Seminar Carlos Gershenson: Self-organizing urban transportation systems

Self-organizing urban transportation systems

Carlos Gershenson
Universidad Nacional Autónoma de México & Vrije Universiteit Brussel

Tuesday, September 22nd, 2009.

Urban transportation is a complex phenomenon. Since many agents are interacting in parallel, it is difficult to predict the future state of a transportation system. Because of this, optimization techniques tend to give obsolete solutions, as the problem changes before it can be optimized. An alternative lies in seeking adaptive solutions. This adaptation can be achieved with self-organization. In a self-organizing transportation system, the elements of the system follow local rules to achieve a global solution. Like this, when the problem changes the system can adapt by itself to the new configuration.
In this talk, I will review recent, current, and future work on self-organizing transportation systems. Self-organizing traffic lights have proven to improve traffic flow considerably over traditional methods. In public transportation systems, simple rules are being explored to prevent the "equal headway instability" phenomenon. The methods we have used can be also applied to other urban transportation systems and their generality will be discussed.