Peer Kröger
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Peer Kröger has a total of 126 co-authors. Here are the top co-authors by number of shared publications:
- Daniyal Kazempour
Kiel University
13 shared publications - Matthias Renz
Kiel University
6 shared publications - Andrea Göhring
Cluster of Excellence ROOTS, Kiel University - Kiel, Germany
4 shared publications - Claudius Zelenka
Faculty of Engineering, Department of Computer Science Kiel University Kiel Germany
4 shared publications - Gisela Grupe
Ludwig-Maximilians-Universität
4 shared publications - Maximilian Hünemörder
Southern Denmark University
3 shared publications
Latest Publications
- 11 October 2026
- Published by Springer Nature
- p. 223-232
Capacity-Aware Cluster Rebalancing for Efficient R-Tree Indexing
- 11 October 2026
- Published by Springer Nature
- p. 223-232
Capacity-Aware Cluster Rebalancing for Efficient R-Tree Indexing
- 6 October 2026
- Published by Springer Nature
- p. 672-689
Building a Deep Temporal Emulator of an Ocean Carbon Uptake Model
- 6 October 2026
- Published by Springer Nature
- p. 672-689
Building a Deep Temporal Emulator of an Ocean Carbon Uptake Model
- Open Access
- 24 July 2026
- Published by Peer Community In in Peer Community Journal
- Vol. 6
Bridging Interdisciplinary Research Data Management and Data Science through Modular Research Processes
Transparent and reproducible research in large interdisciplinary projects, such as the Cluster of Excellence ROOTS, depends on combining two perspectives that are usually considered separately: the stewardship concerns of Research Data Management (RDM) and the analytical concerns of Data Science. This paper introduces Modular Research Processes (MRP), a framework representing a research endeavour as distinct, referable products and procedures described at several layers of granularity, expressing resolution rather than a hierarchy of authority. MRP provides a shared representation in which RDM and Data Science converge, focusing on the point where (materialised) products are FAIR-ready, while their procedural counterparts are not (yet) – a discrepancy that motivates the development of systematic descriptions and, where applicable, recommendations. We organise the framework’s value into three application fields: illustration and communication; documentation and organisation; as well as comparison and recommendation. Our examinations and activities are based on the Data Management and Data Science Platform (DMDSP), which is an organisational component of the ROOTS project. We present the DMDSP as an example of how such a bridge can be shaped, aligned with an existing ecosystem, filled with collaborative method development, and sustained through training. Beyond the platform, it is MRP that is the transferable contribution, with a claim to cross-disciplinary and cross-project application. We report on two archaeological analyses that have been reconstructed as first proofs of concept, techniques for early empirical feedback on the conditions for adoption of an MRP-related tool, and a research roadmap anchored in this beginning.
- 24 July 2026
- Published by Peer Community In in Peer Community Journal
- Vol. 6
Open AccessBridging Interdisciplinary Research Data Management and Data Science through Modular Research Processes
Transparent and reproducible research in large interdisciplinary projects, such as the Cluster of Excellence ROOTS, depends on combining two perspectives that are usually considered separately: the stewardship concerns of Research Data Management (RDM) and the analytical concerns of Data Science. This paper introduces Modular Research Processes (MRP), a framework representing a research endeavour as distinct, referable products and procedures described at several layers of granularity, expressing resolution rather than a hierarchy of authority. MRP provides a shared representation in which RDM and Data Science converge, focusing on the point where (materialised) products are FAIR-ready, while their procedural counterparts are not (yet) – a discrepancy that motivates the development of systematic descriptions and, where applicable, recommendations. We organise the framework’s value into three application fields: illustration and communication; documentation and organisation; as well as comparison and recommendation. Our examinations and activities are based on the Data Management and Data Science Platform (DMDSP), which is an organisational component of the ROOTS project. We present the DMDSP as an example of how such a bridge can be shaped, aligned with an existing ecosystem, filled with collaborative method development, and sustained through training. Beyond the platform, it is MRP that is the transferable contribution, with a claim to cross-disciplinary and cross-project application. We report on two archaeological analyses that have been reconstructed as first proofs of concept, techniques for early empirical feedback on the conditions for adoption of an MRP-related tool, and a research roadmap anchored in this beginning.
