Evolving Dynamic Fault Mitigation Strategies in a Robot Swarm for Collective Transport

Individual robots can fail - swarm as a whole should keep working.

Both faulty and non-faulty robots can take actions to mitigate effects of fault.

Neural network that takes in various sensor values from robot and choose an action optimal for the performance of the swarm as a whole. Therefore there’s no explicit fault detection step, just automatic performing of various mitigation steps to optimise performance.

Created 5/2/2025
Tended
  • 5/2/2025