Indiana University-Purdue University Indianapolis (IUPUI) / The use of Global Positioning Systems (GPS) and/or accelerometers to identify trips and transportation modes such as walking, running, bicycling or motorized transportation has been an active goal in multiple disciplines such as Transportation Engineering, Computer Science, Informatics and Public Health. The purpose of this study was to review existing methods that determined trip and travel mode from raw Global Positioning System (GPS) and accelerometer data, and test a select group of these methods. The study had three specific aims: (1) Create a systematic review of existing literature that explored various methods for determining trip and travel mode from GPS and/or accelerometer data, (2) Collect a convenience sample of subjects who were assigned a GPS and accelerometer unit to wear while performing and logging travel bouts consisting of walking, running, bicycling and driving, (3) Replicate selected method designs extracted from the systematic review (aim 1) and use subject data (aim 2) to compare the methods. The results were be used to examine which methods are effective for various modes of travel.

5 pp. / Precision herbicide application is an effective tool for placing soil incorporated herbicides which have a tendency for soil adherence. And while field implementation depends on previous knowledge of soil textural variability (soil test and texture evaluations), site-specific technologies show promise for Arizona vegetable producers in non-uniform soils. Regardless of the method used for textural characterization, growers should keep in mind that textural differences do not change in the short/medium term, so the costs associated with defining texture-based management zones can be spread over many years.

In construction industry Gantt charts, network methods and Line of Balance (LOB) methods are generally used for planning and control of projects. Networking techniques such as Critical Path Method (CPM) and Program Evaluation and Review Technique (PERT) require technical knowledge and could be difficult to implement without proper scheduling background. Gantt charts and LOB techniques are usually easier to implement, however, similar to network techniques these techniques lack visualization. A graphical based scheduling and progress control method could improve existing techniques with visualization so that schedule information could be understood easily by the project participants. Geographic Information Systems (GIS) can be helpful for improving the traditional scheduling methods by utilizing spatial information. In this study, a prototype Global Positioning System (GPS) and GIS based LOB method is proposed. To illustrate the benefits, the proposed method will be applied to a pipeline project.

Recurrent congestion in urban traffic networks, especially on arterials, is a growing problem. Non-recurrent congestion, mainly due to incidents, only aggravates the problem. Any solution requires monitoring of the network, for which many developing countries, such as Turkey, do not have the traditional surveillance systems on arterials mainly due to high costs. An alternative solution is the utilization of Global Positioning System (GPS) technology, which is increasingly used in traffic monitoring. It is easy and cheap to obtain the GPS track information,even in real-time, from a probe-vehicle or a fleet of vehicles / and spatial variation of speed and travel time of the vehicle(s) in a network can be determined. GPS-based data, especially with only one probe-vehicle, would not provide information on the concurrent states of upstream and downstream traffic, needed to define the state of traffic in a network. To overcome this obstacle, a methodology based on statistical analysis of archival traffic conditions obtained through different sources is proposed to analyze traffic fluctuations and identify daily traffic pattern. As a result, bottleneck and resulting queues can be detected on a corridor. Thus, it enables detection of recurrent congestion and queues that may result from incidents. The proposed methodology is tested on a corridor the roadway between METU and Kizilay of in&ouml / n&uuml / Boulevard. The results show that the methodology can effectively identify bottleneck locations on the corridor and also an incident observed during the data collection is detected correctly by the proposed algorithm.

