Funded Grants

Emergent cooperative search in natural and engineered systems

Trillions of T cells are flowing through your arteries and crawling through your tissues as they search for pathogens. Without a blueprint of your body or centralized instructions, they protect you from flu, nascent tumors and their own uncontrolled proliferation.

Uncountable numbers of ants are crawling across forest canopies, desert sands and maybe your kitchen counter as they search for food or prey. Each species has evolved its own decentralized strategy that tailors a small repertoire of sensing, navigation and communication behaviors to forage effectively in its environment. With these collective search strategies, ants have dominated ecosystems across the globe for millions of years.

A lone billion dollar robot moves a few meters across the Martian surface as it searches for signs of life. It stops and waits for its next command to arrive from Earth. It is easier to send a robot to Mars than to make it cooperate with others or navigate autonomously.

Spectacularly successful decentralized collective behaviors have evolved in ant colonies and immune systems, but no cooperative robotic system has functioned in the real world. Despite successes in assembly lines and warehouses, robots cannot yet cooperate with each other or interact with complex environments.

How does cooperative behavior emerge in complex systems? What behaviors in individuals result in cooperative collectives that effectively work toward a common goal without central control? How could such behaviors have evolved in biology and how can they be designed into technology?

With a team of students and collaborators, I seek answers to these questions through empirical studies, computer simulations and, increasingly, mathematical theory. Our research examines emergent cooperative search through:

  • Field studies of ants searching for food in a variety of resource distributions
  • Lab studies of T cells searching in tissues and lymph nodes
  • Engineering cooperative robotic swarms that navigate and search autonomously
  • Agent based models of search processes in all three systems
  • Mathematical analyses of how search efficiency depends on sensory capabilities, environmental heterogeneity, communication and other behaviors.

We study these three systems to understand how cooperative search emerges at different scales, in different environments, and with different levels of experimental control and analytical precision.

Our simulations test how sensing, navigation and communication behaviors affect collective search success as individuals interact with each other and with complex environments. We propose to develop a mathematical framework that predicts performance of cooperative search strategies based on their ability to extract information from the environment and communicate information between individuals.

Our work is applicable in vaccine design, immunotherapy and swarm robotics. However, we do not envision a one-way street between theory and applications. We will apply theory as we develop it to reveal gaps in our understanding and errors in our assumptions.

Through empirical studies, engineering, simulation and mathematical analysis, we will deepen understanding of how cooperative search emerges in natural systems and how it can be designed into human-built systems.