Global and Asymptotically Efficient Localization from Range Measurements

作     者：Zeng, Guangyang Mu, Biqiang Chen, Jiming Shi, Zhiguo Wu, Junfeng 

作者机构：The College of Control Science and Engineering The State Key Laboratory of Industrial Control Technology Zhejiang University Hangzhou310027 China Key Laboratory of Systems and Control Institute of Systems Science Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China The College of Information Science and Electronic Engineering Zhejiang University Hangzhou310027 China The School of Data Science Chinese University of Hong Kong Shenzhen Shenzhen China The State Key Laboratory of Industrial Control Technology Zhejiang University Hangzhou China 

出 版 物：《arXiv》 (arXiv)

年 卷 期：2022年

核心收录：

主　　题：Spurious signal noise 

摘      要：We consider the range-based localization problem, which involves estimating an object s position by using m sensors, hoping that as the number m of sensors increases, the estimate converges to the true position with the minimum variance. We show that under some conditions on the sensor deployment and measurement noises, the LS estimator is strongly consistent and asymptotically normal. However, the LS problem is nonsmooth and nonconvex, and therefore hard to solve. We then devise realizable estimators that possess the same asymptotic properties as the LS one. These estimators are based on a two-step estimation architecture, which says that any √m-consistent estimate followed by a one-step Gauss-Newton iteration can yield a solution that possesses the same asymptotic property as the LS one. The keypoint of the two-step scheme is to construct a √mconsistent estimate in the first step. In terms of whether the variance of measurement noises is known or not, we propose the Bias-Eli estimator (which involves solving a generalized trust region subproblem) and the Noise-Est estimator (which is obtained by solving a convex problem), respectively. Both of them are proved to be √m-consistent. Moreover, we show that by discarding the constraints in the above two optimization problems, the resulting closed-form estimators (called Bias-Eli-Lin and Noise-Est-Lin) are also √m-consistent. Plenty of simulations verify the correctness of our theoretical claims, showing that the proposed two-step estimators can asymptotically achieve the Cramer-Rao lower bound. Copyright © 2022, The Authors. All rights reserved.