There is a growing recognition of the role that walking and cycling can make in reducing traffic congestion and air pollution whilst also contributing to improved personal health and wellbeing. While studies suggest that infrastructure is required to promote walking and cycling, there is a lack of evidence at the micro level on how interventions aimed at improving connectivity for walking and cycling influence travel behaviour and whether they promote a modal shift away from short car journeys. The aim of this study was to investigate the extent to which the implementation of a high quality traffic free route, delivered by a recent programme targeted at everyday walking and cycling in the UK - the Sustrans Connect2 Programme - influenced individuals' day-to-day travel decisions, changed the spatial and temporal nature of their journeys and impacted on overall carbon emissions from motorised travel. To achieve this aim an in-depth longitudinal panel study of a community of residents living next to a totemic Connect2 scheme in Penarth, Cardiff, was conducted. A panel of purposively selected participants (N=50) were interviewed and asked to record their travel behaviour using personal GPS devices and travel diaries over two seasonally matching 7-day time periods in 2011 and 2012. This novel GPS based mixed-method approach provided a detailed account of participants' travel behaviour in the local area (n=2664 journeys) and a comprehensive understanding of how, why and for whom the Connect2 intervention was likely to influence travel behaviour and the longevity of effects. The findings revealed that participants used the new Connect2 scheme regularly during the period of the study (36% in 2011; 26% in 2012); however, the new scheme was likely to have a greater impact for recreational journeys rather than for everyday travel. Spatial data provided new insights into the complexities of walking behaviour and factors influencing cycling for everyday travel or recreation, including route choice decisions, destinations where activities were conducted and the role of the new Connect2 infrastructure in supporting this. Further findings support the potential of active travel in replacing short car trips (20%) and its impact on carbon emissions from personal travel (4.9% among the study sample). However, results suggest that the new Connect2 scheme alone was unlikely to promote a significant change in travel behaviour and carbon emissions from (displaced) car journeys. The study contributes to the debate on the effectiveness of interventions targeted at promoting walking and cycling and the importance of wider infrastructural improvements that may be required to encourage their wider uptake. The combination of methods for data collection developed and employed in this study also helps to inform future travel behaviour research.

The examined area is related to previously mapped localities in the Novohradské Mountains and the foothills and creates a comprehensive insight into the landscape, not only from the geomorphological point of view. The goal is to create a geomorphological map at a scale of 1:25000, and some more detailed plans of the most interesting localities, using GPS technology. The maps were processed using ArcGIS 9.1 software and ZABAGED maps. The text section of the thesis contains a characteristics of various relief forms in the studied locations. Furthermore, an overall analysis of the physical-geographical elements that influence the landscape character was elaborated. The fieldwork was executed in 2010 and 2011.

