Heidrun Belzner

A New Measure of Travel Time Reliability for In-Vehicle Navigation Systems

This article introduces a new measure of travel time reliability for implementation in the dynamic routing algorithm of an intelligent car navigation system. The measure is based on the log-normal distribution of travel time on a link and consists of two indices corresponding to the extreme values of the distribution, such that they reflect the shortest and longest travel times that may be experienced on the link. Through a series of mathematical manipulations, the indices are expressed in terms of the characteristic values of the speed distribution on the link. An expression relating the indices of a route and the indices of the individual links forming it is derived. The accuracy of the measure is then assessed through a field experiment and the results are presented.

Image based fog detection in vehicles

Modern vehicles are equipped with many cameras and their use in many practical applications is extensive. Detecting the presence of fog from images of a camera mounted in vehicles is a very challenging task with the potential to be used in many practical applications. Approaches introduced until now analyze properties of local objects in the image like lane markings, traffic signs, back lights of vehicles in front or head lights of approaching vehicles. By contrast to all these related works we propose to use image descriptors and a classification procedure in order to distinguish images with fog present from those free of fog. These image descriptors are global and describe the entire image using Gabor filters at different frequencies, scales and orientations. Our experiments demonstrated hight potential of the proposed method for fog detection on daytime images.

Scene understanding from a moving camera for object detection and free space estimation

Modern vehicles are equipped with multiple cameras which are already used in various practical applications. Advanced driver assistance systems (ADAS) are of particular interest because of the safety and comfort features they offer to the driver. Camera based scene understanding is an important scientific problem that has to be addressed in order to provide the information needed for camera based driver assistance systems. While frontal cameras are widely used, there are applications where cameras observing lateral space can deliver better results. Fish eye cameras mounted in the side mirrors are particularly interesting, because they can observe a big area on the side of the vehicle and can be used for several applications for which the traditional front facing cameras are not suitable. We present a general method for scene understanding using 3D reconstruction of the environment around the vehicle. It is based on pixel-wise image labeling using a conditional random field (CRF). Our method is able to create a simple 3D model of the scene and also to provide semantic labels of the different objects and areas in the image, like for example cars, sidewalks, and buildings. We demonstrate how our method can be used for two applications that are of high importance for various driver assistance systems - car detection and free space estimation. We show that our system is able to perform in real time for speeds of up to 63 km/h.

Seamless Traffic Information in Dynamic Navigation Up to Date

Preciseness and reliability of traffic data as an input for GPS navigation units systems means a competition relevant advantage for system provider. It is shown that many data sources for those systems are potentially available today. Most of these sources however either lack in penetration or do not provide sufficient network coverage. In conclusion this means that relying solely on a single data source will not be sufficient to build a high quality navigation solution. This paper gives an overview of potential data sources and elaborates their pros and cons. Relying on existing infrastructure and vehicle probe data “premium quality” can be achieved, on highways at least. This does not fulfil the customer’s expectance of a premium navigation system if arterials are not covered in the same quality as on highways. Significant reduction in travel time is the only valid criteria of quality of recommended changes in routing. Additionally real time traffic information and dynamic route guidance presumably leverages the highest potential especially on secondary road networks within large metropolitan areas. In order to continuously fulfil customer expectations BMW Group engineers collaborate with service providers and engage in various research projects to bring together multiple data sources and develop new data collection methods.

Soft Discretization in a Classification Model for Modeling Adaptive Route Choice with a Fuzzy ID3 Algorithm

This study introduces a way to overcome the sensitivity of decision trees used for route choice behavior studies by using fuzzy logic while preserving the advantages of decision trees and the C4.5 algorithm, namely, comprehensibility and ease of application. Soft discretization of continuous values in fuzzy decision trees can provide a more robust classification. Also, the use of fuzzy logic makes it possible to accommodate qualitative attributes describing route characteristics. Apart from these features, fuzzy decision tree learning algorithms are also capable of assigning numeric values on decisions about the degree of certainty of each recommendation emanating from the fuzzy reasoning. This feature makes it possible to solve the multiple suggestion problem whereby the classic decision tree may suggest that more than one route is optimal. To investigate improvements resulting from the application of fuzzy decision tree learning algorithms, software for an adaptive route choice model using a fuzzy decision tree learning algorithm, fuzzy ID3, was developed and simulation experiments with the model were carried out. The comparison of results with the nonfuzzy adaptive route choice model indicates better predictive accuracy and more effective applicability for the fuzzy model in practice.

Traffic information with seamless coverage

Dynamic navigation functions such as alternative route suggestions are only as good as the traffic information data on which they are based. In order to guarantee high product quality for its telematics services, BMW Group has been seeking new methods of gaining access to accurate traffic data across the entire road network.

Verkehrsinformationen Ohne Weisse Flecken

Dynamische Navigationsfunktionen wie Umleitungsempfehlungen konnen nur so gut sein wie ihre zugrunde liegenden Daten uber Verkehrsstorungen. Zur Sicherung der Produktqualitat ihrer Telematikdienste sah sich die BMW Group deshalb veranlasst, nach neuen Wegen zur netzweiten Verbesserung der Verkehrsdatenerfassung zu suchen.

A Hybrid Model for Forecasting Local Traffic Parameters

Abstract Predicting traffic is certainly one of the greatest challenges facing current traffic engineering research. The work presented here focuses on the prediction of local traffic parameters such as the local mean speed aggregated over short periods such as one minute. For this purpose a new hybrid model comprising an ARIMA 1 model and a neural network has been developed and validated with real traffic data. This model has been evaluated and compared with conventional methods of traffic prediction using independent real test data. It was found that the new hybrid model produces more accurate and more reliable results than conventional methods.