基于支持向量机分类决策的行人导航零速修正方法
|
摘要
为了减小行人导航过程中的误差,提出基于支持向量机分类决策的零速反馈修正方法。根据行人足部运动特点,构建行人足部运动模型,利用支持向量机决策方法对足部运动样本数据进行训练和提取数据特征,建立超平面方程。通过超平面函数对行人足部运动数据进行分类和决策,辨别区分静止段和运动段。在零速静止段,对惯性导航解算的速度、角速度和方向进行修正,利用扩展卡尔曼滤波递推方法进行方向、速度和位置误差跟踪。进行了行人按既定路径的行走跟踪实验,结果表明,设计的行人导航系统能够使行人行走轨迹与设定路径完全吻合,多次测试数据最大误差小于2. 5%,平均误差为1. 94%。因此,基于支持向量机分类决策的行人导航零速修正方法能够准确地对行人轨迹进行跟踪和定位。
In order to reduce the error in the pedestrian navigation process,a zero-speed feedback correction method based on support vector machine classification decision is proposed. According to the characteristics of pedestrian foot movement,the pedestrian foot motion model is constructed. The support vector machine decision method is used to train the foot motion sample data and extract the data features to construct the hyperplane equation.The classification and decision of pedestrian foot motion data are classified by the hyperplane function,and the stationary segment and the motion segment are distinguished. In the zero-speed stationary section,the speed,angular velocity and direction of the inertial navigation solution are corrected,and the extended Kalman filter recursive method is used to track the direction,velocity and position error. The walking tracking experiment of pedestrians according to the established path is carried out. The results show that the pedestrian navigation system can make the pedestrian walking trajectory completely match the set path. The maximum error of multiple test data is less than 2. 5%,and the average error is 1. 94%. Therefore,the pedestrian navigation zero-speed correction method based on support vector machine classification decision can accurately track and locate the pedestrian trajectory.
In order to reduce the error in the pedestrian navigation process,a zero-speed feedback correction method based on support vector machine classification decision is proposed. According to the characteristics of pedestrian foot movement,the pedestrian foot motion model is constructed. The support vector machine decision method is used to train the foot motion sample data and extract the data features to construct the hyperplane equation.The classification and decision of pedestrian foot motion data are classified by the hyperplane function,and the stationary segment and the motion segment are distinguished. In the zero-speed stationary section,the speed,angular velocity and direction of the inertial navigation solution are corrected,and the extended Kalman filter recursive method is used to track the direction,velocity and position error. The walking tracking experiment of pedestrians according to the established path is carried out. The results show that the pedestrian navigation system can make the pedestrian walking trajectory completely match the set path. The maximum error of multiple test data is less than 2. 5%,and the average error is 1. 94%. Therefore,the pedestrian navigation zero-speed correction method based on support vector machine classification decision can accurately track and locate the pedestrian trajectory.
引文
1 Saska M,Baca T,Thomas J,et al. System for deployment of groups of unmanned micro aerial vehicles in GPS-denied environments using onboard visual relative localization[J]. Autonomous Robots,2017,41(4):919-944
2 Yang T,Li G,Li J,et al. A ground-based near infrared camera array system for UAV auto-landing in GPS-denied environment[J]. Sensors,2016,1393(16):1-20
3 Guo K,Qiu Z,Meng W,et al. Ultra-wideband based cooperative relative localization algorithm and experiments for multiple unmanned aerial vehicles in GPS denied environments[J]. International Journal of Micro Air Vehicles,2017,9(3):169-186
4 Han H,Wang J,Wang J,et al. Performance analysis on carrier phase-based tightly-coupled GPS/BDS/INS integration in GNSS degraded and denied environments[J]. Sensors,2015,15(4):8685-8711
5 Shen C,Bai Z,Cao H,et al. Optical flow sensor/INS/magnetometer integrated navigation system for MAV in GPS-denied environment[J]. Journal of Sensors,2016(11),6105803
6 王亚娜,蔡成林,李思民,等.基于行人航迹推算的室内定位算法研究[J].电子技术应用,2017,43(4):86-89Wang Yana,Cai Chenglin,Li Simin,et al. Research on indoor location algorithm based on pedestrian track estimation[J]. Electronic Technology Applications,2017,43(4):86-89
7 Tian X,Chen J,Han Y,et al. Pedestrian navigation system using MEMS sensors for heading drift and altitude error correction[J]. Sensor Review,2017,37(3):270-281
8 Chang L,Li J,Chen S. Initial alignment by attitude estimation for strapdown inertial navigation systems[J]. IEEE Transactions on Instrumentation and Measurement,2015,64(3):784-794
