Process Pig
Data-driven process analysis for context-sensitive management of functional areas to reduce emissions and promote animal welfare in free-ventilated pig barns with open-air runs.
Project description
The aim is to develop a monitoring tool of behaviour patterns in freely ventilated pig houses with structurally separated functional areas depending on climatic conditions. Through this sensor- and behaviour-oriented monitoring, animal welfare-adapted husbandry conditions and a reduction of ammonia emission areas can be promoted. With the development of a tool for monitoring, an aid for the daily animal control and for the control of the climatic conditions in the stable is to be created. In addition to various sensors, the automatically recorded behaviour of the pigs becomes the central indicator for the evaluation of the animal welfare-oriented, climatic conditions and thus serves to investigate the causes of non-compliance with functional areas. Based on video data and using machine-learning methods, animal behaviour is detected and used to create process models to classify different behaviours in temporal and spatial resolution. Deviations from this behaviour are detected and made available to the user in a clear form as an indication and alarm message. With this information advantage, it is possible to use changes in animal behaviour for the control of different actuators for the stable climate, such as the opening and closing of ventilation flaps, box lids or wind protection nets, in order to make adjustments to the climatic conditions before climatic stress situations arise.