Spatial statistics and distributed estimation by robotic sensor networks

Networks of environmental sensors are playing an increasingly important role in scientific studies of the ocean, rivers, and the atmosphere. Robotic sensors can improve the efficiency of data collection, adapt to changes in the environment, and provide a robust response to individual failures. Their operation must be driven by statistically-aware algorithms that make the most of the network capabilities for data collection and fusion. At the same time, such algorithms need to be distributed and scalable to make robotic networks capable of operating in an autonomous and robust fashion. The combination of these two objectives, complex statistical modeling and distributed coordination, presents grand technical challenges: traditional statistical modeling and inference assume full availability of all measurements and central computation. While the availability of data at a central location is certainly a desirable property, the paradigm for distributed motion coordination builds on partial, fragmented information. This work surveys recent progress at bridging the gap between sophisticated statistical modeling and distributed motion coordination.