Urban environments present significant challenges to commercial Global Navigation Satellite Systems (GNSS) receivers due to degraded satellite visibility and Non-line-of-sight (NLOS) receptions. Mitigating NLOS receptions for GNSS is essential, especially for safety-critical and reliability-critical location-based applications. Traditional physical error channel propagation modeling encountered bottlenecks since the NLOS and multipath errors cannot be modeled accurately in complex urban environments. Data-driven methods show significant potential for effectively classifying GNSS Line-of-sight (LOS) and NLOS. This paper proposes the CarNet - a generative Convolutional Neural Network (CNN)-based GNSS LOS/NLOS classifier by transforming multivariate time-series data into images. CarNet comprises two modules: an image generator and an image classifier. The image generator enriches and augments the original 1-dimension feature vector into 2-dimension feature maps and the image classifier uses an inception-based CNN to realize multi-scale feature extraction and classification. The proposed architecture is trained and tested on more than 6 h of real vehicle data collected in different challenging environments (about 1.6 million samples). A thorough benchmark is conducted, comparing CarNet against the existing mainstream Artificial Intelligence (AI) methods. The results with cross-validation on unseen data indicate that CarNet achieves the highest accuracy, i.e., 81.47% while maintaining the optimal balance between precision for both classes: 83.3% for LOS and 70.99% for NLOS. Finally, positioning accuracy is assessed using a reweighting strategy based on the LOS/NLOS information predicted by CarNet. The assessment of total datasets shows that CarNet weighting can achieve the best accuracy compared to the traditional weighting schemes based on signal-to-noise ratio or satellite elevation. CarNet shows strong potential for embedding into GNSS receivers to enhance positioning accuracy in complex urban environments, benefiting a wide range of location-based applications such as autonomous driving , emergency response, and urban logistics.

The Kalman filter is widely used in GNSS/INS integrated navigation systems to fuse information, resulting in high precision and robust positioning performance. In the Kalman filter, the accuracy of the process noise covariance matrix directly affects the precision of the positioning results. We propose a Soft Actor-Critic (SAC) algorithm based on Gated Recurrent Unit neural networks (GRU-SAC) to dynamically adjust the process noise covariance matrix online using sequential observation data to improve positioning accuracy. We model the decision-making process as a Partially Observable Markov Decision Process (POMDP) and incorporate multiple information sources as the system state. The GRU network is used to extract temporal features from the navigation data and to address memory consumption issues commonly associated with POMDPs. And the SAC algorithm continuously adjusts the process noise covariance based on observations from the Kalman filter, allowing the algorithm to perform better in complex, dynamic, and changing navigation environments. Additionally, we provide detailed design and deployment strategies for both loosely-coupled and tightly-coupled systems. Extensive experiments have been conducted to validate the effectiveness of our method. The results show that our approach generalizes well across a wide range of preset process noise covariance matrices and performs excellently in suppressing error drift during GNSS outages.

By using only accelerometers, gyroscopes, and magnetometers, inertial navigation systems can continuously track a pedestrian’s position without relying on external signals, making them a smart choice for seamless location-based services. While many AI-based pedestrian inertial navigation models already exist, most of them are generic and trained on a large amount of data. These models face accuracy limitations due to their inability to account for individual walking characteristics and various scenarios. In this paper, we propose a novel approach to pedestrian inertial navigation tailored to individual users. Our method uses gait segmentation techniques to leverage the cyclical nature of human locomotion, allowing for personalized modeling of walking patterns. A key advantage of our approach is its ability to achieve high accuracy with relatively small training sets—about 36 times smaller than the dataset used to train the state-of-the-art model, RoNIN. Through extensive real-world evaluation, covering 42.6 km with seven profiles learned from six volunteers, we show that our tailored models consistently outperform the RoNIN algorithm. Specifically, we report a stride length error of $0.07 \pm 0.07 \mathrm{~m}$ and a stride angular error of 7.72° ± 7.07°. Furthermore, the proposed method, when evaluated against a real-time implementation of another state-of-the-art generic model, IMUNet, shows superior performance in both walking distance and direction estimations.

