Multi-robot system performance can be evaluated and benchmarked in generalized search and task allocation problems (STAP) [1]. In STAP robots search for tasks and recruit other robots to help solve and complete the discovered tasks. Multi-agent coordination algorithms, like swarm intelligence and game theory, are of special interest when assessing multi-robot system performance in STAP. Also, system dimensions like the ability of inter-robot communication, robot computing power and unit design cost can be simulated and tested using this benchmark problem. It would be interesting to research if there exist any swarm behaviour that could approximate the optimal game theoretic coordination algorithm in terms of system performance in STAP? A strong candidate would be to analyse quorum sensing mechanisms in this context, e.g. estimating swarm density using collision frequency or some signal aggregation. Consequently, this thesis would evaluate system performance when comparing quorum sensing mechanisms to the optimal auction algorithm in simulated multi-agent STAP. The simulated results could be validated in the ITS robot lab using up to 20 TurtleBots.
Note: This thesis is available for completion as either a short or a long thesis.
[1] M. Minos-Stensrud, H. Jonas Fossum Moen and J. Dyre Bjerknes, "Information sharing in multi-agent search and task allocation problems," 2021 IEEE Symposium Series on Computational Intelligence (SSCI), 2021, pp. 1-7, doi: 10.1109/SSCI50451.2021.9660121.