Comparison of real-time network traffic estimator models in gain scheduler middleware by unmanned ground vehicle network-based controller

We propose a gain scheduler middleware (GSM) module, for use in IP network-based control applications. The network traffic estimator (NTE) module of GSM must provide good estimation of the real-time network delay in order for the overall IP network-based control to provide satisfactory results. In this paper, we demonstrate the advantages of the NTE module vs. non-delay estimation control. Five delay estimation models are introduced and compared by using three typical network delay datasets. A network-based control of an unmanned ground vehicle (UGV) path-tracking problem is used to illustrate the performance of the NTE model. Simulation results show that the generalized exponential model outperforms the other models on the three criteria, including the UGV tracking error, travelling time and the computation time of the estimation model.