Search

Trajectory data mining is a lively research field in the domain of spatio-temporal data mining. Trajectory pattern mining comprises a set of specific pattern mining methods, which are applied as consecutive steps on a trajectory with the goal to extract and classify re-occurring spatio-temporal patterns. Despite the common nature and frequent usage of such methods by the GIScience community, a methodological approach is missing so far, especially when it comes to the use of machine learning-based classification methods. The current work closes this gap by proposing and evaluating a machine learning-based 3-steps trajectory data mining methodology using the detection and classification of stop points in vehicle trajectories as example. The work describes in detail the applied methodologies with respect to the three mining steps ‘stop detection’, ‘feature extraction’ and ‘classification in traffic-relevant and non-traffic-relevant stops’ and evaluates six machine learning-based classification algorithms using a real-world dataset of 15,498 vehicle trajectories with 5,899 detected stops (thereof 2,032 manually classified). Due to its exemplary nature, the presented methodology is suited to act as blueprint for similar trajectory data mining problems.

Automated driving systems are in need of accurate localization, i.e., achieving accuracies below 0.1&nbsp, m at confidence levels above 95%. Although during the last decade numerous localization techniques have been proposed, a common methodology to validate their accuracies in relation to a ground-truth dataset is missing so far. This work aims at closing this gap by evaluating four different methods for validating localization accuracies of a vehicle’s position trajectory to different ground truths: (1) a static driving-path, (2) the lane-centerline of a high-definition (HD) map with validated accuracy, (3) localized vehicle body overlaps of the lane-boundaries of a HD map, and (4) longitudinal accuracy at stop points. The methods are evaluated using two localization test datasets, one acquired by an automated vehicle following a static driving path, being additionally equipped with roof-mounted localization systems, and a second dataset acquired from manually-driven connected vehicles. Results show the broad applicability of the approach for evaluating localization accuracy and reveal the pros and cons of the different methods and ground truths. Results also show the feasibility of achieving localization accuracies below 0.1&nbsp, m at confidence levels up to 99.9% for high-quality localization systems, while at the same time demonstrate that such accuracies are still challenging to achieve.

In the context of intelligent transport systems, dynamic route planning is an eagerly researched topic. While the research community during the last decade has focused on performance and scalability of dynamic route planning algorithms, the question concerning route qualities has been widely neglected. The current work contributes to this question by proposing a methodology to assess the quality of arbitrary routes based on seven spatio‐temporal route quality metrics (five spatial and two temporal ones). The methodology is evaluated by calculating quality metrics for 45 routes being derived from three dynamic route planning systems for three selected origin–destination pairs during one day. For objectively assessing route qualities, the approach matches the route planning results to a reference digital road network and uses reference travel times for calculating quality metrics. A spatio‐temporal cross‐evaluation illustrates different quality aspects in the context of both quality dimensions and demonstrates the usefulness of the proposed approach.

Technological progress with regard to various sensors and mobile devices is constant. In the field of movement data analysis in particular, this has led to new opportunities thanks to data sources such as Global Positioning Systems (GPS). In recent years, many research groups have developed new approaches for analysing this data. Most of these approaches are computationally intensive and unable to deliver results in a reasonable time when run on a mobile device. This paper presents a light-weight approach, called the “cookie-cutter”, which follows an alternative path by using an Eulerian model to determine stays of individuals within reasonable computation time. The quality measures used in this work show that the approach is promising with regard to both accuracy and computing requirements.