Indoor Localization Problem Research Articles

Cooperative simultaneous localization and mapping by exploiting multipath propagation

An affordable and reliable indoor positioning is a highly needed service. Moreover, maps of the indoor environment are vital to many applications. In this paper, a method for joint localization and mapping using multipath delay estimates is developed. Required high-resolution estimates of multipath delays may be obtained using radio frequency or acoustic measurements among a set of nodes in a network. In this paper, the problem is modeled in two-dimensional space with arbitrary node configuration and assuming a convex polygonal room shape. Joint localization and mapping is formulated as an optimization problem. It is subdivided and relaxed into two convex subproblems, which can be solved in an alternating manner. A method for data association and a low-complexity mapping algorithm stemming from Hough transform are proposed. Both the estimation performance and identifiability of the indoor localization problem are improved. Moreover, a basic map of the propagation environment is produced.

Indoor Positioning of a Robotic Walking Assistant for Large Public Environments

Indoor localization and position tracking are essential to support applications and services for ambient-assisted living. While the problem of indoor localization is still open and already quite complex per se, in large public places, additional issues of cost, accuracy, and scalability arise. In this paper, the position estimation and tracking technique developed within the project devices for assisted living (DALi) is described, analyzed through simulations, and finally validated by means of a variety of experiments on the field. The goal of the DALi project is to design a robotic wheeled walker guiding people with psychomotor problems. Indeed, people with motor or cognitive impairments are often afraid of moving in large and crowded environments (e.g., because they could lose the sense of direction). In order to mitigate this problem, the position tracking approach described in this paper is based on multisensor data fusion and it is conceived to assure a good tradeoff between target accuracy, level of confidence, and deployment costs. Quite interestingly, the same approach could be used for indoor automated guided vehicles and robotics.

Robot Localization Using the Phase of Passive UHF-RFID Signals Under Uncertain Tag Coordinates

An indoor global localization problem for a mobile robot based on Radio Frequency IDentification (RFID) is considered. The localization system consists of a reader installed on the robot which measures the phase of UHF-RFID signals coming from a set of passive tags deployed on the ceiling of the environment. Assuming only an approximate information is available at the beginning on the position of the tags, this paper presents an algorithm, based on a Multi-Hypothesis Extended Kalman Filter, which improves the initial estimate on the tag coordinates while simultaneously localizing the robot. Simulative and experimental results are reported to illustrate the effectiveness of the proposed approach.

ASC localization in noisy environment based on wireless sensor network

This study investigates indoor localization problem of robot or a customer at shopping mall environment. To improve the localization accuracy, a sensor fusion-based approach is employed, which comb...

Profiling-Based Indoor Localization Schemes

In this paper, we consider the indoor localization problem, i.e., identifying the Cartesian coordinates of an object or a person under the roof. To solve this problem we consider an RF-based localization method called profiling , a two-step process, where a radio map of the monitored area is first constructed by collecting signal strength from known locations. An unknown location is then predicted using this radio map as a reference. In this paper, we first propose a K nearest neighbor (KNN) profiling-based localization method dubbed LEMON (location estimation by mining oversampled neighborhoods). It is based on a low-cost, low-power wireless devices and ensures good accuracy compared to the state-of-the-art. We then propose a variant of LEMON called combinatorial localization which exhaustively searches for the best possible set of nearest neighbors. We further define a Bayesian network model for the same localization problem. The performance of these methods is evaluated through extensive experiments in various indoor areas. We found an interesting outcome that the simple KNN-based approach can offer better localization accuracy compared to other complex localization methods. Thus we further enhance the performance of the KNN-based approach using multiple RF channels.

Multivariable fuzzy inference system for fingerprinting indoor localization

The emergence of wireless sensor network has raised the need for cheap wireless indoor localization technique. This paper considers the problem of fingerprinting indoor localization based on signal strength measurements RSS. A new approach based on Fuzzy logic has been put forward. The proposal makes use of k-nearest neighbor classification in signal space. The localization of target node is then determined as a weighted combination of nearest fingerprints. The weights are determined using Takagi–Sugeno fuzzy controller with two inputs. A new enhancement to the kNN is proposed to enhance the accuracy of location estimation; this enhancement allows the kNN to outlier some miss elected neighbors based on triangular area measurements. The performance of the developed estimation algorithm has been evaluated using both Monte Carlo simulations and real testbed scenarios while compared to other alternative approaches.