In dieser Arbeit wird demonstriert, wie die Messreihen eines in einem Triebwagen eingesetzten GPS-Empfangsgerätes genutzt werden können, um Informationen zu Bahninfrastruktur und Bahnbetrieb im betrachteten Netz zu ermitteln. Dazu wird das relationale Datenbanksystem PostgreSQL mit dessen Erweiterung PostGIS eingesetzt. Im ersten Teil werden geografische Positionen von Bahnhöfen, Haltepunkten, Haupt-, Vor- und Langsamfahrsignalen, sowie deren signalisierte Geschwindigkeitsbeschränkungen berechnet. Es werden Algorithmen vorgestellt, die die Positionsberechnung von regelmäßig angefahrenen Bahnhöfen und Haltepunkten ebenso ermöglichen, wie die von Haupt- und Langsamfahrsignalen. Es wird gezeigt, dass mit den gewählten Ansätzen die Berechnung der Vorsignalpositionen nicht möglich ist. Darüber hinaus werden Algorithmen zur Berechnung von Halte- und Fahrzeiten an bzw. zwischen den vorher errechneten Betriebsstellen erläutert. Die gewonnenen Informationen werden im letzten Teil in einem Algorithmus genutzt, um Ankunftsprognosen einzelner Fahrten an beliebigen geografischen Stellen der Strecke erstellen zu können. GPS-Daten unterliegen verschiedenen Ungenauigkeiten, die betrachtet werden müssen, um aussagekräftige Ergebnisse liefern zu können. In der Arbeit wird deshalb ausführlich auf die Messungenauigkeiten und Messfehler der betrachteten Daten eingegangen. Ausnahmefall- und Sonderfallbetrachtungen und -behandlungen machen einen großen Teil der Lösungsentwicklungen aus.:1 Einleitung 1.1 Zielsetzung 1.1.1 Aufgabenstellung 1.1.2 Anwendungsfälle 1.2 Datenquelle 1.3 Methodik und Aufbau 1.4 Beschreibung der verwendeten Systemumgebung 2 Begriffsklärungen 3 Vergleichbare Arbeiten 3.1 GNSS zur verbesserten Echtzeitlokalisierung von Fahrzeugen 3.2 Einsatz von digitalen Karten und GIS zusätzlich zu GNSS 3.3 Vermessung von Gleisstrecken mittels GNSS 3.4 Einordnung der Arbeit 4 Beschreibung und Vorbereitung der Rohdaten 4.1 Aufbau der GPRMC-Rohdaten 4.2 Konvertierung der Rohdaten 4.2.1 Vorbereitung der Logdateien 4.2.2 Import der Logdatei 4.2.3 Anlegen von Indexes 4.2.4 Materialisierte Sichten 5 Clustering der Daten 23 5.1 Temporaler Schnitt der Daten 5.2 Ansatz 1: k-Means-Clustering 5.2.1 Definition 5.2.2 Berechnung der Cluster 5.2.3 Fazit 5.3 Ansatz 2: Dichtebasiertes Clustering 5.3.1 Definition 5.3.2 Fazit 5.4 Ansatz 3: Gruppieren nach Koordinatenwerten - Rasterbasiertes Clustering 5.4.1 Berechnung der Cluster 5.4.2 Vorteile des Verfahrens 5.4.3 Nachteile des Verfahrens 5.4.4 Fazit 5.5 Ansatz 4: Snap-To-Track 5.5.1 Vorbereitung der Daten 5.5.2 Clustering 5.5.3 Vorteile des Verfahrens 5.5.4 Nachteile des Verfahrens 5.5.5 Fazit 5.6 Vergleich: Rasterbasiertes Clustering und Snap-To-Track 5.6.1 Fazit 6 Auswertungen mit rasterbasierten Clustern 6.1 Informationen aus Durchschnittsgeschwindigkeiten 6.2 Informationen aus Maximalgeschwindigkeiten 6.3 Informationen aus Standardabweichung der Geschwindigkeiten 7 Kontextabhängige Optimierungen der Datenmenge 7.1 Reduzierung von Datensätzen mit 0 km/h-Messungen 7.2 Zusammenführung mehrerer Messreihen 7.2.1 Zusammenführung von zwei Messreihen 7.2.2 Zusammenführung von mehr als zwei Messreihen 7.2.3 Fazit 8 Berechnung von Infrastruktur 8.1 Kriterien für die zu ermittelnde Infrastruktur 8.2 Regelmäßig bediente Betriebsstellen 8.2.1 Abfrage von Referenzdaten 8.2.2 Grundlage: Rasterbasiertes Clustering 8.2.2.1 Ergebnisse 8.2.3 Grundlage: Snap-To-Track-Clustering 8.2.3.1 Ergebnisse und Nachbesserungen 8.2.4 Vergleich beider Verfahren 8.2.5 Fazit 8.3 Berechnung der Ausrichtung der Betriebsstellen 8.3.1 Vergleichsfunktion für Richtungsangaben 8.3.2 Fazit 8.4 Berechnung von Haupt- und Vorsignalen, sowie Geschwindkeitsbeschränkungen 8.4.1 Berechnung von Haupt- und Vorsignalen für eine einzelne Fahrt 8.4.1.1 Beschreibung des Algorithmus für Signale 8.4.1.2 Umsetzung des Algorithmus für Signale als Datenbankabfrage 8.4.1.3 Beschreibung des Algorithmus für Geschwindigkeitsbeschränkungen 8.4.1.4 Einordnung des Algorithmus und Fazit 8.4.2 Berechnung von Haupt- und Vorsignalen für alle Fahrten 8.4.2.1 Berechnung der Positionen der Hauptsignale 8.4.2.2 Fazit 8.4.3 Berechnung von Geschwindigkeitsbeschränkungen für alle Fahrten 8.4.3.1 Beschreibung des Algorithmus 8.4.3.2 Umsetzung des Algorithmus als Datenbankabfrage 8.4.3.3 Fazit 9 Berechnung von betrieblichen Aspekten 9.1 Haltezeiten an Betriebsstellen 9.1.1 Auswirkungen der verwendeten Clusteringverfahren und Daten auf die Haltezeitberechnungen 