Nigel Groome Studentship: NatureNet: AI systems for conservation and the management of human-wildlife conflict

Human-wildlife conflict is widespread, yet current monitoring systems aiming at their reduction remain costly, vulnerable, and difficult to scale. This project will focus on computational and engineering innovation by developing wildlife tracking technologies, including integrating advanced AI-based analytics to create a novel
prototype for conflict mitigation.

The work will involve developing a data processing pipeline capable of handling complex behavioural datasets from biologgers and autonomous monitoring units, enabling accurate movement and interaction analysis without reliance on GPS-based technology. In the final phase, the prototype will be tested in real-world conditions such as involving predators and/or semi-domesticated livestock, in collaboration with local and Indigenous stakeholders (Europe, the Global South).

This project develops novel computational methods to meet conservation needs, delivering a transformative solution for reducing human-wildlife conflict. 

This project would suit individuals with an interest in technology development and computational methods to be applied towards wildlife monitoring and conservation.

The student will join two complementary research environments: The C-Wild Warrington lab (Ecology and Conservation) and the Machine Learning and Robotics research group (Artificial Intelligence, Data Analysis and Systems). This will provide interdisciplinary training, enabling the student to integrate ecological knowledge with advanced computational methods for developing and testing innovative wildlife monitoring solutions.

The student should have strong computational skills, including programming (Python or similar), data analysis, and familiarity with machine learning for time-series and sensor data. Basic knowledge of designing and building electronic units (e.g., biologgers, IoT systems), experience managing large datasets, and prior exposure to working in wild or remote field settings are essential.