP algorithm,the gravity-matched inertial navigation is carried out under different gravity background field environments.Through moving local window calculation method on the gravity field trajectory to analyze various gravity field feature judgment formulas,and select the local gravity field standard deviation and latitude and longitude direction correlation coefficient as the quantitative indicators of matching region judgement,and simulate and compare the matching ability of the three algorithms in different environments.