Cooperative multiagent systems for the optimization of urban traffic

The risks and benefits of trusting others in a cooperative context is discussed and the notion of social rationality is used to establish these ideas in the realm of autonomous rational agents (utility maximizers). A traffic simulation is introduced to test the ideas presented in this work. The simulation consists of traffic lights controlled by cooperative autonomous agents, each of whom is given mechanisms to implement social rationality. Results using cooperative traffic agents are compared to results of control simulations where non-cooperative agents were deployed. Results predictably show a loss of local efficiency and a gain of global efficiency with the cooperative agents in comparison to their non-cooperative counterparts.