Pedestrian GNSS/PDR Integrated Navigation System with Graph Optimization

doi:10.12382/bgxb.2022.0557

Abstract

Abstract:

The integration of a navigation system based on the Global Navigation Satellite System (GNSS) and the Pedestrian Dead Reckoning (PDR) with inertial measurement data is a widely used and reliable navigation solution. To further improve the positioning accuracy of the GNSS/PDR integrated navigation system, we propose a GNSS/PDR integrated navigation method based on graph optimization. By constructing a factor graph to represent the probabilistic dependence between states and measurement information, all past states are iteratively estimated at each step as the unknowns, and the optimal estimation of the states is obtained by minimizing the global cost function. Compared with the KF algorithm, this new system can further reduce average positioning error and improve positioning accuracy. Results from two different real scene results show that, compared with KF, the average values of the horizontal positioning errors are reduced by more than 40%, verifying the algorithm’s effectiveness in improving the positioning accuracy.

Key words: pedestrian navigation, integrated navigation, global navigation satellite system, factor graph optimization

CLC Number:

TN96
TH89

ZHU Jianliang, WANG Liya, BO Yuming. Pedestrian GNSS/PDR Integrated Navigation System with Graph Optimization[J]. Acta Armamentarii, 2023, 44(10): 3137-3145.

