Small and Adrift with Self-Control: Using the Environment to Improve Autonomy

We present information theoretic search strategies for single and multirobot teams to localize the source of a chemical spill in turbulent flows. In this work, robots rely on sporadic and intermittent sensor readings to synthesize information maximizing exploration strategies. Using the spatial distribution of the sensor readings, robots construct a belief distribution for the source location. Motion strategies are designed to maximize the change in entropy of this belief distribution. In addition, we show how a geophysical description of the environmental dynamics can improve existing motion control strategies. This is especially true when process and vehicle dynamics are intricately coupled with the environmental dynamics. We conclude with a summary of current efforts in robotic tracking of coherent structures in geophysical flows. Since coherent structures enables the prediction and estimation of the environmental dynamics, we discuss how this geophysical perspective can result in improved control strategies for autonomous systems. M. Ani Hsieh, Hadi Hajieghrary, and Dhanushka Kularatne Scalable Autonomous Systems Lab, Drexel University, Philadelphia, PA 19104, USA e-mail: {mhsieh1,hadi.hajieghrary,dnk32}@drexel.edu Christoffer R. Heckman Autonomous Robotics & Perception Group at the University of Colorado, Boulder, CO 80309 USA e-mail: [email protected] Eric Forgoston Department of Mathematical Sciences, Montclair State University, Montclair, New Jersey 07043, USA e-mail: [email protected] Ira B. Schwartz Nonlinear Systems Dynamics Section, Plasma Physics Division, Code 6792, U.S. Naval Research Laboratory, Washington, DC 20375, USA e-mail: [email protected] Philip A. Yecko Albert Nerken School of Engineering, The Cooper Union, New York, NY 10003, USA e-mail: