Bratislav Predic

ExposureSense: Integrating daily activities with air quality using mobile participatory sensing

With an increasing number of rich embedded sensors, like accelerometer and GPS, smartphone becomes a pervasive people-centric sensing platform for inferring users daily activities and social contexts. Alternatively, wireless sensor network offers a comprehensive platform for capturing the surrounding environmental information using mobile sensing nodes, e.g., the OpenSense project [2] in Switzerland deploying air quality sensors like CO on public transports like buses and trams. The two sensing platforms are typically isolated from each other. In this paper, we build ExposureSense, a rich mobile participatory sensing infrastructure that integrates the two independent sensing paradigms. ExposureSense is able to monitor peoples daily activities as well to compute a reasonable estimation of pollution exposure in their daily life. Besides using external sensor networks, ExposureSense also supports pluggable sensors (e.g., O3) to further enrich air quality data using mobile participatory sensing with smartphones.

Continuous range monitoring of mobile objects in road networks

In contrast to regular queries that are evaluated only once, a continuous query remains active over a period of time and has to be continuously evaluated to provide up to date answers. We propose a method for continuous range query processing for different types of queries, characterized by mobility of objects and/or queries which all follow paths in an underlying spatial network. The method assumes an available 2D indexing scheme for indexing spatial network data. An appropriately extended R*-tree, that primarily is used as an indexing scheme for network segments, provides matching of queries and objects according to their locations on the network or their network routes. The method introduces an additional pre-refinement step which generates main-memory data structures to support efficient, incremental reevaluation of continuous range queries in periodically performed refinement steps.

Web information system for transport telematics and fleet management

In this paper, a Web information system for fleet management and transport telematics is described. The system is developed based on MOWIS - a service-oriented open software platform for location-based and context-aware mobile and Web applications. Based on MOWIS services such as routing service, directory service, traffic information service and moving object service, as well as real AVL bus tracking data, the Web information system developed possess all functionalities needed for vehicle tracking, fleet control and management available for Web fleet control users.

Prediction of bus motion and continuous query processing for traveler information services

The paper presents the methods for prediction of bus arrival times and continuous query processing as foundations of traveler information services. The time series of data from automatic vehicle location (AVL) system, consisting of time, location and speed data, is used with historical statistics and bus schedule information to predict future arrivals and motion. Based on predicted and AVL data, continuous query processing technique is proposed to extend traveler information service with notification/alarm features. Extensive experiments have shown that the proposed algorithm for bus motion prediction is efficient enough to function in real conditions and that augmented with continuous query processing techniques can produce services that useful to the travelers.

Integration of GIS and video surveillance

ABSTRACT This paper discusses the integration of three-dimensional (3D) geographic information systems (GIS) and video surveillance systems using augmented reality (AR) techniques. The motivation for this integration is to overcome problems faced by conventional video surveillance systems. Explicit information concerning which camera currently monitors what area in such systems is missing; therefore, insight into the situation depends heavily on the operator’s training and experience. To ensure the complete coordination and monitoring of a situation in a system with multiple cameras, it is necessary to introduce a single reference system. GIS arises as a natural solution because it not only provides a solid ground truth but also provides semantic information that can be highly important in certain video surveillance applications. To integrate information into a GIS application, that information must be georeferenced. Based on our previous research regarding the addition of georeferencing information to surveillance video, this paper introduces models that can be applied to help integrate video and GIS. With an analogy to Milgram’s continuum between the real world and virtual reality, and analogous to the augmented reality and augmented virtuality in Milgram’s continuum, two models of integration are defined here: GIS-augmented video and video-augmented GIS. Then, we define the architecture of GIS-based video surveillance based on these proposed integration models, and finally, a prototype is implemented. The implemented prototype serves as a basis for analysing possible applications of real-world systems based on the integration of GIS and video.

Incremental evaluation of continuous range queries over objects moving on known network paths

In this paper we address the problem and propose the method for continuous range query processing for mobile objects moving on known network paths. The method assumes that the objects know their destination in advance and move along the best/shortest path to it. The method is based on an available 2D indexing scheme (e.g. R* Tree) for indexing transportation network data. The network R* tree is extended to provide matching of queries and objects according to their locations on the network for stationary objects/queries or their network routes for mobile objects/queries and performing the filter step of the continuous query. The refinement step of the query processing methodology generates main memory data structures that represent temporal query result and support periodic, incremental evaluation to produce result updates.

A Method for Estimating Surveillance Video Georeferences

The integration of a surveillance camera video with a three-dimensional (3D) geographic information system (GIS) requires the georeferencing of that video. Since a video consists of separate frames, each frame must be georeferenced. To georeference a video frame, we rely on the information about the camera view at the moment that the frame was captured. A camera view in 3D space is completely determined by the camera position, orientation, and field-of-view. Since the accurate measuring of these parameters can be extremely difficult, in this paper we propose a method for their estimation based on matching video frame coordinates of certain point features with their 3D geographic locations. To obtain these coordinates, we rely on high-resolution orthophotos and digital elevation models (DEM) of the area of interest. Once an adequate number of points are matched, Levenberg–Marquardt iterative optimization is applied to find the most suitable video frame georeference, i.e., position and orientation of the camera.

Collaborative mobile information system for collection and delivery of traffic information

This paper describes a collaborative mobile information system for collection and delivery of traffic information aimed to support fast, efficient and secure travel of people and transport of goods. The mobile information system provides organization of users into a specific social network dedicated to exchange of traffic information. Each user is able to provide information about traffic events on particular locations, as well as to receive information about traffic events which are collected by other participants in the network according to his/her current location, selected route, preferences and context. Based on that information the users are able to obtain re-routing instructions in order to avoid, or eventually visit the event for which they have received information.

Influence of the different input data volume, on the emergency vehicles relocation optimization process based on the AVL data

If we intent to realize the serious optimization task and if the main goal is the location and relocation of the ambulance vehicles based on the data archived in the history of the movement of these vehicles, at the very beginning we have to make several important decisions. One of the dilemmas is how far to go in the past, or how many destinations to which the vehicle traveled in the past we had to take in account and use as starting point of the analysis. In this paper we try to estimate influence of the different number of destinations to the quality of the optimization solution. For this purpose a specifically defined optimization coefficient is defined and introduced. We use the data extracted from the actual use of the emergency ambulance vehicles in real life and the optimization calculations are carried out using software module made during a live project implementation. The results obtained during this analysis of different scale of input data are illustrated and discussed in several aspects. Addition attention is directed to the most important analysis and that is the behavior and the comparison of the different optimization coefficient values.

Efficient replication of geospatial data for mobile GIS in field work

Replication of geospatial data for mobile GIS in field work is important because it allows field workers access to geospatial data even in situations when a server providing that data is unavailable. Due to limitations of mobile devices, it is important to create a specific data set for each specific task. In this way, mobile GIS can always maintain the highest level of performance. This paper presents a module for efficient replication of geospatial data, as part of a previously developed GIS architecture.