In the recent past, we have witnessed the adoption of different machine learning techniques for indoor positioning applications using WiFi, Bluetooth and other technologies. The techniques range from heuristically derived hand-crafted feature-based traditional machine learning algorithms, feature selection algorithms to the hierarchically self-evolving feature-based Deep Learning algorithms. The transient and chaotic nature of the WiFi/Bluetooth fingerprint data along with different signal sensitivity of different device configurations presents numerous challenges that influence the performance of the indoor localization system in the wild. This article is intended to offer a comprehensive state-of-the-art survey on machine learning techniques that have recently been adopted for localization purposes. Hence, we review the applicability of machine learning techniques in this domain along with basic localization principles, applications, and the underlying problems and challenges associated with the existing systems. We also articulate the recent advances and state-of-the-art machine learning techniques to visualize the possible future directions in the research field of indoor localization.

A Survey of Machine Learning Techniques for Indoor Localization and Navigation Systems

Combining particle filter algorithm with bio-inspired anemotaxis behavior: A smoke plume tracking method and its robotic experiment validation

Indoor localization has the capability to change the way of providing location-based services in a closed environment and has more potential than that of GPS if the present shortcomings can be overcome. Thus, developing a ubiquitous Indoor Localization System (ILS) is the need of the day. WiFi-based indoor localization using smartphones is a promising approach to achieve ubiquity since smartphones are widely available today and most of the buildings are WiFi enabled. However, the significant variation of WiFi signal strengths with ambient conditions as well as device configuration badly affects the localization accuracy. Hence, a ubiquitous ILS, which would be widely available and would not require any specialized hardware support is still out of reach. Thus, this paper surveys state-of-the-art WiFi-based indoor localization techniques with a critical analysis of their applicability as a ubiquitous system. Consequently, the main objective of this paper is to highlight the key research challenges of implementing a WiFi-based ILS when system ubiquity is the prime concern with a discussion on future direction. Some of the experiments have been implemented on a real-life dataset in order to indicate the implementation challenges of such a system. 

/pdf/a-survey-on-ubiquitous-wifi-based-indoor-localization-system-1zqwhlw0.pdf

A survey on ubiquitous WiFi-based indoor localization system for smartphone users from implementation perspectives

Device-free wireless indoor localization is a key enabling technology for the Internet of Things (IoT). Fingerprint-based indoor localization techniques are a commonly used solution. This paper proposes a semi-supervised, generative adversarial network (GAN)-based device-free fingerprinting indoor localization system. The proposed system uses a small amount of labeled data and a large amount of unlabeled data (i.e., semi-supervised), thus considerably reducing the expensive data labeling effort. Experimental results show that, as compared to the state-of-the-art supervised scheme, the proposed semi-supervised system achieves comparable performance with equal, sufficient amount of labeled data, and significantly superior performance with equal, highly limited amount of labeled data. Besides, the proposed semi-supervised system retains its performance over a broad range of the amount of labeled data. The interactions between the generator, discriminator, and classifier models of the proposed GAN-based system are visually examined and discussed. A mathematical description of the proposed system is also presented.

Semi-Supervised Learning with GANs for Device-Free Fingerprinting Indoor Localization

/pdf/a-survey-on-ubiquitous-wifi-based-indoor-localization-system-27p3u9y9.pdf

Vision-and-language navigation (VLN) is a challenging task that requires an agent to navigate in real-world environments by understanding natural language instructions and visual information received in real-time. Prior works have implemented VLN tasks on continuous environments or physical robots, all of which use a fixed camera configuration due to the limitations of datasets, such as 1.5 meters height, 90 degrees horizontal field of view (HFOV), etc. However, real-life robots with different purposes have multiple camera configurations, and the huge gap in visual information makes it difficult to directly transfer the learned navigation model between various robots. In this paper, we propose a visual perception generalization strategy based on meta-learning, which enables the agent to fast adapt to a new camera configuration with a few shots. In the training phase, we first locate the generalization problem to the visual perception module, and then compare two meta-learning algorithms for better generalization in seen and unseen environments. One of them uses the Model-Agnostic Meta-Learning (MAML) algorithm that requires a few shot adaptation, and the other refers to a metric-based meta-learning method with a feature-wise affine transformation layer. The experiment results show that our strategy successfully adapts the learned navigation model to a new camera configuration, and the two algorithms show their advantages in seen and unseen environments respectively.