3-D model-based tracking for UAV indoor localization.

This paper proposes a novel model-based tracking approach for 3-D localization. One main difficulty of standard model-based approach lies in the presence of low-level ambiguities between different edges. In this paper, given a 3-D model of the edges of the environment, we derive a multiple hypotheses tracker which retrieves the potential poses of the camera from the observations in the image. We also show how these candidate poses can be integrated into a particle filtering framework to guide the particle set toward the peaks of the distribution. Motivated by the UAV indoor localization problem where GPS signal is not available, we validate the algorithm on real image sequences from UAV flights.

Horizontal dilution of precision-based ultra-wideband positioning technique for indoor environments

Ultra-wideband (UWB) technology provides considerable performance in many indoor localization problems thanks to its very wide spectrum and high-resolution characteristics. In this paper, we propose using the horizontal dilution of precision (HDOP) to decrease the localization error in indoor environments for UWB localization systems. To achieve this aim, first we determine the positioning accuracy of a commercially available UWB positioning system using laboratory experiments. Next, the results of the position estimations obtained by the HDOP are compared with the experimental results acquired by the UWB positioning system. Finally, we investigate a detailed comparison with the least squares (LS), nonlinear regression (NLR), and iterative nonlinear regression (INR) techniques. In terms of the mean position estimation error, the proposed HDOP technique increases the performance of the UWB positioning system and the LS algorithm by approximately 10% and 3%, respectively. In addition, while the proposed HDOP technique provides localization for all of the test points, both the NLR and INR algorithms perform below the expected levels at the same points.

An indoor localization system based on artificial neural networks and particle filters applied to intelligent buildings

Smart Buildings aim to provide users with seamless, invisible and proactive services adapted to their preferences and needs. These services can be offered intelligently by means of considering the static and dynamical status of the building and the location of its occupants. Furthermore, gathering data about the identity and location of users enables to provide more personalized services, while wasted energy in overuse is reduced. But to cope with these objectives, it is necessary to acquire contextual information, both from users and the environment, using nonintrusive, ubiquitous and cheap technologies. In this work, we propose a low-cost and nonintrusive solution to solve the indoor localization problem focused on satisfying the requirements, in terms of accuracy in localization data, to provide customized comfort services in buildings, such as climate and lighting control, or security, with the goal of ensuring users comfort while saving energy. The proposed localization system is based on RFID (Radio-Frequency Identification) and IR (Infra-Red) data. The solution implements a Radial Basis Function Network to estimate the location of occupants, and a Particle Filter to track their next positions. This mechanism has been tested in a reference building where an automation system for collecting data and controlling devices has been setup. Results obtained from experimental assessments reveal that, despite our localization system uses a relative low number of sensors, estimated positions are really accurate considering the requirements of precision to provide user-oriented pervasive services in buildings.

Indoor Localization With Range-Based Measurements and Little Prior Information

We address the problem of indoor localization with networks of sensors taking range-based measurements in the presence of very little prior information. We propose several robust methods that do not require previous measurement campaigns when a network is deployed. The focus is on networks of ultrawideband sensors, but the proposed range-based methods can also be applied to other types of sensor networks. The location of a target node is estimated from measured distances to anchor nodes of known positions. We allow for the possibility of large errors in the range measurements due to undetected direct path (UDP) propagation conditions. In mitigating the UDP effect, our approach is to combine intermediate location estimates from different subsets of beacons. We propose novel criteria for identifying the combinations that produce bad estimates. These combinations are then discarded in obtaining the final estimates. Simulations reveal that the proposed methods achieve improved performance with respect to that of existing techniques that exploit the same prior information. Under many scenarios, the proposed methods reach the performance of some algorithms that exploit more prior information.