9.1.2 Fazit 9.2 Fahrzeit zwischen Betriebsstellen 9.2.1 Fazit 9.3 Berechnung der Entfernungen zwischen Betriebsstellen 9.3.1 Beschreibung des Algorithmus 9.3.2 Umsetzung des Algorithmus als Datenbankabfrage 9.3.3 Fazit 9.4 Einordnung einzelner Fahrten 9.4.1 Beschreibung des Algorithmus 9.4.2 Umsetzung des Algorithmus als Datenbankabfrage 9.4.3 Beschreibung eines genaueren Algorithmus 9.4.4 Umsetzung des genaueren Algorithmus als Datenbankabfrage 9.4.5 Fazit 9.5 Ankunftsprognosen 9.5.1 Umsetzung des Algorithmus als Datenbankabfrage 9.5.2 Fazit 10 Ausblick 10.1 Offene Fragen der Arbeit 10.1.1 Unregelmäßig oder nur in eine Richtung bediente Haltepunkte, Bedarfshaltepunkte 10.1.2 Nutzung der Standardabweichungen für Geschwindigkeitswerte 10.1.3 Verifikation der berechneten Signalstandorte 10.1.4 Anpassung der gespeicherten Werte 10.1.5 Gleisscharfe Abfragen 10.2 Verbesserungen der demonstrierten Abfragen 10.2.1 Optimierung der Abfrageparameter bzgl. der Genauigkeit 10.2.2 Verbesserung der Fehlertoleranz 10.3 Einbeziehung anderer Forschungsarbeiten 10.4 Big Data, Machine Learning 11 Zusammenfassung und Fazit 11.1 Zusammenfassung der Kapitel 11.2 Fazit der Arbeit Anhang / In this thesis, it is demonstrated how the measurement series of a GPS receiver of a railcar can be used to determine information on railway infrastructure and operations in the related network. For this purpose the relational database system PostgreSQL with its extension PostGIS is used. In the first part, geographic positions of stations, stops, main and approach signals as well as signalized speed limits are calculated. Algorithms are presented that allow the calculation of the positions of regularly served stations, stops and signals. It is shown that the calculation of positions of approach signals is not possible with the selected algorithms. Furthermore, algorithms for the calculation of stopping and running times at or between the previously calculated operating points are explained. The information obtained is used in the last part in an algorithm to be able to generate arrival forecasts of specific trips at arbitrary geographical on-track locations. GPS data are subject to various inaccuracies that must be considered in order to provide meaningful results. The thesis especially deals with the analysis of measurement data regarding their inaccuracies and measurement errors. Therefore, exception and edge case considerations and treatments are a large part of the process of developing appropriate solutions.:1 Einleitung 1.1 Zielsetzung 1.1.1 Aufgabenstellung 1.1.2 Anwendungsfälle 1.2 Datenquelle 1.3 Methodik und Aufbau 1.4 Beschreibung der verwendeten Systemumgebung 2 Begriffsklärungen 3 Vergleichbare Arbeiten 3.1 GNSS zur verbesserten Echtzeitlokalisierung von Fahrzeugen 3.2 Einsatz von digitalen Karten und GIS zusätzlich zu GNSS 3.3 Vermessung von Gleisstrecken mittels GNSS 3.4 Einordnung der Arbeit 4 Beschreibung und Vorbereitung der Rohdaten 4.1 Aufbau der GPRMC-Rohdaten 4.2 Konvertierung der Rohdaten 4.2.1 Vorbereitung der Logdateien 4.2.2 Import der Logdatei 4.2.3 Anlegen von Indexes 4.2.4 Materialisierte Sichten 5 Clustering der Daten 23 5.1 Temporaler Schnitt der Daten 5.2 Ansatz 1: k-Means-Clustering 5.2.1 Definition 5.2.2 Berechnung der Cluster 5.2.3 Fazit 5.3 Ansatz 2: Dichtebasiertes Clustering 5.3.1 Definition 5.3.2 Fazit 5.4 Ansatz 3: Gruppieren nach Koordinatenwerten - Rasterbasiertes Clustering 5.4.1 Berechnung der Cluster 5.4.2 Vorteile des Verfahrens 5.4.3 Nachteile des Verfahrens 5.4.4 Fazit 5.5 Ansatz 4: Snap-To-Track 5.5.1 Vorbereitung der Daten 5.5.2 Clustering 5.5.3 Vorteile des Verfahrens 5.5.4 Nachteile des Verfahrens 5.5.5 Fazit 5.6 Vergleich: Rasterbasiertes Clustering und Snap-To-Track 5.6.1 Fazit 6 Auswertungen mit rasterbasierten Clustern 6.1 Informationen aus Durchschnittsgeschwindigkeiten 6.2 Informationen aus Maximalgeschwindigkeiten 6.3 Informationen aus Standardabweichung der Geschwindigkeiten 7 Kontextabhängige Optimierungen der Datenmenge 7.1 Reduzierung