Cooperation Strategies in a Time-Stepped Simulation of Foraging Robots

Large robotic swarms may be used to carry out tasks such as space exploration, mining, search & rescue operations and more. To enable their use in these fields, the individual robots within a swarm will need to be autonomic, capable of making their own decisions and adjusting their behaviour without relying on regular human intervention. This paper demonstrates the potential for autonomic self-adaptation within a swarm of foraging robots by investigating the performance of different cooperation strategies in different scenarios. The results show that the performances of the strategies are affected by operational conditions that can change over the course of a mission, and that the autonomic capability to self-adapt would prove beneficial. Additionally, the time-stepped simulation used here is compared to the performance of a previous approach using real-time simulation, with a view to identifying which approach is more suitable for embedding within a robot as a means of aiding that autonomic process through simulating potential options. The time-stepped simulation is found to be faster and more efficient, and therefore more suited to embedding.