Adaptive sampling and energy-efficient navigation in time-varying flows

This chapter presents a strategy to enable a team of mobile robots to adaptively sample and track a dynamic spatiotemporal process. We propose a distributed strategy where robots collect sparse sensor measurements, create a reduced -order model (ROM) of the spatiotemporal process, and use this model to estimate field values for areas without sensor measurements of the dynamic process. The robots then use these estimates of the field, or inferences about the process, to adapt the model and reconfigure their sensing locations. We use this method to obtain an estimate for the underlying fl ow field and use that to plan optimal energy paths for robots to travel between sensing locations. We show that the errors due to the reduced -order modeling scheme are bounded, and we illustrate the application of the proposed solution in simulation and compare it to centralized and global approaches. We then test our approach with physical marine robots sampling a spatially nonuniform time -varying process in a water tank.