基于非线性观测器的传感器融合与旋转运动重构  

Sensor Fusion and Rotation Motion Reconstruction Based on Nonlinear Observer

作　　者：黄晓艳[1] HUNAG Xiao-yan(Chongqing Kechuang Vocational College,Chongqing 402160)

机构地区：[1]重庆科创职业学院,重庆402160

出　　处：《武汉职业技术学院学报》2018年第4期72-76,81,共6页Journal of Wuhan Polytechnic

摘　　要：虽然测量和记录人身体在空间中运动的加速度和角速度非常容易,但是从这些旋转位移测量中重建并不是一个简单的任务。为了定义位移量这些量需要集成,信号中存在的噪音对测量结果有明显的影响,限制了该技术在工业应用技术领域的发展,常用的噪声滤波技术(如卡尔曼滤波)对约束非线性运动学问题是一个很大的挑战。通过对刚体假设下的问题给出一个简洁的表述,探讨了非线性状态观测器在测量数据的处理、人体运动的数据融合和重建等方面的应用。在扩展卡尔曼滤波器技术和所提出的方法之间进行比较,需要特别注意影响这两种方法性能的条件。为了更好地说明方法之间的差异,本文对数值实验的结果进行了比较。Although it is very easy to measure and record the acceleration and angular velocity of a person^body in space, reconstruction from these rotational displacement measurements is not a simple task. In order todefine the amount of displacement, these quantities need to be integrated. The noise present in the signal has asignificant influence on the measurement results, which limits the development of the technology in the field ofindustrial application technology. Noise filtering techniques （such as Kalman filter） are a big challenge forconstrained nonlinear kinematics. This paper aims to elaborate on the topic, by providing a concise formulation tothe problem under rigid-body body assumptions and explore the use of nonlinear state-estimators to address theconditioning of the measured data, data fusion and reconstruction of the body motion. A comparison is drawnbetween an extended linear approach （EKF） and the proposed methodology, paying particular attention to theconditions that affect the performance of both methodologies. The paper compares results from numericalexperiments using to better illustrate the differences between methodologies.

关 键 词：传感器融合 旋转运动重建 状态观测器 扩展卡尔曼滤波器 滑模观测器 

分 类 号：G804.49[文化科学—运动人体科学]