Publication

Abstract : We examine the problem of tracking dynamic boundaries occurring in natural phenomena using a network of range sensors. Two main challenges of the boundary tracki ng problem are accurate boundary estimation from noisy observations and continuous tracking of the boundary. We propose Dynamic Boundary Tacking (DBTR), an algorithm that combines the spatial estimation and temporal estimation techniques to effectively track a dynamic boundary. The regression-based spatial estimation technique determines discrete points o n the boundary and estimates a confidence band around the entire boundary. In addition, a Kalman Filter-based temporal estimation technique tracks changes in the boundary and aperiodic ally updates the spatial estimation to meet accuracy requiremen ts. DBTR, provides a low communication overhead solution to track boundaries without requiring prior knowledge about the dynamics. Experimental results demonstrate the effective ness of our algorithm; estimated confidence bands indicate aloss of coverage of less than 2 − 5% for a variety of boundaries with different spatial characteristics