Every year, for ten years now, the IPIN competition has aimed at evaluating real-world indoor localisation systems by testing them in a realistic environment, with realistic movement, using the EvAAL framework. The competition provided a unique overview of the state-of-the-art of systems, technologies, and methods for indoor positioning and navigation purposes. Through fair comparison of the performance achieved by each system, the competition was able to identify the most promising approaches and to pinpoint the most critical working conditions. In 2020, the competition included 5 diverse off-site off-site Tracks, each resembling real use cases and challenges for indoor positioning. The results in terms of participation and accuracy of the proposed systems have been encouraging. The best performing competitors obtained a third quartile of error of 1 m for the Smartphone Track and 0.5 m for the Foot-mounted IMU Track. While not running on physical systems, but only as algorithms, these results represent impressive achievements.

Simultaneous localization and mapping (SLAM) is currently a widely used technology for indoor positioning. Many studies have focused on the smartphone-based localization and navigation for its portability and feature-rich sensors. But due to the low-quality sensors and complex environments, current SLAM-based positioning technology using smartphones still poses great challenges, such as inherent cumulative global drift and potential divergence facing texture-less indoor regions. For the regular gait characteristics of pedestrians when naturally walking, the gait motion model can certainly provide effective observation of the pedestrian motion state. We propose a visual-inertial odometry (VIO) assisted by pedestrian gait information for smartphone-based indoor positioning. This work mainly builds two additional state constraints, pedestrian velocity, and step displacement, obtained by the pedestrian dead reckoning (PDR) algorithm for the visual-inertial tracking system. For each step, the corresponding residual term of step length and velocity constraints is constructed and added to the cost function for nonlinear sliding-window optimization. Furthermore, the step displacement is applied again in the four-degree-of-freedom (4-DOF) graph-based optimization to refine the trajectory. VIO system also assists the PDR algorithm in mode switching, to improve the accuracy of gait information by applying the adaptive step length formula. Field experiments were conducted, and the results indicate that with the aiding of the pedestrian gait information, the accuracy and robustness of the visual-inertial pedestrian tracking system using smartphones have been significantly improved. Compared with the state-of-the-art algorithm monocular visual-inertial navigation system (VINS-MONO), our method improves the accuracy by 54.6% on average in our field tests in challenging environments.

Smartphone-based pedestrian dead reckoning (PDR) has been widely used indoors for continuous localization. However, the specific tracking solutions under different modes are vulnerable to mode transition and thus degrading performance. The robustness of pedestrian navigation may be weakened due to the mix of smartphone motions and walking patterns. Due to this challenge, most existing PDR methods assume that the smartphone is carried in a certain pose, such as handheld horizontally, swinging, calling, and pocketed, ignoring the short period but negative transition impact on tracking, which limits its flexibility when applying in the Internet of Things (IoT) services. To achieve a fully flexible PDR (i.e., FlexPDR), this article enhances the robustness and smoothness of pedestrian tracking during the transition between several phone poses and regular motion modes for the first time. We propose a Bayesian-based real-time multimode recognition method that does not require any posterior information, together with a time-series decomposition approach for adaptively tracking scheme-switching. The proposed FlexPDR system achieved a real-time smartphone indoor positioning with a high position accuracy of 98.11% on the specific situation that mixed mode-switching happens, which outperformed other state-of-the-art methods.

Pedestrian dead reckoning (PDR) is widely used in handheld indoor positioning systems. However, low-cost inertial sensors built into smartphones provide poor-quality measurements, resulting in cumulative error which consists of heading estimation error caused by gyroscope and step length estimation error caused by an accelerometer. Learning more motion features through limited measurements is important to improve positioning accuracy. This paper proposes an improved PDR system using smartphone sensors. Using gyroscope, two motion patterns, walking straight or turning, can be recognised based on dynamic time warp (DTW) and thus improve heading estimation from an extended Kalman filter (EKF). Joint quasi-static field (JQSF) detection is used to avoid bad magnetic measurements due to magnetic disturbances in an indoor environment. In terms of periodicity of angular rate while walking, peak–valley angular velocity detection and zero-cross detection is combined to detect steps. A step-length estimation method based on deep belief network (DBN) is proposed. Experimental results demonstrate that the proposed PDR system can achieve more accurate indoor positioning.

Calibration and initialization of point feature-based SLAM systems are prone to performance degradation in challenging environments where rapid motion and texture-less areas are present. Inaccurate calibration and initialization results can introduce additional errors to backend optimization, thereby affecting the localization performance. Difficulty in convergence also limits the practical application of SLAM systems. The structure line segment features present more stability and compact spatial constraint, which can provide additional structural constraints for visual-inertial state estimation. Thus, in this paper, we propose to incorporate the structure information of extracted line segment features and construct an enhanced online calibration and initialization framework for visual-inertial SLAM systems. The proposed method estimates initial velocity, gravity orientation, gyroscope bias, scale, extrinsic rotation, and the potential time delay between IMU and camera data sequence. A vertical structure direction-aided closed-form solution is proposed, achieving both higher initialization accuracy and less convergence time. Experimental results on the synthetic data sequences and real-world dataset prove that our method can achieve robust and accurate calibration and initialization with the introduction of structure information, which is superior to the state-of-the-art works in a high success rate of up to 85% and less convergence time of initialization to 0.5 s within 15 keyframes.

In an intelligent information society, indoor pedestrian navigation systems based on inertial sensors are becoming increasingly important due to advantages of autonomy and continuity in positioning. However, these systems cannot estimate pedestrian altitude because of the divergence of inertial altitude channels and the complexity of pedestrian movement patterns. To address this challenge, this paper proposes an indoor pedestrian altitude estimation method utilizing the inertial measurement unit (IMU) sensor installed on the pedestrian’s foot. Specifically, the method calculates the height difference of steps with the inertial navigation system (INS), extended Kalman filter (EKF), and zero-velocity update (ZUPT), named the IEZ framework. Then, a convolutional neural networks-support vector machine (CNN-SVM) model is developed to accurately identify complex pedestrian motion patterns including horizontal walking, running, walking stairs, standing, and taking elevators and escalators. Next, the height difference is adaptively corrected according to the motion modes. For horizontal walking and standing, the height difference based on the IEZ framework is revised and remains unchanged; for walking or running stairs, the pedestrian altitude is determined by the stair step model. In the case of taking elevators or escalators, the vertical displacement is obtained by integrating the barometric height measured as EKF observations with the double integral value of the vertical acceleration. Experimental results show that the CNN-SVM model achieves a classification accuracy of 98.8%, and a positioning error for pedestrian altitude estimation of less than 0.6 m with an indoor walking distance of approximately 500 m.