Tracking problem of non-particle morphology based on fixed-time ZNN model

Abstract

This article examines a particular tracking challenge where targets and trackers cannot be conceptualized as static points and are continuously rotating. This poses difficulties in determining the shortest distance between them. To rapidly and precisely ascertain this minimal distance, we formulate an optimization problem. Subsequently, a fixed-time zeroing neural network (ZNN) model is devised to address this optimization challenge. Moreover, the fixed-time stability of the proposed network is established and an estimation of a relatively smaller upper bound of convergence time (UBCT) is derived from previous iterations. Furthermore, the sensitivity of parameters to UBCT is also given. Finally, a specific tracking scenario demonstrates the efficacy and superior performance of our approach.