A nonlinear estimation and control algorithm based on ant colony optimization

Abstract

A new heuristic controller, called Continuous Ant Colony Controller, is proposed for nonlinear stochastic systems. The new controller formulates the states estimation and model predictive control problems as a single stochastic dynamic optimization problem and utilizes a colony of virtual ants to find and track the best state estimation and the best control signal. For this purpose an augmented state space is defined. An integrated cost function is also defined to evaluate the ants within the state space. This function minimizes simultaneously the state estimation error, tracking error, control effort and control smoothness. Ants search the augmented state space dynamically in a similar scheme to the optimization algorithm, known as Continuous Ant Colony System. The performance of the new controller is evaluated for three nonlinear problems. The first problem is a nonlinear cart and spring system, the second problem is a nonlinear Continuous Stirred Tank Reactor, and the third problem is a nonlinear two dimensional engagement between a pursuer and a target. The results verify the successful performance of the proposed algorithm from both estimation and control points of view.