Computer Vision and Geospatial Data Science
Work description
Integration of pasture simulation and animal movement with ML predictions (e.g., meteorology/growth), supporting planning/optimisation modules and the explainability of decisions in the dynamic virtual fence prototype. Tasks: - Contribute to the state of the art and support the writing of technical-scientific documentation (reports and specification notes). - Develop simulation models (e.g., agent-based/spatial) of animal movement and pasture dynamics, calibrated from real data. - Integrate ML forecasts (meteorology, vigour/biomass, mobility) as exogenous inputs to the simulator, including uncertainty propagation. - Design and evaluate planning/optimisation models for dynamic allocation of areas/rotations, coupled with the simulator. - Define validation metrics and protocols (KPIs) and conduct functional tests with pilot data and the prototype. - Explore explainability mechanisms (XAI) to justify simulator/optimizer recommendations to end users.
Academic Qualifications
Master's degree in Computer Engineering, Electrical Engineering and Computer Engineering or related fields;
Minimum profile required
- Average grade in bachelor's and master's degrees above 14.- 1 article accepted for presentation at a conference or publication in a journal.
Preference factors
- Fluency in Portuguese. - Candidate enrolled or attending a doctoral programme. - Solid knowledge of modelling/simulation, optimisation/OR and ML. - Proficiency in Python and good mathematical foundations. - Experience with Pyomo/OR-Tools and agent-based simulation.
Application Period
Since 27 Nov 2025 to 12 Dec 2025
Centre
Enterprise Systems Engineering