- Open Access
- 13 July 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Spatial Algorithms and Systems
- Vol. 12 (4) , 1-2
Introduction to the Special Issue on SIGSPATIAL 2024
- 13 July 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Spatial Algorithms and Systems
- Vol. 12 (4) , 1-2
Open AccessIntroduction to the Special Issue on SIGSPATIAL 2024
- Open Access
- 13 May 2026
- Published by Springer Nature in Production Engineering
- Vol. 20 (2) , 68
Concept-based force-signature analysis of tool-parameter effects in fine blanking
Fine blanking part quality is highly sensitive to tool parameterization, yet cause–effect relationships are obscured by substantial process noise. This limits the utility of both simulation and scalar process metrics for tool optimization and mechanistic understanding. At the same time, high-resolution force-signature data is information-rich and can be used as evidence for learning physically plausible relationships, provided that interpretability is retained. This paper presents a data-to-knowledge pipeline combining finite element method simulation, neural network prediction, and concept extraction to derive interpretable patterns from force time-series and formulate mechanistic hypotheses linking tool parameters to die-roll formation. Controlled tool-parameter variation generates approximately 27,000 strokes with die-roll metrology. Neural networks predict die-roll height and tool parameters with mean absolute percentage errors of 0.6% and 2.6%, respectively. Concept extraction (ECLAD-ts) and segment-aware attribution localize predictive information to peak-load and post-peak regimes. For die-clearance variation, piecewise-linear analysis within concept windows identifies two mechanism-sensitive slope descriptors (Cliff’s $$|\delta | \ge 0.83$$ ) with ordered progressions mirroring the graded die-roll response. The rising-flank slope captures clearance-dependent shearing resistance, while the post-peak decay slope reflects thermo-mechanical softening, reducing flow stress and accelerating the post-peak force decay. Smaller clearance yields higher resistance, faster decay, and reduced die-roll height, consistent with forming theory. The pipeline enables phase-local assessment of tool-parameter effects under process noise and supports testable mechanistic hypotheses for simulation refinement.- 13 May 2026
- Published by Springer Nature in Production Engineering
- Vol. 20 (2) , 68
Open AccessConcept-based force-signature analysis of tool-parameter effects in fine blanking
Fine blanking part quality is highly sensitive to tool parameterization, yet cause–effect relationships are obscured by substantial process noise. This limits the utility of both simulation and scalar process metrics for tool optimization and mechanistic understanding. At the same time, high-resolution force-signature data is information-rich and can be used as evidence for learning physically plausible relationships, provided that interpretability is retained. This paper presents a data-to-knowledge pipeline combining finite element method simulation, neural network prediction, and concept extraction to derive interpretable patterns from force time-series and formulate mechanistic hypotheses linking tool parameters to die-roll formation. Controlled tool-parameter variation generates approximately 27,000 strokes with die-roll metrology. Neural networks predict die-roll height and tool parameters with mean absolute percentage errors of 0.6% and 2.6%, respectively. Concept extraction (ECLAD-ts) and segment-aware attribution localize predictive information to peak-load and post-peak regimes. For die-clearance variation, piecewise-linear analysis within concept windows identifies two mechanism-sensitive slope descriptors (Cliff’s $$|\delta | \ge 0.83$$ ) with ordered progressions mirroring the graded die-roll response. The rising-flank slope captures clearance-dependent shearing resistance, while the post-peak decay slope reflects thermo-mechanical softening, reducing flow stress and accelerating the post-peak force decay. Smaller clearance yields higher resistance, faster decay, and reduced die-roll height, consistent with forming theory. The pipeline enables phase-local assessment of tool-parameter effects under process noise and supports testable mechanistic hypotheses for simulation refinement. - Open Access
- 25 April 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Spatial Algorithms and Systems
- Vol. 12 (2) , 1-2
Introduction to the Special Issue on SIGSPATIAL 2023
- 25 April 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Spatial Algorithms and Systems
- Vol. 12 (2) , 1-2
Open AccessIntroduction to the Special Issue on SIGSPATIAL 2023
- Open Access
- 8 March 2026
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 14 (5) , 507
Representation Learning for Maritime Vessel Behaviour: A Three-Stage Pipeline for Robust Trajectory Embeddings
The growing complexity of maritime navigation creates safety challenges that drive the shift toward autonomous systems. Maritime vessel behaviour modelling is critical for safe and efficient autonomous operations. Representation learning offers a systematic approach to learn feature embeddings encoding vessel behaviour for improved situational awareness and decision-making. We introduce a three-stage representation learning pipeline evaluating six architectures on real-world AIS trajectories. Grouped Masked Autoencoder (GMAE)-Risk Extrapolation (REx) combines group-wise masked autoencoding at the semantic feature level with risk extrapolation regularisation, forcing encoders to learn cross-group dependencies between temporal, kinematic, spatial, and interaction features. DAE and EAE provide robust and uncertainty-aware baselines. Evaluation uses a dual-pipeline framework on two years of Kiel Fjord AIS data (176,787 trajectories, 527,225 segments). Pipeline 1 applies three-stage representation learning using vessel-type classification as encoder selection probe. GMAE-REx achieves 86.03% validation accuracy, outperforming DAE (85.63%), EAE (85.56%), and baselines Transformer (84.93%), TCN (76.27%), LiST (85.12%). Pipeline 2 applies unsupervised clustering to discover intrinsic behavioural structure. Learnt representations consistently outperform expert features on DBCV, conductance, and modularity metrics, organising trajectories by operational context rather than vessel type. This behaviour-oriented organisation enables cross-vessel knowledge transfer for autonomous navigation, VTS monitoring, and safety analysis.
