SinCity: A Pedagogical Testbed For Checking Multi-Agent Learning Techniques

In this paper we present SinCity, a pedagogical testbed to compare multi-agent learning strategies. SinCity has been developed in NetLogo and it can be considered as an extension of the simple predator-prey pursuit problem. In our case, we model a police/thief pursuit in an urban grid environment where several elements (cars, traffic lights, etc.) may interact with the agents in the simulation. SinCity allows to model, in a graphical friendly environment, different strategies for both, the Police and the thief, also implying coordination and communication among the agent set. SinCity is oriented mainly as a pedagogical testbed for learning techniques. The main contributions of the paper are a graphical simulating environment in NetLogo (distributed as open source), a pedagogical testbed for simulating multiagent learning strategies, and the results concerning the simulations performed.