An Automated Analysis Framework for Trajectory Datasets

@article{Glasmacher2022AnAA,
  title={An Automated Analysis Framework for Trajectory Datasets},
  author={Christoph Glasmacher and Robert Krajewski and Lutz Eckstein},
  journal={ArXiv},
  year={2022},
  volume={abs/2202.07438}
}
Trajectory datasets of road users have become more important in the last years for safety validation of highly automated vehicles. Several naturalistic trajectory datasets with each more than 10 000 tracks were released and others will follow. Considering this amount of data, it is necessary to be able to compare these datasets in-depth with ease to get an overview. By now, the datasets’ own provided information is mainly limited to metadata and qualitative descriptions which are mostly not… 
MEAT: Maneuver Extraction from Agent Trajectories
TLDR
An automated methodology is proposed that allows to extract maneuvers from agent trajectories in large-scale dataset analysis and maneuver-specific evaluation of multiple state-of-the-art trajectory prediction models.

References

SHOWING 1-10 OF 32 REFERENCES
The rounD Dataset: A Drone Dataset of Road User Trajectories at Roundabouts in Germany
TLDR
This work presents a new dataset of road user trajectories at roundabouts in Germany using a camera-equipped drone and provides source code for parsing and visualizing the dataset as well as maps of the recording sites.
The highD Dataset: A Drone Dataset of Naturalistic Vehicle Trajectories on German Highways for Validation of Highly Automated Driving Systems
TLDR
This work proposes a novel method to measure data from an aerial perspective for scenario-based validation fulfilling the mentioned requirements, and provides a large-scale naturalistic vehicle trajectory dataset from German highways called highD.
The inD Dataset: A Drone Dataset of Naturalistic Road User Trajectories at German Intersections
TLDR
A comprehensive, large-scale urban intersection dataset with naturalistic road user behavior using camera-equipped drones as successor of the highD dataset, called inD is created and is available online for non-commercial research.
A framework for anomaly detection in maritime trajectory behavior
  • Po-Ruey Lei
  • Computer Science
    Knowledge and Information Systems
  • 2015
TLDR
An anomaly detection algorithm is developed based on a proposed framework for maritime trajectory modeling and anomaly detection, called MT-MAD, in which an indicator is used to evaluate suspicious behavior and scores trajectory behavior according to the defined outlying features.
Anomaly Detection in Trajectory Data with Normalizing Flows
TLDR
The promising results obtained in the performed computational experiments indicate the feasibility of the GRADINGS, specially the variant that considers autoregressive normalizing flows, which provides an unsupervised procedure to identify out of distribution samples.
Vehicle trajectory clustering for traffic intersection surveillance
TLDR
Clustering on the vehicle trajectory dataset is simulated to group the data via similarity function based on Longest Common Subsequence (LCSS) to evaluate the clustering result, Rand Index is implemented to compare the simulated result with the ground-truth result.
Constructing a Highly Interactive Vehicle Motion Dataset
TLDR
A dataset with highly interactive vehicle motions is constructed and a set of metrics to extract the interactive motions in different maps, including the minimum difference of time to collision point (MDTTC) and duration of waiting period are proposed.
Anomaly Detection Based on the Global-Local Anomaly Score for Trajectory Data
TLDR
A new abnormal score, Global-Local Anomaly Score (GLAS) is proposed, to sensitively quantify the abnormal degree of trajectories with varying local densities, and an effective and fast measure, Extended Power enhanced Euclidean Distance (EPED) is designed to calculate the trajectory similarity.
Top-view Trajectories: A Pedestrian Dataset of Vehicle-Crowd Interaction from Controlled Experiments and Crowded Campus
TLDR
This study proposed two pedestrian trajectory datasets, CITR dataset and DUT dataset, so that the pedestrian motion models can be further calibrated and verified, especially when vehicle influence on pedestrians plays an important role.
...
...