von Datensätzen mit 0 km/h-Messungen 7.2 Zusammenführung mehrerer Messreihen 7.2.1 Zusammenführung von zwei Messreihen 7.2.2 Zusammenführung von mehr als zwei Messreihen 7.2.3 Fazit 8 Berechnung von Infrastruktur 8.1 Kriterien für die zu ermittelnde Infrastruktur 8.2 Regelmäßig bediente Betriebsstellen 8.2.1 Abfrage von Referenzdaten 8.2.2 Grundlage: Rasterbasiertes Clustering 8.2.2.1 Ergebnisse 8.2.3 Grundlage: Snap-To-Track-Clustering 8.2.3.1 Ergebnisse und Nachbesserungen 8.2.4 Vergleich beider Verfahren 8.2.5 Fazit 8.3 Berechnung der Ausrichtung der Betriebsstellen 8.3.1 Vergleichsfunktion für Richtungsangaben 8.3.2 Fazit 8.4 Berechnung von Haupt- und Vorsignalen, sowie Geschwindkeitsbeschränkungen 8.4.1 Berechnung von Haupt- und Vorsignalen für eine einzelne Fahrt 8.4.1.1 Beschreibung des Algorithmus für Signale 8.4.1.2 Umsetzung des Algorithmus für Signale als Datenbankabfrage 8.4.1.3 Beschreibung des Algorithmus für Geschwindigkeitsbeschränkungen 8.4.1.4 Einordnung des Algorithmus und Fazit 8.4.2 Berechnung von Haupt- und Vorsignalen für alle Fahrten 8.4.2.1 Berechnung der Positionen der Hauptsignale 8.4.2.2 Fazit 8.4.3 Berechnung von Geschwindigkeitsbeschränkungen für alle Fahrten 8.4.3.1 Beschreibung des Algorithmus 8.4.3.2 Umsetzung des Algorithmus als Datenbankabfrage 8.4.3.3 Fazit 9 Berechnung von betrieblichen Aspekten 9.1 Haltezeiten an Betriebsstellen 9.1.1 Auswirkungen der verwendeten Clusteringverfahren und Daten auf die Haltezeitberechnungen 9.1.2 Fazit 9.2 Fahrzeit zwischen Betriebsstellen 9.2.1 Fazit 9.3 Berechnung der Entfernungen zwischen Betriebsstellen 9.3.1 Beschreibung des Algorithmus 9.3.2 Umsetzung des Algorithmus als Datenbankabfrage 9.3.3 Fazit 9.4 Einordnung einzelner Fahrten 9.4.1 Beschreibung des Algorithmus 9.4.2 Umsetzung des Algorithmus als Datenbankabfrage 9.4.3 Beschreibung eines genaueren Algorithmus 9.4.4 Umsetzung des genaueren Algorithmus als Datenbankabfrage 9.4.5 Fazit 9.5 Ankunftsprognosen 9.5.1 Umsetzung des Algorithmus als Datenbankabfrage 9.5.2 Fazit 10 Ausblick 10.1 Offene Fragen der Arbeit 10.1.1 Unregelmäßig oder nur in eine Richtung bediente Haltepunkte, Bedarfshaltepunkte 10.1.2 Nutzung der Standardabweichungen für Geschwindigkeitswerte 10.1.3 Verifikation der berechneten Signalstandorte 10.1.4 Anpassung der gespeicherten Werte 10.1.5 Gleisscharfe Abfragen 10.2 Verbesserungen der demonstrierten Abfragen 10.2.1 Optimierung der Abfrageparameter bzgl. der Genauigkeit 10.2.2 Verbesserung der Fehlertoleranz 10.3 Einbeziehung anderer Forschungsarbeiten 10.4 Big Data, Machine Learning 11 Zusammenfassung und Fazit 11.1 Zusammenfassung der Kapitel 11.2 Fazit der Arbeit Anhang

Soil landscape characterization into landform elements for precision agriculture has become an important issue. As soil properties and crop yields change over the landscape, delineating landform elements as a basis for site-specific application of crop inputs has become a reality. Two different methods of delineating landform elements from agricultural fields were tested and compared. The first method delineated landform elements from digital elevation maps with the use of the LandMapR(tm) software, the second method delineated classes from IKONOS high resolution panchromatic images using an unsupervised classification algorithm. The LandMapR(tm) model delineated landform elements from true elevation data collected in the field and was considered the reference dataset to which the image classification maps were compared to. The IKONOS imagery was processed using a combination of one filtering algorithm and one unsupervised classification method prior to being compared to the classified DEM. A total of 20 filtering algorithms and two unsupervised methods were used for each of the five study sites. The study sites consisted of four agricultural fields covered with crop stubble and one field in summer fallow. Image classification accuracy assessment was reported as overall, producer’s and user’s accuracy as well as Kappa statistic. Results showed that filtering algorithms and classification methods had no effects on image classification accuracies. Highest classification accuracy of image map to landform element map comparison achieved for all study sites was 17.9 %. Classification accuracy was affected by the heterogeneity of the ground surface cover found in each field. However, the classification accuracy of the fallow field was not superior to the stubble fields. / May 2006

