Applying swarm algorithms in manufacturing scenarios
Our aim is to design and develop swarm inspired algorithms for various applications such as scanning and monitoring artefacts or carrying loads.
Background
Nature provides solutions to most engineering problems in amazing ways in the form of how organisms operate in their constantly changing environment. These solutions have been tested and adapted over a number of years and hence are robust and very effective in solving problems.
Cooperative transport by a swarm of Quadcopters offers more flexibility and performance when carrying loads that are complex in structural profile and mass. Ensuring that team members of the swarm are optimally placed on these loads as well as able to resist disturbances from the environment during transport are current research challenges.
In our work
- We make use of a decentralised behaviour based subsumption architecture to enable a swarm of Quadcopters to explore an unfamiliar area, find a load and transport it to a target location cooperatively. In the architecture, different behaviours are used.
- We also apply multi-agent reinforcement learning algorithms to this end.
- We apply our developed frameworks on various user-inspired use cases.