Seth Rogers

Mining GPS Traces for Map Refinement

Despite the increasing popularity of route guidance systems, current digital maps are still inadequate for many advanced applications in automotive safety and convenience. Among the drawbacks are the insufficient accuracy of road geometry and the lack of fine-grained information, such as lane positions and intersection structure. In this paper, we present an approach to induce high-precision maps from traces of vehicles equipped with differential GPS receivers. Since the cost of these systems is rapidly decreasing and wireless technology is advancing to provide the communication infrastructure, we expect that in the next few years large amounts of car data will be available inexpensively. Our approach consists of successive processing steps: individual vehicle trajectories are divided into road segments and intersections; a road centerline is derived for each segment; lane positions are determined by clustering the perpendicular offsets from it; and the transitions of traces between segments are utilized in the generation of intersection models. This paper describes an approach to this complex data-mining task in a contiguous manner. Among the new contributions are a spatial clustering algorithm for inferring the connectivity structure, more powerful lane finding algorithms that are able to handle lane splits and merges, and an approach to inferring detailed intersection models.

An adaptive interactive agent for route advice

In this paper, the authors describe the Adaptive Route Advisor system. It uses driver preferences and predicts routes based on a model of these driver preferences. If the predicted route is unsatisfactory the system generates additional routes that are based on driver interaction. The authors describes a pilot study using route selections to construct a personalized model. It is shown that as the accuracy of the preference model increases, the need for interaction decreases.

Mining GPS data to augment road models

Many advanced safety and navigation applications in vehicles require accurate, detailed digital maps, but manual lane measurements are expensive and time-consuming, making automated techniques desirable. This paper describes a data-mining approach to map refinement, using position traces that come from Global Positioning System receivers with differential corrections. The computed lane models enable safety applications, such as lanekeeping, and convenience applications, such as lane-changing advice. Experiments show that, starting from a baseline map that is commercially available, our lane models predict a vehicle’s lane with high accuracy from a small number of passes over a particular road segment. Multiple position traces are a powerful new source of data that enables cheap, automated methods of inducing lane models, as well as other geographic knowledge, like traffic signals and elevations, and potentially impacts any geographic information system with a need to relate to actual behavior.

Adaptive user interfaces for automotive environments

We investigate the use of adaptive user interfaces for in-car information access. These interfaces attempt to efficiently provide content the driver needs and wants, and gather feedback on these preferences through the drivers interaction with the system. In this way, the performance of the system improves as it unobtrusively builds a more accurate model of the user. The three systems presented here are: the Adaptive Route Advisor for navigation; the Adaptive News Reader for news stories; and the Adaptive Place Advisor for restaurant selection. All of these systems provide useful information to a driver, and we argue that they do not negatively impact safety because they are replacing other less effective, information sources. We intend to test this hypothesis in future studies.

Creating and evaluating highly accurate maps with probe vehicles

Many advanced safety and navigation applications in vehicles require accurate, detailed digital maps, but manual lane measurements are expensive and time-consuming, making automated techniques desirable. This paper describes a system that refines and augments commercial digital maps using position traces that come from Global Positioning System receivers with differential corrections. The computed lane models enable safety applications, such as lanekeeping, and convenience applications, such as lane-changing advice. We also present a number of evaluation metrics for estimating the quality of the enhanced maps we generate. Multiple position traces are a powerful new source of data that enables cheap, automated methods of inducing lane models, as well as other geographic knowledge, like traffic signals and elevations, and potentially impacts any geographic information system with a need to relate to actual behavior.

Development and evaluation of a curve rollover warning system for trucks

Truck accidents involving rollover have extremely high fatality rates. About half of all truck drivers killed in accidents last year were involved in a rollover. Freightliners Rollover Stability Advisor alerts drivers to potential rollover situations, training them to avoid excessive rollover risk In this paper, we describe work that extends the Rollover Stability Advisor to predict rollover situations in advance, giving drivers time to slow down before they are in danger. In an empirical evaluation of the rollover prediction accuracy on a sharp curve, we found that 50% of the dangerous passes could be predicted 5 seconds or longer in advance of the maximum rollover risk, and only 10% of the safe passes wrongly predicted a rollover risk. We feel that these are encouraging first results, and further refinement of the algorithms will yield even better results.