PLFDoc: Precision Livestock Farming
Disciplines
Computer Sciences (50%); Animal Breeding, Animal Production (50%)
Keywords
- Livestock,
- Computer Vision,
- Sensors,
- Animal Health,
- Animal Welfare,
- Artificial Intelligence
The doctoral program PLFDoc - Precision Livestock Farming (PLFDoc) is a collaborative project involving the TU Wien, the University of Applied Sciences Upper Austria (Campus Hagenberg), and the University of Veterinary Medicine Vienna. This interdisciplinary training program offers five PhD candidates the opportunity to work as a team on contributing to a more sustainable agriculture with improved livestock farming. The research focus consists of applied basic research, particularly in the application of methods of Explainable Artificial Intelligence (XAI) as well as image and video analysis (Computer Vision, CV), to be used for monitoring livestock. Recent advancements in sensor technologies, computer vision, and data science provide the potential for continuous, real-time monitoring of the health status and well- being of animals, both at the herd and individual level. The application of these tools thus holds the potential to detect behavioral and/or physiological deviations in animals early and to initiate appropriate countermeasures to maintain or enhance health and well-being. This is not only relevant for farmers and veterinarians but also of societal significance. PLFDoc aims to develop new CV- and XAI-based tools, including prediction of parturition, animal identification, and estimation of individual water and feed intake in pigs and cattle. The development of these tools requires the combination of expertise from fields such as veterinary medicine, agricultural sciences, behavioral biology, data science, and engineering. The collaborative research by experts from these disciplines is therefore of utmost importance and forms the research framework for the doctoral program. As part of a team, the PhD candidates pursue common goals and learn from each other. The faculty members of the doctoral program possess the respective expertise needed for optimal personal support of the candidates and for conducting the described research activities. Due to the interdisciplinary approach, the doctoral candidates receive training in a variety of areas. The theoretical, scientific, experimental, networking, and communication skills acquired in PLFDoc are essential foundations for a successful future career in industry or academia.
-
consortium member (01.03.2026 -)
-
consortium member (01.07.2026 -)
-
consortium member (01.10.2023 -)
-
consortium member (01.10.2023 -)
-
consortium member (01.10.2023 -)
- Veterinärmedizinische Universität Wien
Research Output
- 3 Citations
- 14 Publications
- 8 Datasets & models
- 1 Disseminations
- 1 Scientific Awards
- 2 Fundings
-
2026
Title Using posture classification for detecting calving related behaviour in dairy cows Type Conference Proceeding Abstract Author Gosch Conference 59th Annual Conference Physiology and Pathology of Reproduction and 51st Joint Conference on Veterinary and Human Reproductive Medicine, Munich, Germany. Link Publication -
2026
Title Vision-based detection of pain and nest-building behaviors in sows within commercial farrowing pens Type Conference Proceeding Abstract Author Helf Conference Artificial Intelligence 4 Animal Science 2026, Ghent, Belgium -
2026
Title Visions-based detection of pain and nest-building behaviors in sows within commercial farrowing pens Type Conference Proceeding Abstract Author Helf Conference Austrian Symposium on AI, Robotics, and Vision Pages 267-272 Link Publication -
2025
Title How toMeasure Explainability andInterpretability ofMachine Learning Results; In: Genetic Programming Theory and Practice XXI DOI 10.1007/978-981-96-0077-9_18 Type Book Chapter Publisher Springer Nature Singapore -
2026
Title Skeleton integrity: A method for the efficient fine-tuning of pose estimation models for pigs DOI 10.1016/j.biosystemseng.2025.104380 Type Journal Article Author Brunner D Journal Biosystems Engineering -
2026
