Motivated by the lateral line system of fish, arrays of flow sensors have been proposed as a new sensing modality
for underwater robots. Existing studies on such artificial lateral lines (ALLs) have been mostly focused on the
localization of a fixed underwater vibrating sphere (dipole source). In this paper we examine the problem of
tracking a moving dipole source using an ALL system. A nonlinear estimation problem is formulated based on an
analytical model for the moving dipole-generated flow field, which is subsequently solved with the Gauss-Newton
method. The effectiveness of the proposed approach is illustrated with simulation results.
The lateral line system, consisting of arrays of neuromasts functioning as flow sensors, is an important sensory
organ for fish that enables them to detect predators, locate preys, perform rheotaxis, and coordinate schooling.
Creating artificial lateral line systems is of significant interest since it will provide a new sensing mechanism for
control and coordination of underwater robots and vehicles. In this paper we propose recursive algorithms for
localizing a vibrating sphere, also known as a dipole source, based on measurements from an array of flow sensors.
A dipole source is frequently used in the study of biological lateral lines, as a surrogate for underwater motion
sources such as a flapping fish fin. We first formulate a nonlinear estimation problem based on an analytical
model for the dipole-generated flow field. Two algorithms are presented to estimate both the source location and
the vibration amplitude, one based on the least squares method and the other based on the Newton-Raphson
method. Simulation results show that both methods deliver comparable performance in source localization. A
prototype of artificial lateral line system comprising four ionic polymer-metal composite (IPMC) sensors is built,
and experimental results are further presented to demonstrate the effectiveness of IPMC lateral line systems and
the proposed estimation algorithms.