Figures/Tables 15

Fig.1 Principle diagram of position updating

Fig.2 Principle diagram of pedestrian positioning based on smartphone function

Fig.3 Factor graph of local function decomposition

Fig.4 Factor graph of variable dependency relation

Fig.5 FGO-GNSS/PDR integrated navigation method

Fig.6 Flow chart of the solving process

Fig.7 Dataset collecting equipment

Fig.8 Hand-held experimental 1 horizontal position errors and distribution

Table 1 Hand-held experiment 1 statistical results of horizontal position errors

方法	误差均值/m	误差均方根/m
KF-GNSS/PDR	1.48	0.77
FGO-GNSS/PDR	0.83	0.50

Fig.9 Hand-held experiment 2 horizontal position errors and distribution

Table 2 Hand-held experiment 2 statistical results of horizontal position errors

方法	误差均值/m	误差均方根/m
KF-GNSS/PDR	1.78	1.18
FGO-GNSS/PDR	1.00	0.61

Fig.10 Knapsack experiment 1 horizontal position errors and distribution

Fig.11 Knapsack experiment 2 horizontal position errors and distribution

Table 3 Knapsack experiment 1 statistical results of horizontal position errors

方法	误差均值/m	误差均方根/m
KF-GNSS/PDR	1.21	0.84
FGO-GNSS/PDR	0.72	0.53

Table 4 Knapsack experiment 2 statistical results of horizontal position errors

方法	误差均值/m	误差均方根/m
KF-GNSS/PDR	2.34	1.39
FGO-GNSS/PDR	1.03	0.71

References 21

[1]	郑学理, 付敬奇. 基于PDR和RSSI的室内定位算法研究[J]. 仪器仪表学报, 2015, 36(5): 1177-1185.
	ZHENG X L, FU J Q. Study on PDR and RSSI based indoor localization algorithm[J]. Chinese Journal of Scientific Instrument, 2015, 36(5): 1177-1185. (in Chinese)
[2]	CHEN G L, MENG X L, WANG Y J, et al. Integrated WiFi/PDR/Smartphone using an unscented Kalman filter algorithm for 3D indoor localization[J]. Sensors, 2015, 15(9): 24595-24614. doi: 10.3390/s150924595 pmid: 26404314
[3]	沈世斌, 谢非, 赵静, 等. 基于相位控制的惯性与卫星超紧组合导航系统信号解调方法[J]. 兵工学报, 2020, 41(3): 495-506. doi: 10.3969/j.issn.1000-1093.2020.03.010
	SHEN S B, XIE F, ZHAO J, et al. A signal demodulation method for ultra-tightly coupled INS/GNSS navigation system based on phase control[J]. Acta Armamentarii, 2020, 41(3): 495-506. (in Chinese) doi: 10.3969/j.issn.1000-1093.2020.03.010
[4]	YU Y, CHEN R Z, CHEN L, et al. A novel 3-D indoor localization algorithm based on BLE and multiple sensors[J]. IEEE Internet of Things Journal, 2021, 8(11): 9359-9372. doi: 10.1109/JIOT.2021.3055794 URL
[5]	JIANG C H, CHEN Y W, XU B, et al. Vector tracking based on factor graph optimization for GNSS NLOS bias estimation and correction[J]. IEEE Internet of Things Journal, 2022, 9(17): 16209-16221. doi: 10.1109/JIOT.2022.3150764 URL
[6]	周瑞, 罗磊, 李志强, 等. 一种基于智能手机传感器的行人室内定位算法[J]. 计算机工程, 2016, 42(11): 22-26. doi: 10.3969/j.issn.1000-3428.2016.11.004
	ZHOU R, LUO L, LI Z Q, et al. An indoor pedestrian positioning algorithm based on smartphone sensor[J]. Computer Engineering, 2016, 42(11): 22-26. (in Chinese) doi: 10.3969/j.issn.1000-3428.2016.11.004
[7]	肖烜, 王清哲, 程远, 等. 捷联惯导系统/里程计高精度紧组合导航算法[J]. 兵工学报, 2012, 33(4): 395-400.
	XIAO X, WANG Q Z, CHENG Y, et al. High accuracy navigation algorithm for tightly coupled INS/Odometer[J]. Acta Armamentarii, 2012, 33(4): 395-400. (in Chinese)
[8]	刘明雍, 胡俊伟, 李闻白. 一种基于改进无迹卡尔曼滤波的自主水下航行器组合导航方法研究[J]. 兵工学报, 2011, 32(2): 252-256.
	LIU M Y, HU J W, LI W B. Research on integrated navigation for autonomous underwater vehicle based on an improved unscented kalman filter[J]. Acta Armamentarii, 2011, 32(2): 252-256. (in Chinese)
[9]	徐昊玮, 廉保旺, 刘尚波. 基于滑动窗迭代最大后验估计的多源组合导航因子图融合算法[J]. 兵工学报, 2019, 40(4): 807-819. doi: 10.3969/j.issn.1000-1093.2019.04.016
	XU H W, LIAN B W, LIU S B. Multi-source integrated navigation algorithm for iterated maximum posteriori estimation based on sliding-window factor graph[J]. Acta Armamentarii, 2019, 40(4): 807-819. (in Chinese) doi: 10.3969/j.issn.1000-1093.2019.04.016
[10]	SUN R, ZHANG Z X, CHENG Q, et al. Pseudorange error prediction for adaptive tightly coupled GNSS/IMU navigation in urban areas[J]. GPS Solutions, 2022, 26(1): 1-13. doi: 10.1007/s10291-021-01184-1
[11]	吴有龙, 王晓鸣, 曹鹏. 抗差估计及Allan方差在车载组合导航系统中的应用研究[J]. 兵工学报, 2013, 34(7): 889-895. doi: 10. 3969/ j. issn. 1000-1093. 2013. 07. 015
	WU Y L, WANG X M, CAO P. The application of robust estimation and allan variance method in land vehicle navigation[J]. Acta Armamentarii, 2013, 34(7): 889-895. (in Chinese)
[12]	黄欣, 熊智, 许建新, 等. 基于零速/航向自观测/地磁匹配的行人导航算法研究[J]. 兵工学报, 2017, 38(10): 2031-2040. doi: 10.3969/j.issn.1000-1093.2017.10.020
	HUANG X, XIONG Z, XU J X, et al. Research on pedestrian navigation algorithm based on zero velocity update/heading error self-observation/geomagnetic matching[J]. Acta Armamentarii, 2017, 38(10): 2031-2040. (in Chinese) doi: 10.3969/j.issn.1000-1093.2017.10.020
[13]	KUANG J, NIU X J, CHEN X G. Robust pedestrian dead reckoning based on MEMS-IMU for smartphones[J]. Sensors, 2018, 18(5): 1391. doi: 10.3390/s18051391 URL
[14]	LI X H, WEI D Y, LAI Q F, et al. Smartphone-based integrated PDR/GPS/Bluetooth pedestrian location[J]. Advances in Space Research, 2017, 59(3): 877-887. doi: 10.1016/j.asr.2016.09.010 URL
[15]	YE J H, LI Y X, LUO H, et al. Hybrid urban canyon pedestrian navigation scheme combined PDR, GNSS and beacon based on smartphone[J]. Remote Sensing, 2019, 11(18): 2174. doi: 10.3390/rs11182174 URL
[16]	HSU L T, GU Y L, KAMIJO S. Sensor integration of 3D map aided GNSS and smartphone PDR in urban canyon with dense foliage[C]// Proceedings of IEEE/ION PLANS 2016. Savannah,GA,US:IEEE, 2016: 85-90.
[17]	REHMAN A, LIU Q, WU Z D, et al. PDR/GNSS fusion algorithm based on joint heading estimation[C]// Proceedings of China Satellite Navigation Conference. Singapore:Springer, 2019: 326-339.
[18]	HOU X Y, BERGMANN J. Pedestrian dead reckoning with wearable sensors: a systematic review[J]. IEEE Sensors Journal, 2022, 21(1): 143-152. doi: 10.1109/JSEN.7361 URL
[19]	JIANG C H, CHEN Y W, CHEN C, et al. Implementation and performance analysis of the PDR/GNSS integration on a smartphone[J]. GPS Solutions, 2022, 26: 81. doi: 10.1007/s10291-022-01260-0
[20]	JIANG C H, CHEN S, CHEN Y W, et al. GNSS vector tracking method using graph optimization[J]. IEEE Transactions on Circuits and Systems II:Express Briefs, 2020, 68(4): 1313-1317.
[21]	JIANG C H, CHEN S, CHEN Y W, et al. Superior position estimation based on optimization in GNSS[J]. IEEE Communications Letters, 2020, 25(2): 479-483. doi: 10.1109/COML.4234 URL