마찰 보상과 지도 정합에 의한 미끄럼 조향 이동로봇의 실내 주행

This paper deals with the indoor localization problem for a SSMR (Skid Steering Mobile Robot) subjected to wheel-ground friction and with one LRF (Laser Range Finder). In order to compensate for some friction effect, a friction related coefficient is estimated by the recursive least square algorithm and appended to the maneuvering command. Also to reduce odometric information based localization errors, the lines are extracted with scan points of LRF and matched with the ones of the corresponding map built in advance. The present friction compensation and scan map matching schemes have been applied to a laboratory SSMR, and experimental results are given to validate the localization performance along an indoor corridor.

Indoor Localization via Discriminatively Regularized Least Square Classification

In this paper, we address the received signal strength (RSS)-based indoor localization problem in a wireless local area network (WLAN) environment and formulate it as a multi-class classification problem using survey locations as classes. We present a discriminatively regularized least square classifier (DRLSC)-based localization algorithm that is aimed at making use of the class label information to better distinguish the RSS samples taken from different locations after proper transformation. Besides DRLSC, two other regularized least square classifiers (RLSCs) are also presented for comparison. We show that these RLSCs can be expressed in a unified problem formulation with a closed-form solution and convenient assessment of the convexity of the problem. We then extend the linear RLSCs to their nonlinear counterparts via the kernel trick. Moreover, we address the missing value problem, utilize clustering to reduce the training and online complexity, and introduce kernel alignment for fast kernel parameter tuning. Experimental results show that, compared with other methods, the kernel DRLSC-based algorithm achieves superior performance for indoor localization when only a small fraction of the data samples are used.

UDP identification and error mitigation in toa-based indoor localization systems using neural network architecture

Time-of-Arrival (ToA) based localization has attracted considerable attention for solving the very complex and challenging problem of indoor localization, mainly due to its fine range estimation process. However, ToA-based localization systems are very vulnerable to the blockage of the direct path (DP) and occurrence of undetected direct path (UDP) conditions. Erroneous detection of other multipath components as DP, which corresponds to the true distance between transmitter and receiver, introduces substantial ranging and localization error into ToA-based systems. Therefore, in order to enable robust and accurate ToA-based indoor localization, it is important to identify and mitigate occurrence of DP blockage. In this paper we present two methodologies to identify and mitigate the UDP conditions in indoor environments. We first introduce our identification technique which utilizes the statistics of radio propagation channel metrics along with binary hypothesis testing and then we introduce our novel identification technique which integrates the same statistics into a neural network architecture. We analyze each approach and the effects of neural network parameters on the accuracy of the localization system. We also compare the results of the two approaches in a sample indoor environment using both real-time measurement and ray tracing simulation. The identification metrics are extracted from wideband frequency-domain measurements conducted in a typical office building with a system bandwidth of 500 MHz, centered around 1 GHz. Then we show that with the knowledge of the channel condition, it is possible to improve the localization performance by mitigating those UDP-induced ranging errors. Finally, we compare the standard deviation of localization error of traditional localization system and UDP identification-enhanced localization system with their respective lower bound.

Indoor fuzzy self-localization using fuzzy segments

HE indoor localization problem consists on estimating the robot’s location using absolute (reference-based systems) and relative (also called dead-reckoning) position measurements [3], which give the situation of the environment around the robot and the feedback about its driving actions respectively. Basically, it consists on answering the question Where am I? [16] from the robot’s point of view given: a representation of the world where the robot is navigating and sensor information. Localization methods are usually distinguished according to the type of problem they tackle [13]; in particular, i) tracking or local techniques, and ii) global techniques. Commonly, local techniques present advantages in accuracy and efficiency, while global techniques are much more robust [14]. What makes this problem difficult is the presence of uncertainty in the measurements because it usually results hard to represent noisy measures. These measures are usually stated by giving a range of values that are likely to enclose the real value. From the probabilistic point of view, this range is usually designated by repeating the measure under the same conditions enough times for modeling the uncertainty using a probabilistic function. In other words, probabilistic functions should be experimentally calculated which implies reproducing the measure under the same conditions. However, sometimes it is not possible to repeat the measure or reproduce the factors that influence this. Techniques based on fuzzy logic only require an approximate sensor model [6] of the measures. Moreover, position measurements may be affected by different types