Research On High Precision Positioning Technology Of Motion Platforms And Performance Analysis

In recent years,motion platforms such as drones and unmanned vehicles have been widely used in agriculture,industry,and even national defense.In application practice,the location and motion information of the platform itself are often indispensable.And the research of motion state estimation is also getting more and more attention.In the actual environment,the location information of the anchor node often has a certain error.How to obtain higher-precision positioning information under the influence of this error is a hot and difficult point of current research.In addition,how to use the observation information to estimate the position and velocity information of the moving platform at the same time is also an important research direction.Based on the assumption that the anchor node has errors,this thsis studies the positioning algorithm of the moving platform from two aspects: joint positioning of multiple nodes and estimation of the motion state of the moving array.The main research work of this thesis is as follows:(1)Based on the position error of the anchor node,a system model of multi-platform joint positioning is constructed,and three positioning algorithms are proposed based on the system model.Based on the traditional cooperative positioning model,this paper considers the position error of anchor nodes,and adds the ranging information between anchor nodes as prior information to estimate the location of blind nodes and correct the position of anchor nodes.Based on the above system model,this paper derives the Cramér-Rao Lower Bound(CRLB)for multi-platform co-location.Then,this paper proposes an iterative method and two convex relaxation algorithms based on Maximum Likelihood Estimate(MLE).First,this paper deduces the Newton iteration method for solving the MLE localization problem,which can provide high-precision localization estimation under good environmental conditions.Secondly,this paper uses the second-order cone relaxation and semi-definite relaxation algorithms respectively to transform the original MLE problem into a convex optimization problem,which can provide a good initial solution for the iterative method.Finally,this paper combines the Two Step Weighted Least Squares(TSWLS)algorithm considering the anchor node error and the two convex relaxation algorithms above with the iterative method to test the positioning effect in different environments.The simulation results show that the positioning algorithm proposed in this paper has higher positioning accuracy than the traditional TSWLS algorithm,and combining the TSWLS algorithm and the two convex relaxation algorithms with the iterative method can obtain the positioning results close to CRLB.Among them,the convex optimization problem after semi-positive definite relaxation can be solved to obtain the initial solution closest to the real value,and can reach CRLB through iteration,but the computational complexity is high and the operation time is the longest.(2)On the basis of the system model discussed above,a multi-time joint motion array state estimation system model is constructed,and three motion state estimation algorithms are proposed based on this model.In this thesis,the measurement information of the adjacent moments of the motion array is used to solve the simultaneous solution.Under the condition that the motion model is unknown and there is no motion information measurement method,a system model for estimating the position and velocity of each node of the motion array is provided.First,this thesis derives the CRLB for estimating position and velocity based on the system model.Then,based on the MLE problem,this thesis proposes an iterative method and two convex relaxation algorithms using second-order cone relaxation and positive semi-definite relaxation respectively.In the simulation test,the convex relaxation algorithm provides a good initial solution for the iterative method,and the estimation accuracy of the motion state is improved in the iteration.The simulation results show that,combining the convex relaxation algorithm and iterative algorithm proposed in this thesis,the position and velocity estimation effect of the motion array close to CRLB can be obtained in different environments.