Title Improved and interpretable accelerometer-based farrowing prediction DOI 10.1016/j.biosystemseng.2025.104381 Type Journal Article Author Mayrhuber E Journal Biosystems Engineering -
2025
Title Towards Automated and Interpretable Decision Support Systems for Precision Livestock Farming Using Evolutionary Computing DOI 10.1145/3712255.3734314 Type Conference Proceeding Abstract Author Mayrhuber E Pages 2600-2603 Link Publication -
2025
Title 14. Leveraging Vision Mamba for cow pose estimation DOI 10.1016/j.anscip.2025.08.169 Type Journal Article Author Daadouch S Journal Animal - Science proceedings Pages 515-517 -
2025
Title 89. Evaluation of YOLO-Pose for cattle pose estimation from videos DOI 10.1016/j.anscip.2025.08.244 Type Journal Article Author Zhao K Journal Animal - Science proceedings Pages 658-660 -
2025
Title 136. Pose classification for detection of calving related behavior in dairy cows DOI 10.1016/j.anscip.2025.08.291 Type Journal Article Author Gosch M Journal Animal - Science proceedings Pages 745-747 -
2025
Title Explainable Artificial Intelligence and Computer Vision as Powerful Tools for Precision Livestock Farming. Type Conference Proceeding Abstract Author Brunner Conference 18. Forschungsforum der österreichischen Fachhochschulen Pages 552-553 Link Publication -
2025
Title Automated Body Measurement of Sows in Feeding Stations Using Multiple Cameras Type Conference Proceeding Abstract Author Helf Conference 1st EAAP Conference on Artificial Intelligence 4 Animal Science, 2025, Zurich, Switzerland Pages 63 Link Publication -
2024
Title Exploring the Potential of Symbolic Regression in Precision Livestock Farming. Type Conference Proceeding Abstract Author Mayrhuber Conference 11th European Conference on Precision Livestock Farming, Bologna, Italy 9 - 12 September 2024 Pages 1324-1331 -
2024
Title Skeleton-based computer vision algorithms for detecting behaviour of dairy cows prior to calving Type Conference Proceeding Abstract Author Gosch Conference 11th European Conference on Precision Livestock Farming. Bologna, Italy. Pages 274-283 Link Publication
-
0
Link
Title Pre-farrowing activity dataset Type Database/Collection of data Public Access Link Link -
0
Title YOLO Pose models trained for cattle pose estimation DOI 10.1016/j.anscip.2025.08.244 Type Computer model/algorithm Public Access -
0
Link
Title Pre-farrowing behavior dataset DOI 10.34749/3061-1466.2026.36 Type Database/Collection of data Public Access Link Link -
0
Link
Title Cow calving postures dataset DOI 10.1016/j.anscip.2025.08.291 Type Database/Collection of data Public Access Link Link -
2026
Link
Title Pig pose estimation dataset DOI 10.17632/sn49zt6jpw.1 Type Database/Collection of data Public Access Link Link -
2025
Link
Title Vision Mamba-based Keypoint R-CNN model for cow pose estimation DOI 10.1016/j.anscip.2025.08.169 Type Computer model/algorithm Public Access Link Link -
2025
Link
Title MLP classifyer for classifying calving-related body-, head- and tail postures of cows approaching parturition DOI 10.1016/j.anscip.2025.08.291 Type Computer model/algorithm Public Access Link Link -
2025
Link
Title K-nearest neighbors (KNN) classifyer for classifying calving-related body-, head- and tail postures of cows approaching parturition DOI 10.1016/j.anscip.2025.08.291 Type Computer model/algorithm Public Access Link Link
-
2025
Title Presentation to the industry partner - memeber of the industry advisory board of PLFDoc Type A talk or presentation
-
2025
Title Frontiers in Precision Livestock Farming Award Type Poster/abstract prize Level of Recognition Continental/International
-
2026
Title Computer vision assisted phenotyping, early detection and genetic analysis of abnormal behavior in weaned piglets Type Research grant (including intramural programme) Start of Funding 2026 Funder BMLUK - Federal Ministry of Agriculture and Forestry, Climate and Environmental Protection, Regions and Water Management -
2025
Title VetIdeas challenge - 3rd place for PhD student of PLFDoc - Mathias Gosch Type Travel/small personal Start of Funding 2025 Funder University of Veterinary Medicine Vienna