The aim is to develop intelligent swarm systems in which aerial, ground-based and maritime drones can collaborate autonomously on tasks such as search and rescue operations, environmental monitoring, precision agriculture and the protection of critical infrastructure. Military applications are also included within the project.
“The project focuses on designing and deploying autonomous drone swarms. It does not only involve aerial systems, but also drones operating on land and in water. Our main contribution will be the airborne component,” says Rikard Tyllström of the Department of Aviation and Aeronautical Sciences at Lund University School of Aviation in Ljungbyhed.
The Lund researchers will contribute to the development, testing and validation of the aerial systems. One of the key challenges is creating a framework in which different types of drones, each with distinct capabilities, can work together effectively.
“Some drones may be designed for long endurance, while others are built for speed or heavy loads. Together, they should be able to solve tasks in ways that no single platform can achieve on its own,” says Rikard Tyllström.
Artificial intelligence and digital twins
At the heart of the initiative is the development of swarm systems capable of self-configuration and continuous adaptation to changing conditions through artificial intelligence and digital twins. A key component will be SwarmOS, a new operating system designed to integrate diverse hardware platforms into a single intelligent system.
The software will be developed as open source, enabling both researchers and other stakeholders to use and further develop the technology.
“We are building an ecosystem that can both be used and expanded upon. For us, this is a very large-scale project,” says Rikard Tyllström.
As part of the initiative, new testing environments will also be established where autonomous drone swarms can be evaluated in realistic operational scenarios. One ambition is to create an environment in which unmanned aerial vehicles can operate alongside manned aircraft within the same airspace.
Shared responsibility between humans and autonomous systems
At the same time, researchers will examine several fundamental questions related to the safe governance and deployment of autonomous swarms. Rogier Woltjer, associate professor and senior lecturer at the Department of Aviation and Aeronautical Sciences, leads two research tracks focusing on risk management and evaluation methodologies.
“There is still relatively little research on how humans should interact with entire swarms of autonomous systems, what risks may emerge, and how new swarm concepts should be evaluated. Research on large swarms has not progressed to the same extent as research on individual drones,” he says.
The research will investigate how control and responsibility can be shared between humans and autonomous systems, how multiple operators can coordinate their efforts, and how swarms can continue functioning even when individual units fail or operating conditions change rapidly.
The work will proceed in stages, beginning with simulations, followed by trials involving real drone swarms, and ultimately testing in operationally relevant environments.
Legal aspects
Researchers from the Department of Technology and Society are also participating in the project. Stefan Larsson, associate professor of technology and social change, and his research group will examine how technical solutions interact with legal, organisational and societal requirements.
“These are socio-technical questions. For example, how organisations collaborating around drone swarms can align their values, norms and working methods. There are also important legal and governance-related issues that need to be addressed,” says Stefan Larsson.
By combining advanced AI, autonomous systems and research on human-technology interaction, the researchers hope to create the flexible and robust swarm systems of the future, with applications ranging from rescue operations to the protection of critical infrastructure.