In this report we attempt to provide insights to how a resource efficient solution for transportation mode recognition can be implemented on a smartphone using the accelerometer and magnetometer as sensors for data collection. The proposed system uses a hierarchical classification process where instances are first classified as vehicles or non-vehicles, then as wheel or rail vehicles, and lastly as belonging to one of the transportation modes: bus, car, motorcycle, subway, or train. A virtual gyroscope is implemented as a low-power source of simulated gyroscope data. Features are extracted from the accelerometer, magnetometer and virtual gyroscope readings that are sampled at 30 Hz, before they are classified using machine learning algorithms from the WEKA machine learning library. An Android application was developed to classify real-time data, and the resource consumption of the application was measured using the Trepn profiler application. The proposed system achieves an overall accuracy of 82.7% and a vehicular accuracy of 84.9% using a 5 second window with 75% overlap while having an average power consumption of 8.5 mW. / I denna rapport försöker vi ge insikter om hur en resurseffektiv lösning för transportlägesigenkänning kan implementeras på en smartphone genom att använda accelerometern och magnetometern som sensorer för datainsamling. Det föreslagna systemet använder en hierarkisk klassificeringsprocess där instanser först klassificeras som fordon eller icke-fordon, sedan som hjul- eller järnvägsfordon, och slutligen som tillhörande ett av transportsätten: buss, bil, motorcykel, tunnelbana eller tåg. Ett virtuellt gyroskop implementeras som en lågenergi källa till simulerad gyroskopdata. Olika särdrag extraheras från accelerometer, magnetometer och virtuella gyroskopläsningar som samlas in vid 30 Hz, innan de klassificeras med hjälp av maskininlärningsalgoritmer från WEKA-maskinlärningsbiblioteket. En Android-applikation har utvecklats för att klassificera realtidsdata, och programmets resursförbrukning mättes med hjälp av Trepn profiler-applikationen. Det föreslagna systemet uppnår en övergripande noggrannhet av 82.7% och en fordonsnoggrannhet av 84.9% genom att använda ett 5 sekunders fönster med 75% överlappning med en genomsnittlig energiförbrukning av 8.5 mW.

The focus of this dissertation is on safety improvement at intersections and presenting how Vehicle/Bicycle-to-Infrastructure Communications can be a potential countermeasure for crashes resulting from drivers' and cyclists' violations at intersections. The characteristics (e.g., acceleration capabilities, etc.) of transportation modes affect the violation behavior. Therefore, the first building block is to identify the users' transportation mode. Consequently, having the mode information, the second building block is to predict whether or not the user is going to violate. This step focuses on two different modes (i.e., driver violation prediction and cyclist violation prediction). Warnings can then be issued for users in potential danger to react or for the infrastructure and vehicles so they can take appropriate actions to avoid or mitigate crashes. A smartphone application was developed to collect sensor data used to conduct the transportation mode recognition task. Driver violation prediction task at signalized intersections was conducted using observational and simulator data. Also, a naturalistic cycling experiment was designed for cyclist violation prediction task. Subsequently, cyclist violation behavior was investigated at both signalized and stop-controlled intersections. To build the prediction models in all the aforementioned tasks, various Artificial Intelligence techniques were adopted. K-fold Cross-Validation as well as Out-of-Bag error was used for model selection and validation. Transportation mode recognition models contributed to high classification accuracies (e.g., up to 98%). Thus, data obtained from the smartphone sensors were found to provide important information to distinguish between transportation modes. Driver violation (i.e., red light running) prediction models were resulted in high accuracies (i.e., up to 99.9%). Time to intersection (TTI), distance to intersection (DTI), the required deceleration parameter (RDP), and velocity at the onset of a yellow light were among the most important factors in violation prediction. Based on logistic regression analysis, movement type and presence of other users were found as significant factors affecting the probability of red light violations by cyclists at signalized intersections. Also, presence of other road users and age were the significant factors affecting violations at stop-controlled intersections. In case of stop-controlled intersections, violation prediction models resulted in error rates of 0 to 10 percent depending on how far from the intersection the prediction task is conducted. / Ph. D.