9 韩勇强,田晓春.基于足部安装MIMU的行人导航系统设计[J].导航定位与授时,2018,5(1):22-27Han Yongqiang,Tian Xiaochun. Design of pedestrian navigation system based on MIMU installed on foot[J]. Navigation Positioning and Timing,2018,5(1):22-27
10 Wang Zhelong,Zhao Hongyu,Qiu Sen,et al. Stance-phase detection for ZUPT-aided foot-mounted pedestrian navigation system[J].IEEE/ASME Transactions on Mechatronics,2015,20(6):3170-3181
11 Li Xiaofang,Mao Yuliang,Xie Ling,et al. Applications of zero-velocity detector and Kalman filter in zero velocity update for inertial navigation system[C]//2014 IEEE Chinese Guidance,Navigation and Control Conference(CGNCC). New York:IEEE,2014:1760-1763
12 崔潇,秦永元,周琪,等.鞋式个人导航系统算法和试验研究[J].测控技术,2013,32(3):138-142Cui Xiao,Qin Yongyuan,Zhou Qi,et al. Algorithm and test research on shoe-mounted personal navigation system[J]. Measurement and Control Technology,2013,32(3):138-142
13 樊启高,孙艳,孙璧文,等.基于GLRT零速检测算法的行人室内定位系统[J].传感技术学报,2017,30(11):1706-1711Fan Qigao,Sun Yan,Sun Biwen,et al. Pedestrian indoor positioning system based on GLRT zero velocity detection algorithm[J].Acta Sensing Technology,2017,30(11):1706-1711
14 汪海燕,黎建辉,杨风雷.支持向量机理论及算法研究综述[J].计算机应用研究,2014,31(5):1281-1286Wang Haiyan,Li Jianhui,Yang Fenglei. Review of support vector machine theory and algorithms[J]. Computer Applications,2014,31(5):1281-1286
15 Jiménez A R,Seco F,Prieto J C,et al. Indoor pedestrian navigation using an INS/EKF framework for yaw drift reduction and a footmounted IMU[C]//2010 7th Workshop on Positioning Navigation and Communication(WPNC). New York:IEEE,2010:135-143
2 Yang T,Li G,Li J,et al. A ground-based near infrared camera array system for UAV auto-landing in GPS-denied environment[J]. Sensors,2016,1393(16):1-20
3 Guo K,Qiu Z,Meng W,et al. Ultra-wideband based cooperative relative localization algorithm and experiments for multiple unmanned aerial vehicles in GPS denied environments[J]. International Journal of Micro Air Vehicles,2017,9(3):169-186
4 Han H,Wang J,Wang J,et al. Performance analysis on carrier phase-based tightly-coupled GPS/BDS/INS integration in GNSS degraded and denied environments[J]. Sensors,2015,15(4):8685-8711
5 Shen C,Bai Z,Cao H,et al. Optical flow sensor/INS/magnetometer integrated navigation system for MAV in GPS-denied environment[J]. Journal of Sensors,2016(11),6105803
6 王亚娜,蔡成林,李思民,等.基于行人航迹推算的室内定位算法研究[J].电子技术应用,2017,43(4):86-89Wang Yana,Cai Chenglin,Li Simin,et al. Research on indoor location algorithm based on pedestrian track estimation[J]. Electronic Technology Applications,2017,43(4):86-89
7 Tian X,Chen J,Han Y,et al. Pedestrian navigation system using MEMS sensors for heading drift and altitude error correction[J]. Sensor Review,2017,37(3):270-281
8 Chang L,Li J,Chen S. Initial alignment by attitude estimation for strapdown inertial navigation systems[J]. IEEE Transactions on Instrumentation and Measurement,2015,64(3):784-794
9 韩勇强,田晓春.基于足部安装MIMU的行人导航系统设计[J].导航定位与授时,2018,5(1):22-27Han Yongqiang,Tian Xiaochun. Design of pedestrian navigation system based on MIMU installed on foot[J]. Navigation Positioning and Timing,2018,5(1):22-27
10 Wang Zhelong,Zhao Hongyu,Qiu Sen,et al. Stance-phase detection for ZUPT-aided foot-mounted pedestrian navigation system[J].IEEE/ASME Transactions on Mechatronics,2015,20(6):3170-3181
11 Li Xiaofang,Mao Yuliang,Xie Ling,et al. Applications of zero-velocity detector and Kalman filter in zero velocity update for inertial navigation system[C]//2014 IEEE Chinese Guidance,Navigation and Control Conference(CGNCC). New York:IEEE,2014:1760-1763
12 崔潇,秦永元,周琪,等.鞋式个人导航系统算法和试验研究[J].测控技术,2013,32(3):138-142Cui Xiao,Qin Yongyuan,Zhou Qi,et al. Algorithm and test research on shoe-mounted personal navigation system[J]. Measurement and Control Technology,2013,32(3):138-142
13 樊启高,孙艳,孙璧文,等.基于GLRT零速检测算法的行人室内定位系统[J].传感技术学报,2017,30(11):1706-1711Fan Qigao,Sun Yan,Sun Biwen,et al. Pedestrian indoor positioning system based on GLRT zero velocity detection algorithm[J].Acta Sensing Technology,2017,30(11):1706-1711
14 汪海燕,黎建辉,杨风雷.支持向量机理论及算法研究综述[J].计算机应用研究,2014,31(5):1281-1286Wang Haiyan,Li Jianhui,Yang Fenglei. Review of support vector machine theory and algorithms[J]. Computer Applications,2014,31(5):1281-1286
15 Jiménez A R,Seco F,Prieto J C,et al. Indoor pedestrian navigation using an INS/EKF framework for yaw drift reduction and a footmounted IMU[C]//2010 7th Workshop on Positioning Navigation and Communication(WPNC). New York:IEEE,2010:135-143