不依赖于精确初始坐标的车联网相对定位坐标估计算法

发布时间：2018-08-02 10:45

【摘要】：在车联网定位中,GPS(Global Positioning System)信号长时间较差甚至中断会导致GPS定位结果不可靠甚至不可用,无法为相对定位算法提供可靠的精确初始坐标.针对这一问题,该文对车辆相对位置坐标估计方法展开研究,结合TOA(Time of Arrival)测距技术,将相对位置坐标估计问题转化为非线性规划问题.为减小非线性规划问题中初始坐标对算法结果的影响,将外部罚函数法与Powell算法结合,利用外部罚函数法"能够从非可行解出发逐步逼近可行域"的特点优化最优化方法,解决算法对初始坐标的敏感特性;利用Powell算法能够"逼近局部最优解"的特点作为最优化求解方法,用于求解目标函数最优解.提出一种不依赖于精确初始坐标的相对定位(Exact Initial Coordinate Free Relative Localization,EICFRL)算法,实现车联网高精度相对定位.在算法验证中,该文采用两种TOA节点部署方案,分别为单点部署方案和基于几何约束的多点部署方案.在多点部署方案中,利用车辆固有形状属性,形成基于车型的几何约束,增加非线性规划问题可行域限制.为验证该文算法可行性及有效性,该文在仿真实验中设置不同测距误差、连通性、车辆数目等条件,并在实际环境中实验验证,将该算法结果与Powell算法、LM(Levenberg-Marquard)算法、CRLB(Cramer-Rao Lower Bound)进行对比.实验结果显示,该文算法定位精度提高超过50%.当使用多点部署方案时,算法定位误差进一步减小约为30%(仿真环境)和23%(实测环境).
[Abstract]:Long time poor or even interrupted (Global Positioning System) signals in networked vehicle positioning may lead to unreliable or even unusable GPS positioning results, which can not provide reliable initial coordinates for relative positioning algorithms. In order to solve this problem, this paper studies the method of vehicle relative position coordinate estimation. Combining with TOA (Time of Arrival) ranging technology, the problem of relative position coordinate estimation is transformed into a nonlinear programming problem. In order to reduce the influence of initial coordinates on the results of nonlinear programming problems, the external penalty function method is combined with the Powell algorithm to optimize the optimization method using the characteristic that the external penalty function method "can approach the feasible region step by step from the infeasible solution". The sensitivity of the algorithm to the initial coordinates is solved, and the Powell algorithm is used to solve the optimal solution of the objective function by making use of the characteristic that the local optimal solution can be approximated by the Powell algorithm. An algorithm of relative positioning (Exact Initial Coordinate Free Relative localization EICFRL, which does not depend on the exact initial coordinates, is proposed to realize the high precision relative positioning in the vehicle network. In the algorithm verification, two kinds of TOA node deployment schemes are adopted, one is single-point deployment scheme and the other is multi-point deployment scheme based on geometric constraints. In the multi-point deployment scheme, the inherent shape attributes of the vehicle are used to form the geometric constraints based on the vehicle model, which increases the limitation of the feasible region of the nonlinear programming problem. In order to verify the feasibility and effectiveness of the proposed algorithm, different ranging errors, connectivity and vehicle number are set up in the simulation experiment. The results of the algorithm are compared with that of the Powell algorithm, LM (Levenberg-Marquard) algorithm (Cramer-Rao Lower Bound). The experimental results show that the accuracy of the algorithm is improved by more than 50%. When the multi-point deployment scheme is used, the location error of the algorithm is further reduced to about 30% (simulation environment) and 23% (measured environment).
【作者单位】：北京科技大学计算机与通信工程学院;北京科技大学融合网络与泛在业务工程技术研究中心;
【基金】：国家自然科学基金(61304257) 北京市自然科学基金(4152036) 中央高校基本科研业务费(FRF-TP-15-026A2)资助北京科技大学与台北科技大学学术合作专题研究计划经费辅助~~
【分类号】：U495

【参考文献】