改进灰色模型预测跌倒解算算法
|
摘要
针对行人跌倒装置跌倒判断率低,成本高,便捷性不强等问题,提出一种改进灰色模型预测跌倒解算算法。算法利用迭代扩展卡尔曼滤波法(iteration extended Kalman filter,IEKF)将超宽带(ultra wide band,UWB)与惯导数据相融合得到实时综合位置,并通过迭代灰色MGM(1,1)模型得到预测综合位置,提高定位精度。将惯导装置获取的加速度、角速度和磁通量通过四元数运算得到姿态角,利用最小二乘法线性拟合得到倾角,同时融入综合位置进行三级跌倒判断,有效地提高了跌倒判断准确率。实验结果表明,采用改进算法可以较好地区分正常姿态与跌倒姿态,跌倒判断准确度高达97. 5%。
Aiming at the problems of low falling judgment rate,high cost and low convenience of pedestrian falling device,an improved grey model algorithm for predicting falling was proposed. Iterative extended Kalman filter method (IEKF) was used to combine the ultra wide band (UWB) with the inertia derivative to obtain the real-time integrated position. The iterative gray model was used to obtain the predicted comprehensive position. The position accuracy was improved. The attitude angle was obtained by quaternion calculation of acceleration. Angular velocity and magnetic flux was obtained by inertial navigation device. The inclination angle was obtained by linear fitting of least square method. At the same time,it was integrated into the comprehensive position for three-level fall judgment,which effectively improved the accuracy of fall judgment. Experimental results show that the improved algorithm can be used to distinguish the normal posture from the fall posture,and the fall judgment accuracy is up to 97. 5%.
Aiming at the problems of low falling judgment rate,high cost and low convenience of pedestrian falling device,an improved grey model algorithm for predicting falling was proposed. Iterative extended Kalman filter method (IEKF) was used to combine the ultra wide band (UWB) with the inertia derivative to obtain the real-time integrated position. The iterative gray model was used to obtain the predicted comprehensive position. The position accuracy was improved. The attitude angle was obtained by quaternion calculation of acceleration. Angular velocity and magnetic flux was obtained by inertial navigation device. The inclination angle was obtained by linear fitting of least square method. At the same time,it was integrated into the comprehensive position for three-level fall judgment,which effectively improved the accuracy of fall judgment. Experimental results show that the improved algorithm can be used to distinguish the normal posture from the fall posture,and the fall judgment accuracy is up to 97. 5%.
引文
1陈鹏,涂亚庆,童俊平,等.可穿戴式跌倒检测智能系统设计[J].传感器与微系统,2017,36(2):114-116Chen Peng,Tu Yaqing,Tong Junping,et al.Wearable intelligent fall detection system design[J].Transducer and Microsystem Technologies,2017,36(2):114-116
2张军建,赵捷,安佰京,等.基于三轴加速度传感器的跌倒检测研究[J].现代生物医学进展,2014,14(18):3585-3588Zhang Junjian,Zhao Jie,An Baijing,et al.Fall detection based on triaxial acceleration sensor[J].Progress in Modern Biomedicine,2014,14(18):3585-3588
3肖竹.超宽带定位与RAKE接收关键技术研究[D].西安:西安电子科技大学,2009Xiao Zhu.Research on the key technologies of UWB location and RAKE reception[D].Xi'an:Xidian University,2009
4张家田,王望,严正国.基于MEMS传感器的运动轨迹的追踪系统[J].现代电子技术,2016,39(12):117-119,122Zhang Jiatian,Wang Wang,Yan Zhengguo.MEMS sensor-based motion track tracking system[J].Modern Electronics Technique,2016,39(12):117-119,122
5刘铮.UKF算法及其改进算法的研究[D].长沙:中南大学,2009Liu Zheng.UKF algorithm and its improvement[D].Changsha:Central South University,2009