Soil landscape characterization into landform elements for precision agriculture has become an important issue. As soil properties and crop yields change over the landscape, delineating landform elements as a basis for site-specific application of crop inputs has become a reality. Two different methods of delineating landform elements from agricultural fields were tested and compared. The first method delineated landform elements from digital elevation maps with the use of the LandMapR(tm) software, the second method delineated classes from IKONOS high resolution panchromatic images using an unsupervised classification algorithm. The LandMapR(tm) model delineated landform elements from true elevation data collected in the field and was considered the reference dataset to which the image classification maps were compared to. The IKONOS imagery was processed using a combination of one filtering algorithm and one unsupervised classification method prior to being compared to the classified DEM. A total of 20 filtering algorithms and two unsupervised methods were used for each of the five study sites. The study sites consisted of four agricultural fields covered with crop stubble and one field in summer fallow. Image classification accuracy assessment was reported as overall, producer’s and user’s accuracy as well as Kappa statistic. Results showed that filtering algorithms and classification methods had no effects on image classification accuracies. Highest classification accuracy of image map to landform element map comparison achieved for all study sites was 17.9 %. Classification accuracy was affected by the heterogeneity of the ground surface cover found in each field. However, the classification accuracy of the fallow field was not superior to the stubble fields.

Soil landscape characterization into landform elements for precision agriculture has become an important issue. As soil properties and crop yields change over the landscape, delineating landform elements as a basis for site-specific application of crop inputs has become a reality. Two different methods of delineating landform elements from agricultural fields were tested and compared. The first method delineated landform elements from digital elevation maps with the use of the LandMapR(tm) software, the second method delineated classes from IKONOS high resolution panchromatic images using an unsupervised classification algorithm. The LandMapR(tm) model delineated landform elements from true elevation data collected in the field and was considered the reference dataset to which the image classification maps were compared to. The IKONOS imagery was processed using a combination of one filtering algorithm and one unsupervised classification method prior to being compared to the classified DEM. A total of 20 filtering algorithms and two unsupervised methods were used for each of the five study sites. The study sites consisted of four agricultural fields covered with crop stubble and one field in summer fallow. Image classification accuracy assessment was reported as overall, producer’s and user’s accuracy as well as Kappa statistic. Results showed that filtering algorithms and classification methods had no effects on image classification accuracies. Highest classification accuracy of image map to landform element map comparison achieved for all study sites was 17.9 %. Classification accuracy was affected by the heterogeneity of the ground surface cover found in each field. However, the classification accuracy of the fallow field was not superior to the stubble fields.