Hidden Markov Model-based Pedestrian Navigation System using MEMS Inertial Sensors

Zhang, Yingjun; Li, Wen; Yang, Xuefeng; Xing, Shengwei

doi:10.1515/msr-2015-0006

Cited by 10 publications

(5 citation statements)

References 15 publications

(17 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the sensors are fixed on the user’s foot, the stance phase of the foot can be easily detected in measurements and the periodic zero velocity updates (ZUPT) are performed to bound the error [ 29 ]. Zhang et al [ 30 ] introduced a method describing zero velocity detection with the hidden Markov model, and four states are used to describe the walking motion. Radu and Marina [ 31 ] proposed a localization system called HiMLoc for pedestrians holding their smartphones in hand or in a pocket.…”

Section: Solution Background and Related Workmentioning

confidence: 99%

Grid-Based Bayesian Filtering Methods for Pedestrian Dead Reckoning Indoor Positioning Using Smartphones

Opiela

Galčík

2020

Sensors

View full text Add to dashboard Cite

Indoor positioning systems for smartphones are often based on Pedestrian Dead Reckoning, which computes the current position from the previously estimated location. Noisy sensor measurements, inaccurate step length estimations, faulty direction detections, and a demand on the real-time calculation introduce the error which is suppressed using a map model and a Bayesian filtering. The main focus of this paper is on grid-based implementations of Bayes filters as an alternative to commonly used Kalman and particle filters. Our previous work regarding grid-based filters is elaborated and enriched with convolution mask calculations. More advanced implementations, the centroid grid filter, and the advanced point-mass filter are introduced. These implementations are analyzed and compared using different configurations on the same raw sensor recordings. The evaluation is performed on three sets of experiments: a custom simple path in faculty building in Slovakia, and on datasets from IPIN competitions from a shopping mall in France, 2018 and a research institute in Italy, 2019. Evaluation results suggests that proposed methods are qualified alternatives to the particle filter. Advantages, drawbacks and proper configurations of these filters are discussed in this paper.

show abstract

Section: Solution Background and Related Workmentioning

confidence: 99%

Grid-Based Bayesian Filtering Methods for Pedestrian Dead Reckoning Indoor Positioning Using Smartphones

Opiela

Galčík

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…The second factor is that the inclinometer is sensitive to temperature, so its output angle could be greater or smaller than the actual angle when the temperature fluctuates [12]. Using this method, the inclinometer has to be calibrated properly because of manufacturing defects, like misalignment of the inertial sensor axes during construction, soldering, or packaging [13].…”

Section: Measurement Modelmentioning

confidence: 99%

Rotating Shaft Tilt Angle Measurement Using an Inclinometer

Luo

Wang

Shen

et al. 2015

Measurement Science Review

View full text Add to dashboard Cite

This paper describes a novel measurement method to accurately measure the rotating shaft tilt angle of rotating machine for alignment or compensation using a dual-axis inclinometer. A model of the rotating shaft tilt angle measurement is established using a dual-axis inclinometer based on the designed mechanical structure, and the calculation equation between the rotating shaft tilt angle and the inclinometer axes outputs is derived under the condition that the inclinometer axes are perpendicular to the rotating shaft. The reversal measurement method is applied to decrease the effect of inclinometer drifts caused by temperature, to eliminate inclinometer and rotating shaft mechanical error and inclinometer systematic error to attain high measurement accuracy. The uncertainty estimation shows that the accuracy of rotating shaft tilt angle measurement depends mainly on the inclinometer uncertainty and its uncertainty is almost the same as the inclinometer uncertainty in the simulation. The experimental results indicate that measurement time is 4 seconds; the range of rotating shaft tilt angle is 0.002° and its standard deviation is 0.0006° using NS-5/P2 inclinometer, whose precision and resolution are ±0.01° and 0.0005°, respectively.

show abstract

“…Thus, the zero velocity update (ZUPT) algorithm has been proposed and widely used to mitigate the accumulated error [22]. The elegance of the framework lies in the fact that the foot swings to stance phase periodically during most types of human locomotion, such as walking, running, and ascending or descending stairs [23,24]. Once a stance phase is detected, the error of velocity output and zero velocity can be employed as the pseudo measurement so as to correct accumulated errors.…”

Section: Introductionmentioning

confidence: 99%

Foot-Mounted Pedestrian Navigation Method by Comparing ADR and Modified ZUPT Based on MEMS IMU Array

Chen

2020

Sensors

View full text Add to dashboard Cite

Using an MEMS Inertial Measurement Unit (MEMS IMU) array mounted on foot is a feasible approach to improve the pedestrian tracking accuracy for the pedestrian navigation system (PNS). Based on the in-house developed IMU array, the paper proposes a new integrated framework that combines adaptive deck reckoning (ADR) with the modified zero velocity update (ZUPT). In the proposed ADR algorithm, the IMUs with large drift errors on the array are selected and removed according to the step length and the track angle computed by each IMU on the array. Then, by using the step length and the track angle of each step computed by remaining IMUs, the foot position extracted from the traditional ZUPT model is estimated on the basis of least squares (LS) so as to improve the traveled distance calculation accuracy. Compared with the traditional IMU array fusion method based on a maximum likelihood estimator (MLE) when it is used in the PNS, which is approximately taking the mean value of array readings, the proposed method is equivalent to adaptively fusing the array readings and thus improves the pedestrian tracking accuracy. To compare the proposed method with MLE, two different types of walking tracks are designed. The 161 m straight line experiments show that the end position by ADR/modified ZUPT method is much closer to the one of the reference trajectory compared with the MLE in repeated walks, and the closed-loop tracks about 300 m show that the positioning error with respect to the total traveled distance is less than 0.6% (1σ), which is higher than 1% (1σ) of MLE.

show abstract

Hidden Markov Model-based Pedestrian Navigation System using MEMS Inertial Sensors

Cited by 10 publications

References 15 publications

Grid-Based Bayesian Filtering Methods for Pedestrian Dead Reckoning Indoor Positioning Using Smartphones

Grid-Based Bayesian Filtering Methods for Pedestrian Dead Reckoning Indoor Positioning Using Smartphones

Rotating Shaft Tilt Angle Measurement Using an Inclinometer

Foot-Mounted Pedestrian Navigation Method by Comparing ADR and Modified ZUPT Based on MEMS IMU Array

Contact Info

Product

Resources

About