Visual Perception Generalization for Vision-and-Language Navigation via Meta-Learning

Prior works in vision-and-language navigation (VLN) focus on using long short-term memory (LSTM) to carry the flow of information on either the navigation model (navigator) or the instruction generating model (speaker).The outstanding capability of LSTM to process inter-modal interactions has been widely verified, however, LSTM neglects the intra-model interactions, leading to negative effect on either navigator or speaker. The performance of attention-based Transformer is satisfactory in sequence-to-sequence translation domains, but Transformer structure implemented directly in VLN has yet been satisfied. In this paper, we propose novel Transformer-based multi-modal frameworks for the navigator and speaker respectively. In our frameworks, the multi-head self-attention with the residual connection is used to carry the information flow. Specially, we set a switch to prevent them from directly entering the information flow in our navigator framework. In experiments, we verify the effectiveness of our proposed approach, and show significant performance advantages over the baselines.

Improved Speaker and Navigator for Vision-and-Language Navigation

In this paper, a real WiFi fingerprint based indoor localization system is considered for experiments, including three primary components: the APP in the smart phone, the server system and the embedded localization algorithm. As we all know, one of the main drawbacks in fingerprint based localization is the labor intensity and time consumption of data collection. This paper proposes an improved graph-based semi-supervised learning (I-GSSL) to better overcome this problem. Apart from taking advantage of the indoor propagation model, the I-GSSL algorithm is proposed to handle the existing out-of-sample problem where an elastic regularization is considered as an extra constraint. Meanwhile, due to unequal amount of location information in the received signal strength (RSS) from different access points (APs) and the redundancy of RSS at APs, a double weighted K nearest neighbor (DWKNN) algorithm is proposed for localization. Experimental results show the proposed scheme achieves a better label propagation and localization accuracy.

Improved Graph-Based Semi-Supervised Learning for Fingerprint-Based Indoor Localization

In this paper, a real WiFi fingerprint-based indoor localization system is considered, where three primary components including the APP in smart phone, the server system and the embedded localization algorithm, have been designed. This paper proposes a dedicated data preprocessing algorithm to solve the singular-collection problem. Furthermore, the issue of WiFi access point (AP) missing is discussed and the theoretical analysis is presented under the condition of a two-AP scenario. Finally, because of unequal amount of location information contained in received signal strength (RSS) from different AP, the weighted RSS (WRSS) and the iterative weighted K nearest neighbor (IWKNN) algorithm are proposed for localization. Experimental results shows the proposed scheme achieves a competitive localization accuracy.

WiFi Fingerprint Based Indoor Localization with Iterative Weighted KNN for WiFi AP Missing

Vision-and-Language Navigation in Continuous Environments (VLN-CE) is a navigation task that requires an agent to follow a language instruction in a realistic environment. The understanding of environments is a crucial part of the VLN-CE task, but existing methods are relatively simple and direct in understanding the environment, without delving into the relationship between language instructions and visual environments. Therefore, we propose a new environment representation in order to solve the above problems. First, we propose an Environment Representation Graph (ERG) through object detection to express the environment in semantic level. This operation enhances the relationship between language and environment. Then, the relational representations of object-object, object-agent in ERG are learned through GCN, so as to obtain a continuous expression about ERG. Sequentially, we combine the ERG expression with object label embeddings to obtain the environment representation. Finally, a new cross-modal attention navigation framework is proposed, incorporating our environment representation and a special loss function dedicated to training ERG. Experimental result shows that our method achieves satisfactory performance in terms of success rate on VLN-CE tasks. Further analysis explains that our method attains better cross-modal matching and strong generalization ability.

Ting Wang

Papers

Visual Perception Generalization for Vision-and-Language Navigation via Meta-Learning

Improved Speaker and Navigator for Vision-and-Language Navigation

Improved Graph-Based Semi-Supervised Learning for Fingerprint-Based Indoor Localization

WiFi Fingerprint Based Indoor Localization with Iterative Weighted KNN for WiFi AP Missing

Graph based Environment Representation for Vision-and-Language Navigation in Continuous Environments