Effective Combination of Search Policy Based on Probability and Entropy for Heterogeneous Mobile Sensors

In this paper, we consider a problem of searching several static targets with a team of mobile sensors consisting of two types of sensors which have different sensor ranges and detection probabilities. Roughly speaking, searching policies are classified into two approaches: one is based on the probability of the presence of targets, and the other is based on the entropy evaluating the uncertainty of their presence. The search performance of the team depends on which policy each sensor adopts. First, we show that a change in the number of targets affects the superiority or inferiority of the search policy based on probability, and the search policy based on entropy. Then, we clarify which combination of search policies is effective in a search by a team of heterogeneous sensors.