For positioning tasks of mobile robots in indoor environments, the emerging positioning technique based on visual inertial odometry (VIO) is heavily influenced by light and suffers from cumulative errors, which cannot meet the requirements of long-term navigation and positioning. In contrast, positioning techniques that rely on indoor signal sources such as 5G and geomagnetism can provide drift-free global positioning results, but their overall positioning accuracy is low. In order to obtain higher precision and more reliable positioning, this paper proposes a fused 5G/geomagnetism/VIO indoor localization method. Firstly, the error back propagation neural network (BPNN) model is used to fuse 5G and geomagnetic signals to obtain more reliable global positioning results; secondly, the conversion relationship from VIO local positioning results to the global coordinate system is established through the least squares principle; and finally, a fused 5G/geomagnetism/VIO localization system based on the error state extended Kalman filter (ES-EKF) is constructed. The experimental results show that the 5G/geomagnetism fusion localization method overcomes the problem of low accuracy of single sensor localization and can provide more accurate global localization results. Additionally, after fusing the local and global positioning results, the average positioning error of the mobile robot in the two scenarios is 0.61 m and 0.72 m. Compared with the VINS-mono algorithm, our approach improves the average positioning accuracy in indoor environments by 69.0% and 67.2%, respectively.

Global navigation satellite system (GNSS) signals are easily blocked by urban canyons, tree-lined roads, and overpasses in urban environments, making it impossible to ensure continuous and reliable positioning using only GNSS, even with the widely used precise point positioning and real-time kinematic (PPP-RTK). Since the inertial navigation system (INS) and GNSS are complementary, a tightly coupled PPP-RTK/INS model is developed to improve the positioning performance in these GNSS-challenged scenarios, in which the atmospheric corrections are used to achieve a rapid ambiguity resolution and the mechanization results from INS are utilized to assist GNSS preprocessing, re-fixing, and reconvergence. The experiment was conducted using three sets of vehicle-mounted data, and the performance of low-cost receiver and microelectromechanical system (MEMS) inertial measurement unit (IMU) was compared. The result shows that the positioning accuracy of PPP-RTK/INS can reach 2 cm in the horizontal component and 5 cm in the vertical component in the open environment. In the complex urban environment, continuous and reliable positioning can be ensured during GNSS short interruption, ambiguity can be instantaneously re-fixed with the assistance of INS, and decimeter-level positioning accuracy can be achieved. As a result, the horizontal positioning errors of more than 95% of the total epochs were within 20 cm. In addition, average positioning accuracy better than 15 cm and 30 cm in the horizontal and vertical components, respectively, can be obtained using the low-cost receiver and MEMS IMU. Compared with tactical IMU, the improvements in positioning accuracy and the ambiguity fixing rate using the geodetic receiver were more significant.

With the urgent need of precise positioning faced by internet of things (IoT) applications, the universality and cost of indoor positioning devices become key factors. Since the 5G network has been widely deployed, new opportunity is brought by tightly fusing the traditional low-cost sensors, i.e., the magnetometer. In this study, using 5G channel state information (CSI) and geomagnetic data, a multi-input convolutional neural network (CNN) localization system was proposed. First, to generate a data tensor that is easy for CNN processing, the raw data was reconstructed individually. Then, to comprehensively incorporate the features of 5G CSI and geomagnetic strength data, the ReLU function was chosen as the activation function of the convolutional layer. After that, a multi-input CNN was trained using the incorporated geomagnetic strength and CSI amplitude in the off-line side, and the trained CNN was recorded as a location fingerprint, which can be used for the user position prediction. Finally, in the online side and using a multi-input CNN, the 2-D coordinates are estimated and tested indoors in a typical conference room scenario. The results showed that longer sampling time of fingerprint data result in better uniqueness of the reference point, while the data collection time of locating points does not need to be long. Taking the positioning efficiency and accuracy into consideration, a sampling time of 3s at the reference point and 0.2s at the locating point are recommended. The positioning accuracy using the proposed method was 1.41 m, with an improvement of 22.9% compared with the 5G positioning, and an improvement of 18.0% compared with the 5G and geomagnetic fusing positioning using single CNN.