Jay A Farrell | Search PubMed

PubMed2006

Chemical plume source localization.

This paper addresses the problem of estimating a likelihood map for the location of the source of a chemical plume using an autonomous vehicle as a sensor probe in a fluid flow. The fluid flow is assumed to have a high Reynolds number. Therefore, the dispersion of the chemical is dominated by turbulence, resulting in an intermittent chemical signal. The vehicle is capable of detecting above-threshold chemical concentration and sensing the fluid flow velocity at the vehicle location. This paper reviews instances of biological plume tracing and reviews previous strategies for a vehicle-based plume tracing. The main contribution is a new source-likelihood mapping approach based on Bayesian inference methods. Using this Bayesian methodology, the source-likelihood map is propagated through time and updated in response to both detection and nondetection events. Examples are included that use data from in-water testing to compare the mapping approach derived herein with the map derived using a previously existing technique.

Air Pollution↗

PubMed2005

Composite adaptive control with locally weighted statistical learning.

This paper introduces a probably stable learning adaptive control framework with statistical learning. The proposed algorithm employs nonlinear function approximation with automatic growth of the learning network according to the nonlinearities and the working domain of the control system. The unknown function in the dynamical system is approximated by piecewise linear models using a nonparametric regression technique. Local models are allocated as necessary and their parameters are optimized on-line. Inspired by composite adaptive control methods, the proposed learning adaptive control algorithm uses both the tracking error and the estimation error to update the parameters. We first discuss statistical learning of nonlinear functions, and motivate our choice of the locally weighted learning framework. Second, we begin with a class of first order SISO systems for theoretical development of our learning adaptive control framework, and present a stability proof including a parameter projection method that is needed to avoid potential singularities during adaptation. Then, we generalize our adaptive controller to higher order SISO systems, and discuss further extension to MIMO problems. Finally, we evaluate our theoretical control framework in numerical simulations to illustrate the effectiveness of the proposed learning adaptive controller for rapid convergence and high accuracy of control.

Algorithms↗