دانلود رایگان مقاله محلی سازی انسانی بر اساس حسگر اینرسی و اثر انگشت در اینترنت اشیای صنعتی

Abstract

Localization services, especially for human localization, are an indispensable component of most technologies and applications related to the Industrial Internet of Things (IIoT). However, because of the complexity of an industrial environment and the mobility of the subjects, attempts to develop an accurate localization solution face certain difficulties. In this paper, we propose a novel approach that leverages the inertial sensors embedded in smartphones and uses WiFi fingerprints based on the Angle of Arrival (AoA) to assist in localization; this approach is referred to as ISWF for short. By using data from inertial sensors in smartphones and with the supplementary incorporation of fingerprint localization, our approach can overcome the difficulties posed by complex human movements and magnetic interference in an industrial environment. To accurately localize the user’s position, a step length map is designed; its training process includes step boundary detection, step length estimation and orientation angle estimation. Experiments in a realistic environment show that the presented method demonstrates superior localization performance compared with the existing methods.

6. Conclusion

In this paper, we present a new localization method called Inertial Sensors and WiFi Fingerprint localization (ISWF). Our contributions to localization techniques mainly embody in two aspects. Firstly, we introduce a novel fingerprint localiza- tion method based on AoA. It has stronger anti-interference and better stability of localizing than some previous finger- print localization. Secondly, with help of AoA fingerprint, we utilize various inertial sensors to design step length map training scheme for our ISWF algorithm. The training progress main includes three stages: step boundary detec- tion, orientation angle estimation and step length estimation. We’ve developed the algorithm on Android based phones and demonstrated its effectiveness and accuracies in compar- ison with the state-of-the-art techniques. The results show that our localization algorithm can achieve high accuracy in estimating the users’ walking position. In addition, the av- erage localization error is less than 1.4m in our experiment 747 scenarios. It provides a high tracking accuracy for the com- plex indoor environments. Considering the satisfactory per- formance of our algorithm, it can provide the effective local- ization solutions for IIoT and have potential for transforming into the practical localization technique in IIoT scenarios.