In this article, we explore the impact of outfits on AI-based Pedestrian Dead Reckoning (PDR) with a pocket-worn inertial sensor. This PDR mode faces significant variability due to the countless choices of outfits available. We observe significant variations in the inertial signals captured by a pocket-worn device, which are highly influenced by the outfit being worn. To address this, we propose a 2-category classification of outfits as tight or loose, based on their impact on the inertial signals. Notably, AI models trained on tight outfits exhibit poor generalization with loose outfits and vice versa. We highlight this phenomenon by implementing a data-low-cost PDR algorithm based on Support Vector Regression (SVR) and assess its performance on two healthy volunteers and a senior and blind volunteer wearing tight and loose outfits, on real-life situation test tracks spanning approximately 200 to 400 meters.

Global Navigation Satellite System (GNSS)-based navigation is usually considered as not usable indoors where the satellite visibility is degraded, and the complex propagation conditions perturb the GNSS signals with reflection and refraction. However, this paper presents a novel approach called LIGHT(Light Indoor GNSS macHine-learning-based Time difference carrier phase) that can select healthy indoor GNSS carrier phase data thanks to Machine Learning (ML) for positioning. The selected carrier phase data are fed into a Time Difference Carrier Phase (TDCP) based Extended Kalman Filter (EKF) to estimate the user’s velocity. Two indoor scenarios (shopping mall and railway station) are tested over a 2 km total walking distance. It is shown that at least half of the epochs become usable for GNSS TDCP standalone positioning, and the accuracy of the velocity estimates can improve up to 87% in terms of the 75th percentile of the absolute horizontal velocity error compared with the non-ML approach. Furthermore, a newly-designed hybridization filter LIGHT-PDR that fuses the LIGHT algorithm and Pedestrian Dead Reckoning (PDR) solution together is applied to perform seamless indoor/outdoor positioning in a more robust way.

Navigation is one of the major difficulties for Visually Impaired People (VIP). Spatial landmarks are of great importance for them to find their way and orient themselves. In this paper, the most commonly used spatial landmarks by VIP are identified and their geometric constraints are constructed to pave the way for map-matching algorithms. The representative landmarks for VIP were identified through a systematic interview with 12 VIP, whose profiles cover different levels of vision impairments using different assistive mobility aids. Various analyses are performed based on their sensory modality, frequency of use as well as the number of users. Next, the previously identified landmarks are divided into two categories: waypoints and reassurance points, depending on whether they contribute directly to map-matching algorithms. Geometric constraints are designed for each identified landmark to facilitate their integration into the map-matching or path-planning algorithms. Finally, an explicit dictionary of landmarks and their geometric constraints is proposed dedicated to the VIP’s navigation in cities. Through a user centric approach, our method translates the subjective, personal navigation experiences of the VIP into an objective, universally accessible format.

Miniaturized IMU (inertial measurement units) are widely integrated in wearable devices, promoting the versatile and low cost pedestrian inertial navigation technology, especially for indoor environment. In recent years, AI (Artificial Intelli-gence) is applied to improve the performance of this technology. AI methods work with data samples, thus it is important to select a suitable process for segmenting the inertial data sequences. This survey classifies AI methods for pedestrian inertial navigation into two categories, namely human gait driven methods and sampling frequency driven methods, according to their data segmentation process. Human gait driven methods segment the inertial measurement sequence by gait (step or stride) events and learn to infer a gait vector (step/stride length and direction) given a gait segment. Sampling frequency driven methods learn to infer the user's velocity or change in position given a fixed-length segment of inertial measurements. The survey studies the underlying assumptions and their validity of the two categories of AI methods. Two methods (SELDA and RoNIN), each from a category, are chosen for evaluation and comparison, on three testing tracks totaling 770m, covering indoor and outdoor en-vironment, including stairs. The experiments highlight the two methods' advantages and limitations, supporting the theoretical analyses. The selected methods achieve 7m and 12m positioning errors, respectively.

This paper presents a wearable positioning system which is able to provide real-time positioning information for all kinds of environments in a robust way. The system is mainly based on a foot-mounted INS assisted by a GNSS receiver as well as a barometer. The scenario presented in this paper took place during the final competition of the challenge MALIN (MAîtrise de la Localisation INdoor) organized by the DGA (Direction Générale de l’Armement) and the French National Research Agency (ANR). The objective of this challenge is to create a positioning system to track emergency response agents in non-collaborative environments, where GNSS signals are usually defeated. The proposed INS-based foot-mounted system is able to provide highly accurate positioning for various motion types (walking, running, stairs, ladder) thanks to a machine learning-based Zero Velocity Detector (ZVD). The external GNSS receiver is used to capture the GNSS positions in favorable conditions and further to provide absolute position and orientation corrections via a least square-based point pattern matching algorithm. The proposed system is tested over a 2.5 km trajectory in a soldier scenario including complex outdoor and deep indoor environments. The proposed system was able to provide real-time positioning information with an accuracy around 0.3% of error over the total traveled distance. The 75% HPE remains below 8 m and the 75% VPE is under 3 m.

With the highly development of navigation techniques during the past decades, the demand for seamless indoor-outdoor navigation is growing from different application fields especially for the military or the first response emergency services. For military applications, one of the key performance requirements is the availability of the positioning solutions for all kinds of dynamics in different environments. Furthermore, due to the stealth requirement in some military actions, it is impossible for military vehicles or personnel to emit signals which enable to be detected by their opponents. This limitation prevents the use of infrastructure-based cooperative localization techniques.The research work of this paper aims at facing the following challenging issues: firstly, to design a positioning filter which is adaptive to the dynamic changes between walking and driving; secondly, to find an approach that correctly identifies the transition between outdoor and indoor with reduced latency; finally, to construct a loosely coupling GNSS/IMU scheme which takes into account the GNSS signal distortion in indoor and urban spaces.Under this context, we propose a complete indoor-outdoor infrastructure-free positioning prototype including a foot-mounted reference navigation system named Pedestrian Reference System (PERSY) and a Ublox High Sensitivity GNSS (HS-GNSS) receiver (M8P). A loosely-coupled architecture between GNSS receiver and the PERSY is employed by using an indicator of horizontal position accuracy PACCH provided by the GNSS Ublox M8P receiver. This indicator allows qualifying the position solutions delivered by the GNSS receiver as well as detecting the transition of indoor/outdoor, which helps the PERSY to update with absolute positions from GNSS. This positioning prototype can take advantage of both GNSS and PERSY so as to realize a seamless indoor-outdoor positioning for pedestrians and vehicles. The proposed system is evaluated in two scenarios over respectively 2.17 km and 2.68 km including indoor , outdoor and in-vehicle phases. The median horizontal position errors for the two scenarios are respectively 2.23 m and 1.93 m.

Integrity is one criteria to evaluate GNSS performance, which was first introduced in the aviation field. It is a measure of trust which can be placed in the correctness of the information supplied by the total system. In recent years, many GNSS-based applications emerge in the urban environment including liability critical ones, so the concept of integrity attracts more and more attention from urban GNSS users. However, the algorithms developed for the aerospace domain cannot be introduced directly to the GNSS land applications. This is because a high data redundancy exists in the aviation domain and the hypothesis that only one failure occurs at a time is made, which is not the case for the urban users. The main objective of this paper is to provide an overview of the past and current literature discussing the GNSS integrity for urban transport applications so as to point out possible challenges faced by GNSS receivers in such scenario. Key differences between integrity monitoring scheme in aviation domain and urban transport field are addressed. And this paper also points out several open research issues in this field.