Traffic congestion has become one of the major challenging problems of modern life in many urban areas. This growing problem leads to negative environmental impacts, wasted fuel, lost productivity, and increased travel time. In big cities, trains and buses bring riders to transit stations near shopping and employment centers, but riders then need another transportation mode to reach their final destination, which is known as the last mile problem. A smart bike-sharing system (BSS) can help address this problem and encourage more people to ride public transportation, thus relieving traffic congestion. At the strategic level, we start with proposing a novel two-layer hierarchical classifier that increases the accuracy of traditional transportation mode classification algorithms. In the transportation sector, researchers can use smartphones to track and obtain information of multi-mode trips. These data can be used to recognize the user's transportation mode, which can be then utilized in several different applications; such as planning new BSS instead of using costly surveys. Next, a new method is proposed to quantify the effect of several factors such as weather conditions on the prediction of bike counts at each station. The proposed approach is promising to quantify the effect of various features on BSSs in cases of large networks with big data. Third, these resulted significant features were used to develop state-of-the-art toolbox algorithms to operate BSSs efficiently at two levels: network and station. Finally, we proposed a quality-of-service (QoS) measurement, namely Optimal Occupancy, which considers the impact of inhomogeneity in a BSS. We used one of toolbox algorithms modeled earlier to estimate the proposed QoS. Results revealed that the Optimal Occupancy is beneficial and outperforms the traditionally-known QoS measurement. / PHD / A growing population, with more people living in cities, has led to increased pollution, noise, congestion, and greenhouse gas emissions. One possible approach to mitigating these problems is encouraging the use of bike-sharing systems (BSSs). BSSs are an integral part of urban mobility in many cities and are sustainable and environmentally friendly. As urban density increases, it is likely that more BSSs will appear due to their relatively low capital and operational costs, ease of installation, pedal assistance for people who are physically unable to pedal for long distances or on difficult terrain, and the ability to track bikes in some cases. This dissertation is a building block for a smart BSS in the strategic level, which could be used in real and different applications. The main aims of the dissertation are to boost the redistribution operation, to gain new insights into and correlations between bike demand and other factors, and to support policy makers and operators in making good decisions regarding planning new or existing BSS. This dissertation makes many significant contributions. These contributions include novel methods, measurements, and applications using machine learning and statistical learning techniques in order to design a smart BSS. We start with proposing a novel framework that increases the accuracy of traditional transportation mode classification algorithms. In the transportation sector, researchers can use smartphones to track and obtain information of multi-mode trips. These data can be used to recognize the user’s transportation mode, which can be then used in planning new BSS. Next, a new method is proposed to quantify the effect of several factors such as weather conditions on the prediction of bike station counts. Third, we use state-of-the-art data analytics to develop a toolbox to operate BSSs efficiently at two levels: network and station. Finally, we propose a quality-of-service (QoS) measurement, which considers the impact of inhomogeneity of BSS properties.

Aujourd'hui, la plupart des internautes ont leurs données dispersées dans plusieurs appareils, applications et services. La gestion et le contrôle de ses données sont de plus en plus difficiles. Dans cette thèse, nous adoptons le point de vue selon lequel l'utilisateur devrait se voir donner les moyens de récupérer et d'intégrer ses données, sous son contrôle total. À ce titre, nous avons conçu un système logiciel qui intègre et enrichit les données d'un utilisateur à partir de plusieurs sources hétérogènes de données personnelles dans une base de connaissances RDF. Le logiciel est libre, et son architecture innovante facilite l'intégration de nouvelles sources de données et le développement de nouveaux modules pour inférer de nouvelles connaissances. Nous montrons tout d'abord comment l'activité de l'utilisateur peut être déduite des données des capteurs de son téléphone intelligent. Nous présentons un algorithme pour retrouver les points de séjour d'un utilisateur à partir de son historique de localisation. À l'aide de ces données et de données provenant d'autres capteurs de son téléphone, d'informations géographiques provenant d'OpenStreetMap, et des horaires de transports en commun, nous présentons un algorithme de reconnaissance du mode de transport capable de retrouver les différents modes et lignes empruntés par un utilisateur lors de ses déplacements. L'algorithme reconnaît l'itinéraire pris par l'utilisateur en retrouvant la séquence la plus probable dans un champ aléatoire conditionnel dont les probabilités se basent sur la sortie d'un réseau de neurones artificiels. Nous montrons également comment le système peut intégrer les données du courrier électronique, des calendriers, des carnets d'adresses, des réseaux sociaux et de l'historique de localisation de l'utilisateur dans un ensemble cohérent. Pour ce faire, le système utilise un algorithme de résolution d'entité pour retrouver l'ensemble des différents comptes utilisés par chaque contact de l'utilisateur, et effectue un alignement spatio-temporel pour relier chaque point de séjour à l'événement auquel il correspond dans le calendrier de l'utilisateur. Enfin, nous montrons qu'un tel système peut également être employé pour faire de la synchronisation multi-système/multi-appareil et pour pousser de nouvelles connaissances vers les sources. Les résultats d'expériences approfondies sont présentés. / Typical Internet users today have their data scattered over several devices, applications, and services. Managing and controlling one's data is increasingly difficult. In this thesis, we adopt the viewpoint that the user should be given the means to gather and integrate her data, under her full control. In that direction, we designed a system that integrates and enriches the data of a user from multiple heterogeneous sources of personal information into an RDF knowledge base. The system is open-source and implements a novel, extensible framework that facilitates the integration of new data sources and the development of new modules for deriving knowledge. We first show how user activity can be inferred from smartphone sensor data. We introduce a time-based clustering algorithm to extract stay points from location history data. Using data from additional mobile phone sensors, geographic information from OpenStreetMap, and public transportation schedules, we introduce a transportation mode recognition algorithm to derive the different modes and routes taken by the user when traveling. The algorithm derives the itinerary followed by the user by finding the most likely sequence in a linear-chain conditional random field whose feature functions are based on the output of a neural network. We also show how the system can integrate information from the user's email messages, calendars, address books, social network services, and location history into a coherent whole. To do so, it uses entity resolution to find the set of avatars used by each real-world contact and performs spatiotemporal alignment to connect each stay point with the event it corresponds to in the user's calendar. Finally, we show that such a system can also be used for multi-device and multi-system synchronization and allow knowledge to be pushed to the sources. We present extensive experiments.