Particle Swarm Optimization for robot target tracking application

Abstract

Tracking is a process of finding continuous path of a selected target. Tracking is important in various civil and defense applications. For tracking applications, various adaptive filters are being used such as Kalman Filter and it's variants. In this paper, a popular meta-heuristic algorithm, namely, Particle Swarm Optimization (PSO) algorithm is applied to track a target which is a robot and it is transmitting noisy data due to installation of cheap quality sensors onboard. These noisy observations are filtered by PSO algorithm in order to estimate true path of the robot. The simulated results are presented by using MATLAB software. From the results, it is observed that the PSO algorithm is very effective algorithm in estimating the target location continuously with almost zero estimation error. Thus, Particle Swarm Optimization algorithm is a suitable algorithm for robot tracking applications and it can also be used for nonlinear data processing in autonomous robot design applications.