Scalar field estimation with mobile sensor networks

In this article, we consider the problem of estimating a scalar field using network mobile sensors which can measure value at their instantaneous location. The to be estimated is assumed represented by positive definite radial basis kernels and use techniques from adaptive control Lyapunov analysis prove stability proposed estimation algorithm. convergence parameter values true guaranteed planning motion satisfy persistence-like conditions. Two kinds algorithms are proposed: (1) where each sensor estimates entire vector, (2) only part vector. Simulations used for validating algorithms. We find that algorithm in case gives more accurate but computationally expensive, whereas performs much faster though less accurate. A modification second seen perform fast while providing good accuracy.