- 8 March 2026
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 14 (5) , 507
Open AccessRepresentation Learning for Maritime Vessel Behaviour: A Three-Stage Pipeline for Robust Trajectory Embeddings
The growing complexity of maritime navigation creates safety challenges that drive the shift toward autonomous systems. Maritime vessel behaviour modelling is critical for safe and efficient autonomous operations. Representation learning offers a systematic approach to learn feature embeddings encoding vessel behaviour for improved situational awareness and decision-making. We introduce a three-stage representation learning pipeline evaluating six architectures on real-world AIS trajectories. Grouped Masked Autoencoder (GMAE)-Risk Extrapolation (REx) combines group-wise masked autoencoding at the semantic feature level with risk extrapolation regularisation, forcing encoders to learn cross-group dependencies between temporal, kinematic, spatial, and interaction features. DAE and EAE provide robust and uncertainty-aware baselines. Evaluation uses a dual-pipeline framework on two years of Kiel Fjord AIS data (176,787 trajectories, 527,225 segments). Pipeline 1 applies three-stage representation learning using vessel-type classification as encoder selection probe. GMAE-REx achieves 86.03% validation accuracy, outperforming DAE (85.63%), EAE (85.56%), and baselines Transformer (84.93%), TCN (76.27%), LiST (85.12%). Pipeline 2 applies unsupervised clustering to discover intrinsic behavioural structure. Learnt representations consistently outperform expert features on DBCV, conductance, and modularity metrics, organising trajectories by operational context rather than vessel type. This behaviour-oriented organisation enables cross-vessel knowledge transfer for autonomous navigation, VTS monitoring, and safety analysis.
- Open Access
- 17 December 2025
- Published by Wiley in Archaeometry
- Vol. 68 (4) , 607-619
Provenance Analysis Based on Cluster In‐Betweenness and Support Vector Machines: Identifying Migrant Candidates Using Multi‐Isotope Fingerprints
Provenance reconstruction using strontium and lead stable isotopes can produce complex multidimensional fingerprints, challenging traditional methods. Identifying nonlocals, who migrated between sites, is a major task. Migrants are identifiable by divergent multi‐isotope fingerprints due to isotopic mixing between origin and destination sites. Detecting early migrants, however, is possible in exceptional cases only. This study used a Support Vector Machine approach to detect migrants as “in‐between instances” between sites. Tested on synthetic data, the model effectively identified IBIs under varying conditions. Applied to archaeological multi‐isotope data, it revealed nonlocal individuals, also early migrants, including the approximated years since migration.
- 17 December 2025
- Published by Wiley in Archaeometry
- Vol. 68 (4) , 607-619
Open AccessProvenance Analysis Based on Cluster In‐Betweenness and Support Vector Machines: Identifying Migrant Candidates Using Multi‐Isotope Fingerprints
Provenance reconstruction using strontium and lead stable isotopes can produce complex multidimensional fingerprints, challenging traditional methods. Identifying nonlocals, who migrated between sites, is a major task. Migrants are identifiable by divergent multi‐isotope fingerprints due to isotopic mixing between origin and destination sites. Detecting early migrants, however, is possible in exceptional cases only. This study used a Support Vector Machine approach to detect migrants as “in‐between instances” between sites. Tested on synthetic data, the model effectively identified IBIs under varying conditions. Applied to archaeological multi‐isotope data, it revealed nonlocal individuals, also early migrants, including the approximated years since migration.
- Open Access
- 3 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 25-32
Anomalies In The Depths: Evaluating Unsupervised Anomaly Detection on Real-World Unlabelled Data. A Case Study on Autonomous Ocean Gliders
- 3 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 25-32
Open AccessAnomalies In The Depths: Evaluating Unsupervised Anomaly Detection on Real-World Unlabelled Data. A Case Study on Autonomous Ocean Gliders