RMASBench: a Benchmarking System for Multi-Agent Coordination in Urban Search and Rescue (Demonstration)
Fabio Maffioletti, Riccardo Reffato, Alessandro Farinelli, Alexander Kleiner, Sarvapali D. Ramchurn, Bing Shi · 2013
This demonstration paper illustrates RMASBench, a new bench-marking system based on the RoboCup Rescue Agent simulator. The aim of the system is to facilitate benchmarking of coordination approaches in controlled settings for dynamic rescue scenarios. In particular, the key features of the systems are: i) programming in-terfaces to plug-in coordination algorithms without the need for im-plementing and tuning low-level agents ’ behaviors, ii) implemen-tations of state-of-the art coordination approaches: DSA and Max-Sum, iii) a large scale crowd simulator, which exploits GPUs par-allel architecture, to simulate the behaviour of thousands of agents in real time.