6霍宏伟,张宏科.基于室内无线传感器网络射频信号的老年人跌倒检测研究[J].电子学报,2011,39(1):195-200Huo Hongwei,Zhang Hongke.Study on fall detection of elderly people based on radio frequency signal of indoor wireless sensor network[J].Acta Electronica Sinica,2011,39(1):195-200
7李炳炳.基于MEMS惯性传感器的人体姿态检测系统的研究[D].南京:南京理工大学,2017Li Bingbing.Research on human body attitude detection system based on MEMS inertial sensor[D].Nanjing:Nanjing University of Science and Technology,2017
8刘星.多维MEMS惯性传感器的姿态解算算法研究[D].哈尔滨:哈尔滨工程大学,2013Liu Xing.Research on attitude calculation algorithm of multidimensional MEMS inertial sensor[D].Harbin:Harbin Engineering University,2013
9张浩,刘兴,崔学荣.基于单基站天线阵列的超宽带定位AOA估计方法[J].电子与信息学报,2013,35(8):2024-2028Zhang Hao,Liu Xing,Cui Xuerong.AOA estimation method for uwb location based on single base station an tenna array[J].Journal of Electronics Information Technology,2013,35(8):2024-2028
10黄帅.老年人跌倒检测系统中相关算法的研究及应用[D].北京:清华大学,2012Huang Shuai.Research and application of related algorithms in fall detection system for elderly people[D].Beijing:Tsinghua University,2012
11 Serkan T,Fikri E,Tanzer A H,et al.A hybrid algorithm for fall detection[J].Studies in Health Technology and Informatics,2017,237:163
12 Boubezoul A,EspiéS,Larnaudie B,et al.A simple fall detection algorithm for powered two wheelers[J].Control Engineering Practice,2013,21(3):286-297
2张军建,赵捷,安佰京,等.基于三轴加速度传感器的跌倒检测研究[J].现代生物医学进展,2014,14(18):3585-3588Zhang Junjian,Zhao Jie,An Baijing,et al.Fall detection based on triaxial acceleration sensor[J].Progress in Modern Biomedicine,2014,14(18):3585-3588
3肖竹.超宽带定位与RAKE接收关键技术研究[D].西安:西安电子科技大学,2009Xiao Zhu.Research on the key technologies of UWB location and RAKE reception[D].Xi'an:Xidian University,2009
4张家田,王望,严正国.基于MEMS传感器的运动轨迹的追踪系统[J].现代电子技术,2016,39(12):117-119,122Zhang Jiatian,Wang Wang,Yan Zhengguo.MEMS sensor-based motion track tracking system[J].Modern Electronics Technique,2016,39(12):117-119,122
5刘铮.UKF算法及其改进算法的研究[D].长沙:中南大学,2009Liu Zheng.UKF algorithm and its improvement[D].Changsha:Central South University,2009
6霍宏伟,张宏科.基于室内无线传感器网络射频信号的老年人跌倒检测研究[J].电子学报,2011,39(1):195-200Huo Hongwei,Zhang Hongke.Study on fall detection of elderly people based on radio frequency signal of indoor wireless sensor network[J].Acta Electronica Sinica,2011,39(1):195-200
7李炳炳.基于MEMS惯性传感器的人体姿态检测系统的研究[D].南京:南京理工大学,2017Li Bingbing.Research on human body attitude detection system based on MEMS inertial sensor[D].Nanjing:Nanjing University of Science and Technology,2017
8刘星.多维MEMS惯性传感器的姿态解算算法研究[D].哈尔滨:哈尔滨工程大学,2013Liu Xing.Research on attitude calculation algorithm of multidimensional MEMS inertial sensor[D].Harbin:Harbin Engineering University,2013
9张浩,刘兴,崔学荣.基于单基站天线阵列的超宽带定位AOA估计方法[J].电子与信息学报,2013,35(8):2024-2028Zhang Hao,Liu Xing,Cui Xuerong.AOA estimation method for uwb location based on single base station an tenna array[J].Journal of Electronics Information Technology,2013,35(8):2024-2028
10黄帅.老年人跌倒检测系统中相关算法的研究及应用[D].北京:清华大学,2012Huang Shuai.Research and application of related algorithms in fall detection system for elderly people[D].Beijing:Tsinghua University,2012
11 Serkan T,Fikri E,Tanzer A H,et al.A hybrid algorithm for fall detection[J].Studies in Health Technology and Informatics,2017,237:163
12 Boubezoul A,EspiéS,Larnaudie B,et al.A simple fall detection algorithm for powered two wheelers[J].Control Engineering Practice,2013,21(3):286-297