On Pseudo-perturbation-based Broadcast Control of Multi-agent Systems
Yuji Ito, Md Abdus Samad Kamal, Takayoshi Yoshimura, Shun‐ichi Azuma · 2018
A novel method of coordinating multi-agent systems, termed pseudo-perturbation-based broadcast control (PBC), is proposed. Recently, broadcast control (BC) has been presented to complete global coordination tasks. BC works with a low communication volume because a supervisor sends an identical signal to all agents indiscriminatingly and there is no agent-to-agent communication. However, in realizing such a low volume of communication, BC suffers from the extra random movements of agents required for stochastic optimization. The random movements increase the control cost and reduce the speed of multi-agent coordination. To overcome these problems, PBC is derived by introducing virtual random movements instead of physical movements. This paper theoretically obtains the following results for PBC. First, the convergence of coordination is twice as quickly as that of BC. Second, the travel distances of agents are shorter than those of BC. Finally, various coordination tasks are asymptotically achieved with probability 1. The effectiveness of PBC is confirmed through a numerical simulation of an assignment task.