Piotr Korbel

Mixed-mode wireless indoor positioning system using proximity detection and database correlation

The paper presents a prototype mixed-mode wireless indoor positioning and navigation system. The main goal of the system is to provide accurate and reliable location information for visually impaired users. The system also enables access to location related context information. The radio nodes of the network can operate in two power modes providing basis for both rough and precise user positioning. The paper describes an overview of the system architecture and the user interface suited to the needs of the visually impaired. Then, the details of the implemented positioning methods are presented. Finally, the results of experimental evaluation of positioning accuracy obtained using different classification strategies are discussed.

A radio network for guidance and public transport assistance of the visually impaired

The paper presents a prototype radio network aiding the visually impaired in using the public transport. The main purpose of the network is to provide accurate and reliable location information as well as to enable access to location related context information. The data from radio nodes are used to facilitate the navigation in urban and indoor areas. Keeping up the connection with dedicated system servers gives access to additional services, e.g. enables the use of passenger information system and provides access to a database storing information on points of interest (POI) of different categories. The paper describes an overview of the system architecture. Then, the details of the positioning methods implemented in the solution are presented. Finally, the user-system interaction scenarios are discussed.

CNN-based object recognition with deformable grids and multiple-feature image representation

The following paper presents a new direction that opens for deformable grid-based object recognition methods, due to introduction of their efficient, parallel implementations. A substantial increase in object recognition performance can be expected when several different features are used to build a class prototype. This would imply extending complexity of image analysis, through an application of several image characteristics in image-model matching. To make such an approach computationally feasible, a CNN is considered as ultra-fast tool for performing grid-matching process. Sample task of face recognition, which is well-suited for being tackled with deformable grids, is used to evaluate a performance of the proposed approach, yielding an expected increase in correct classification rate.

POI explorer - A sonified mobile application aiding the visually impaired in urban navigation

The paper presents POI Explorer mobile application aiding the visually impaired in spatial orientation and in urban navigation. A user equipped with a smartphone with accelerometer, electronic compass, mobile data transmission and positioning capabilities can access information on nearby points of interest. Maintaining data connection with dedicated system servers provides access to additional services facilitating the navigation in urban areas. The paper describes an overall architecture of the system. Then, the details of the user interface of the application are presented. The user interface of the application was designed to meet both the needs of visually impaired users and the requirements imposed by dynamic data changes. It employs a unique combination of text-to-speech and sonification techniques to ensure clarity of messages as well as high responsiveness of the application. The results of experiments performed in areas with different densities of points of interest proved the usability of the proposed approach.

Modeling of elastic inter-node bounds in Cellular Neural Network-based implementation of the deformable grid paradigm

The following paper extends a method of deformable grid paradigm implementation by providing modeling of elastic node interactions. The proposed method is based on parallel estimation of between-node distances, which is done by execution of appropriately arranged constrained diffusion procedure. Since, due to the introduced node interaction mechanism, nodes can get arbitrarily displaced, an appropriate method of modeling of the other component of the deformable grid paradigm - grid-image image interaction s also provided

Different Approaches to Aiding Blind Persons in Mobility and Navigation in the “Naviton” and “Sound of Vision” Projects

In this chapter, we summarize several years of research efforts aimed at building ICT (Information and Communications Technologies) based systems for aiding the blind in travel and navigation at the Lodz University of Technology, mainly as part of the “Naviton” project (http://www.naviton.pl), from a Polish Ministry of Higher Education grant. We report on different approaches we undertook in these challenging goals which comprise the following prototype solutions: (1) sonified stereovision system for obstacle avoidance and environment imaging, (2) radio beacons for local navigation, (3) remote assistance system, (4) mobile navigation applications, (5) real-time tracking of public transport vehicles, (6) haptic imaging . We shortly describe these technologies and discuss user feedback from the trials of these technological aids. Finally, we point out the key objectives and first results of a Horizon 2020 project entitled “Sound of Vision: natural sense of vision through acoustics and haptics” (http://www.soundofvision.net) that started in 2015.

Parallel implementation of elastic grid matching using cellular neural networks

The following paper presents a method that allows for a parallel implementation of the most computationally expensive element of the deformable template paradigm, which is a grid-matching procedure. Cellular Neural Network Universal Machine has been selected as a framework for the task realization. A basic idea of deformable grid matching is to guide node location updates in a way that minimizes dissimilarity between an image and grid-recorded information, and that ensures minimum grid deformations. The proposed method provides a parallel implementation of this general concept and includes a novel approach to grids elasticity modeling. The method has been experimentally verified using two different analog hardware environments, yielding high execution speeds and satisfactory processing accuracy.

Crowdsourcing based terminal positioning using multidimensional data clustering and interpolation

Recent years can be characterized by the rapid increase of mobile device usage in peoples lives. Contemporary mobile devices are equipped with many sensors and have high computational and processing capabilities. In a crowdsourcing systems, mobile users participate constructively in specific information handling. Data collected by crowd of mobile devices and stored in a database may help offering new services such as operators radio quality evaluation and tracking of users. In this paper, we focus on mobile positioning by clustering and interpolating data to match fingerprints to positions. Our method is trained during the offline phase and parameters are periodically updated to track possible changes in propagation environment.

Signal parameters identification methods used in wide-area measurement systems

The paper presents a new method identifying the magnitude and phase of signals measured in an electrical power system. The method can be used in Wide-Area Measurement Systems (WAMSs). It applies the signals sampling with varied frequency as well as an adaptive radial artificial neural network, which makes it possible to calculate the signals parameters with the accuracy specified in IEEE C37.118.1TM. The results obtained by means of this new method were compared with those received by applying quadrature filters defined in the same standard.

Crowdsourcing based terminal positioning using linear and non-linear interpolation techniques

Recent years were marked by a rapid development of mobile technologies. Contemporary mobile devices equipped with powerful processor and a variety of additional sensors open new opportunities to track user position. Crowdsourcing is a new field of research devoted to data retrieval from a crowd of users. Mobile users actively participate in collection of datasets consisting of position, power of received signal and information related to actual location. The mobile user can also inquires about his location or ask for a specific service by sending queries to the crowdsourcing database. In our work, we propose and compare on positioning techniques based on crowdsourcing data using partitioning and interpolation. A taxonomy of methods will be presented and evaluated.