Smartphone-based indoor position and orientation tracking fusing inertial and magnetic sensing

Huang, Cheng-Kai; Zhang, Gong; Jiang, Zhuqing; Li, Chao; Wang, Yupeng; Wang, Xueyang

doi:10.1109/wpmc.2014.7014819

Cited by 16 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ChengKai Huang [ 43 ] proposed a magnetic map-matching algorithm and an improved pedestrian dead reckoning algorithm (IPDR). This improved pedestrian dead reckoning provides efficient indoor navigation.…”

Section: Related Workmentioning

confidence: 99%

An Algorithm with Iteration Uncertainty Eliminate Based on Geomagnetic Fingerprint under Mobile Edge Computing for Indoor Localization

Sun

Liu

et al. 2022

Sensors

View full text Add to dashboard Cite

Indoor localization problems are difficult due to that the information, such as WLAN and GPS, cannot achieve enough precision for indoor issues. This paper presents a novel indoor localization algorithm, GeoLoc, with uncertainty eliminate based on fusion of acceleration, angular rate, and magnetic field sensor data. The algorithm can be deployed in edge devices to overcome the problems of insufficient computing resources and long delay caused by high complexity of location calculation. Firstly, the magnetic map is built and magnetic values are matched. Secondly, orientation updating and position selection are iteratively executed using the fusion data, which gradually reduce uncertainty of orientation. Then, we filter the trajectory from a path set. By gradually reducing uncertainty, GeoLoc can bring a high positioning precision and a smooth trajectory. In addition, this method has an advantage in that it does not rely on any infrastructure such as base stations and beacons. It solves the common problems regarding the non-uniqueness of the geomagnetic fingerprint and the deviation of the sensor measurement. The experimental results show that our algorithm achieves an accuracy of less than 2.5 m in indoor environment, and the positioning results are relatively stable. It meets the basic requirements of indoor location-based services (LBSs).

show abstract

Section: Related Workmentioning

confidence: 99%

An Algorithm with Iteration Uncertainty Eliminate Based on Geomagnetic Fingerprint under Mobile Edge Computing for Indoor Localization

Sun

Liu

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…The estimated step length needs one or more training values to calibrate in the training step length model [26]. A linear step length model is developed using walking frequency and acceleration variance [27,28]. Another nonlinear step length model is one in which the vertical acceleration is segmented and analyzed to estimate step length.…”

Section: Step Length Modelmentioning

confidence: 99%

A Hybrid Dead Reckon System Based on 3-Dimensional Dynamic Time Warping

Chen

Peng

et al. 2019

Electronics

View full text Add to dashboard Cite

In recent years, using smartphones for indoor positioning has become increasingly popular with consumers. This paper presents an integrated localization technique for inertial and magnetic field sensors to challenge indoor positioning without Wi-Fi signals. For dead-reckoning (DR), attitude angle estimation, step length calculation, and step counting estimation are introduced. Dynamic time warping (DTW) usually calculates the distance between the measured magnetic field and magnetic fingerprint in the database. For DR/Magnetic matching (MM), we creatively propose 3-dimensional dynamic time warping (3DDTW) to calculate the distance. Unlike traditional DTW, 3DDTW extends the original one-dimensional signal to a two-dimensional signal. Finally, the weighted least squares further improves indoor positioning accuracy. In the three different experimental scenarios—teaching building, study room, office building—DR/MM hybrid positioning accuracy is about 3.34 m.

show abstract

“…In predicting the orientation of the pedestrians, the number of classes increases when every orientation is predicted. Therefore, most existing studies [ 1 , 2 , 3 , 4 , 5 , 6 , 12 ] have used a method for recognising the orientations by dividing them into N clusters as shown in Figure 2 . For example, the TUD dataset consists of 5228 images of pedestrians with a bounding box and orientation annotations, such as back, front, left, right, left back, right back, left front, and right front.…”

Section: Teacher–student Frameworkmentioning

confidence: 99%

“…There are three types of approaches to POE. Sensor-based approaches use the gyroscope sensor of a smartphone [ 5 ] or a Kinect sensor [ 6 ] because the geometry information brought about by the depth helps overcome the fundamental problems in computer vision, such as a cluttered environment, changes in illumination, and partial occlusions. However, these approaches can only be performed using a smartphone, and if the distance between the pedestrian and the sensor is too great, the recognition may be impaired.…”

Section: Introductionmentioning

confidence: 99%

Estimation of Pedestrian Pose Orientation Using Soft Target Training Based on Teacher–Student Framework

Heo

Nam

2019

Sensors

View full text Add to dashboard Cite

Semi-supervised learning is known to achieve better generalisation than a model learned solely from labelled data. Therefore, we propose a new method for estimating a pedestrian pose orientation using a soft-target method, which is a type of semi-supervised learning method. Because a convolutional neural network (CNN) based pose orientation estimation requires large numbers of parameters and operations, we apply the teacher–student algorithm to generate a compressed student model with high accuracy and compactness resembling that of the teacher model by combining a deep network with a random forest. After the teacher model is generated using hard target data, the softened outputs (soft-target data) of the teacher model are used for training the student model. Moreover, the orientation of the pedestrian has specific shape patterns, and a wavelet transform is applied to the input image as a pre-processing step owing to its good spatial frequency localisation property and the ability to preserve both the spatial information and gradient information of an image. For a benchmark dataset considering real driving situations based on a single camera, we used the TUD and KITTI datasets. We applied the proposed algorithm to various driving images in the datasets, and the results indicate that its classification performance with regard to the pose orientation is better than that of other state-of-the-art methods based on a CNN. In addition, the computational speed of the proposed student model is faster than that of other deep CNNs owing to the shorter model structure with a smaller number of parameters.

show abstract

Smartphone-based indoor position and orientation tracking fusing inertial and magnetic sensing

Cited by 16 publications

References 13 publications

An Algorithm with Iteration Uncertainty Eliminate Based on Geomagnetic Fingerprint under Mobile Edge Computing for Indoor Localization

An Algorithm with Iteration Uncertainty Eliminate Based on Geomagnetic Fingerprint under Mobile Edge Computing for Indoor Localization

A Hybrid Dead Reckon System Based on 3-Dimensional Dynamic Time Warping

Estimation of Pedestrian Pose Orientation Using Soft Target Training Based on Teacher–Student Framework

Contact Info

Product

Resources

About