Using quantitative image analysis and simulations, we first understand the physics of traffic and for the first time: autonomous cars through an active matter model.

​

We find benefits to overall flux through reduced variance and increased response times of autonomous cars. We then explore a simple hacking scenario with disproportionately large countereffects to flux. Surprisingly, we find analogies with the physics of clogging, ​random pinning, and percolation that have been explored extensively in equilibrium systems. We then identify solutions to this example of collective weaponization.

​

Working on these questions through the lens of the anticipate-and-inoculate

mindset has the potential to identify how other emergent collective phenomena can be weaponized and preemptive disarmed. Ultimately, this approach should
lead to safer roads and a greater good.

S Vivek, D Yanni, PJ Yunker, JL Silverberg Collective behavior and emergent risks in a model of human- and autonomously-driven vehicles arXiv:1708.03791 

​